United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-453/R-83-050b
October 1997
Air
&EPA
Pulp, Paper, and Paperboard
Industry - Background
Information for Promulgated
Air Emission Standards
Final
EIS
Manufacturing Processes at
Kraft, Sulfite, Soda, Semi-
Chemical, Mechanical, and
Secondary and Non-wood
Fiber Mills
-------
EPA-453/R-93-050b
Pulp, Paper, and Paperboard Industry
Background Information for
Promulgated Air Emission Standards
Manufacturing Processes at
Kraft, Sulfite, Soda, Semi-Chemical,
Mechanical, and Secondary and
Non-wood Fiber Mills
Emission Standards Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Ofiice of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
October 1997
-------
This report has been reviewed by the Emission Standards Division of
the Office of Air Quality Planning and Standards, U.S.
Environmental Protection Agency, and approved for publication.
Mention of trade names or commercial products is not intended to
constitute endorsement or recommendation for use. Copies of this
report are available through the Library Services Offices (MD-35) ,
U.S. Environmental Protection Agency, Research Triangle Park, N.C.
27711, (919)541-2777, from National Technical Information Services,
5285 Port Royal Road, Springfield, Virginia 22161, (703)487-4650,
or from the internet (http://www.epa.gov/ttn/).
11
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY (EPA)
Background Information
and Final
Environmental Impact Statement
for Hazardous Air Pollutant Emissions
From the Pulp, Paper, and Paperboard Industry
Prepared by:
Jordn (Date)
Director, ^mission Standards Division
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
1. National emission standards for hazardous air pollutants (NESHAP) are being
promulgated for the pulp and paper industry under authority of Section 112 (d)
of the Clean Air Act as amended in 1990. The promulgated NESHAP requires
controls for hazardous air pollutant emissions from wood pulping and bleaching
processes at pulp mills and integrated mills (i.e., mills that combine on-site
production of both pulp and paper) .
2. Copies of this document have been sent to the following Federal Departments:
Labor, Health and Human Services, Defense, Transportation, Agriculture,
Commerce, Interior, and Energy; the National Science Foundation; the Council
on Environmental Quality; members of the State and Territorial Air Pollution
Program Administrators; the Association of Local Air Pollution Control
Offices; EPA Regional Administrators; and other interested parties.
3. For additional information contact:
Ms. Penny Lassiter
Waste and Chemicals Process Group (MD-13)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Telephone: (919) 541-5396
4. Paper copies of this document may be obtained from:
National Technical Information Service (NTIS)
5285 Port Royal Road
Springfield, VA 22161
Telephone; (703) 487-4650
U.S. EPA Library Services Office (MD-35)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Telephone: (919) 541-2777
5 . Electronic copies of this document may be obtained from the EPA Technology
Transfer Network (TTN) on the internet. The TTN may be accessed at
'http://www.epa.gov/ttn/' .
-------
This page intentionally left blank.
IV
-------
TABLE OF CONTENTS
1.0 SUMMARY 1-1
1.1 INTRODUCTION 1-1
1.2 ORGANIZATION OF THIS DOCUMENT 1-2
1.3 SUMMARY OF PUBLIC COMMENTS 1-7
1.4 SUMMARY OF CHANGES SINCE PROPOSAL 1-8
1.4.1 Definition of Source 1-8
1.4.2 Subcategories 1-29
1.4.3 Control Applicability Determination . . . 1-30
1.4.4 Kraft Standards 1-31
1.4.5 Sulfite Standards 1-35
1.4.6 Soda and Semi-Chemical Mill Standards . . 1-36
1.4.7 Bleaching System Standards 1-36
1.4.8 Compliance Schedule 1-38
1.4.9 Test Methods 1-40
1.4.10 Control Device Downtime 1-41
1.4.11 Equipment Enclosures, Closed-Vent Systems,
and Control Equipment 1-42
1.4.12 Interaction With The Resource Conservation
and Recovery Act (RCRA) 1-44
1.5 SUMMARY OF IMPACT OF FINAL STANDARDS 1-44
1.5.1 Emission Impacts 1-44
1.5.2 Energy Impacts 1-45
1.5.3 Secondary Environmental Impacts 1-45
1.5.4 Cost Impacts 1-46
1.5.5 Economic Impacts 1-46
2.0 INDUSTRY CHARACTERIZATION 2-1
2.1 ADEQUACY OF DATA FOR STANDARDS DEVELOPMENT . . . .2-1
2.2 EMISSION FACTORS 2-2
2.2.1 Data and Approach Used 2-2
2.2.2 Emission Factor Document 2-6
2.3 MODEL PROCESS UNITS 2-12
3.0 SUBCATEGORIZATION 3-1
4.0 BASIS OF STANDARDS 4-1
4.1 STATUTORY INTERPRETATION 4-1
v
-------
TABLE OF CONTENTS
4.1.1 88th percentile vs. 94th percentile
Interpretation 4-1
4.1.2 MACT Floor on a "Per Unit" vs. "Whole Mill"
Basis 4-4
4.1.3 Legal Requirement to Base MACT Floor on
Actual Data 4-7
4.1.4 Legal Requirement to Re-propose 4-8
4.1.5 Control Devices Were Not Installed to
Reduce
HAP 4-10
4.1.6 Authority to Regulate Process Wastewater 4-11
4.2 DEFINITION OF SOURCE 4-12
4.2.1 Plant-wide Definition of Source 4-12
4.2.2 Definition Should Be Limited Within Each
Subcategory 4-18
4.2.3 Woodpiles, Power Systems, and (Methanol)
Recovery 4-19
4.3 MACT 4-20
4.3.1 MACT Floor Level of Control Technologies 4-20
4.3.2 MACT Floor Control Applicability 4-65
4.3.3 Beyond the MACT Floor Level of Control. . 4-80
4.3.4 Downtime and Back-up Control Technologies 4-81
4.4 FORMAT OF THE STANDARDS 4-88
4.4.1 General Comments 4-88
4.4.2 Pulping System 4-101
4.4.3 Bleaching System 4-101
4.4.4 Process Wastewater System 4-105
5.0 PULPING AREA 5-1
5.1 CONTROL OPTIONS 5-1
5.1.1 General 5-1
5.1.2 Existing Combustion Devices 5-7
5.1.3 Design Incinerators 5-12
5.1.4 Enclosures and Gas Collection Systems . . 5-19
5.2 COSTS 5-20
5.3 SECONDARY IMPACTS OF PULPING CONTROLS 5-23
6.0 BLEACHING AREA 6-1
6.1 CONTROL OPTIONS 6-1
vi
-------
TABLE OF CONTENTS
6.1.1 Scrubbers 6-1
6.1.2 Incineration 6-6
6.1.3 Process Modifications 6-7
6.2 MISCELLANEOUS BLEACHING COMMENTS 6-11
7.0 PROCESS WASTEWATER AREA 7-1
7.1 DESIGN STEAM STRIPPER 7-1
7.2 APPLICABILITY CUTOFFS 7-2
7.3 HARDPIPING AND BIOLOGICAL TREATMENT 7-4
7.4 WASTEWATER COLLECTION AND TRANSPORT SYSTEMS .... 7-8
7.5 SECONDARY IMPACTS OF WASTEWATER CONTROLS .... 7-11
7.6 COSTS 7-12
7.7 OTHER 7-16
8.0 MONITORING 8-1
8 . 1 GENERAL 8-1
8.2 CONTINUOUS EMISSIONS MONITORING 8-8
8.3 PARAMETER MONITORING 8-14
8.4 LEAK DETECTION AND INSPECTION 8-24
8.5 BYPASS VENTS 8-27
9.0 TEST METHODS AND PROCEDURES 9-1
9.1 FIVE-MILL SAMPLING PROGRAM 9-1
9.2 REQUIRED TEST METHODS 9-3
10.0 RECORDKEEPING AND REPORTING 10-1
11.0 COST/ECONOMIC IMPACTS 11-1
11.1 COST IMPACTS 11-1
11.2 ECONOMIC IMPACTS 11-3
12.0 BENEFITS 12-1
13.0 EMISSIONS AVERAGING 13-1
14.0 DEFINITIONS 14-1
15.0 INTEGRATED RULE INTERACTION 15-1
15.1 GENERAL 15-1
vii
-------
TABLE OF CONTENTS
Page
15.2 EFFLUENT GUIDELINES 15-5
15.3 COMBUSTION MACT 15-8
16.0 INTERACTION WITH OTHER RULES 16-1
16.1 NEW SOURCE REVIEW/PREVENTION OF SIGNIFICANT
DETERIORATION 16-1
16.2 RESOURCE CONSERVATION AND RECOVERY ACT/BOILERS AND
INDUSTRIAL FURNACES 16-12
16.3 SECTION 112 RULES (112(g), 112 (j), 112(r)) . . . 16-16
16.4 NEW SOURCE PERFORMANCE STANDARDS AND CONTROL
TECHNOLOGY GUIDELINES 16-21
16.5 GENERAL PROVISIONS 16-25
16.6 PROJECT XL 16-26
17.0 SCHEDULE ISSUES 17-1
17.1 RULEMAKING SCHEDULE 17-1
17.2 COMPLIANCE SCHEDULE ' 17-2
18.0 MECHANICAL PULPING MILL, SECONDARY FIBER
PULPING MILL, NON-WOOD FIBER PULPING MILL,
AND PAPERMAKING SYSTEM (MACT III) COMMENTS 18-1
19.0 MISCELLANEOUS COMMENTS 19-1
19.1 INFORMATION OMITTED FROM THE DOCKET 19-1
19.2 ADDITIONS OR CORRECTIONS TO THE PROPOSED RULE . . 19-1
19.3 OTHER COMMENTS 19-2
20.0 ENVIRONMENTAL AND COST IMPACTS 20-1
20.1 INTRODUCTION 20-1
20.1.1 Section Organization 20-2
20.2 PROCESS DESCRIPTIONS AND EMISSIONS ESTIMATES . . 20-3
20.2.1 Industry Characterization 20-3
20.2.2 System Approach to Emissions Estimation . 20-4
20.2.3 Baseline Emissions 20-9
20.3 EMISSION CONTROL TECHNIQUES 20-12
20.3.1 Applicable Control Techniques for Vents . 20-15
20.3.2 Applicable Control Technologies for
Pulping
Process Condensates 20-17
viii
-------
TABLE OF CONTENTS
20.4 CONTROL OPTIONS AND ENVIRONMENTAL IMPACTS . . . . 20-19
20.4.1 Control Options 20-20
20.4.2 Environmental Impacts 20-22
20.5 ESTIMATED CONTROL COSTS 20-34
20.5.1 Control Cost Methodology 20-34
20.5.2 National Costs for the Control Options . 20-43
20.6 DATA BASE SYSTEM FOR ESTIMATING NATIONAL IMPACTS 20-43
20.6.1 Data Base Revisions 20-43
20.6.2 Calculation of Impacts 20-47
20.6.3 Effects of OW Changes on Impacts .... 20-49
20.7 SUMMARY OF ENVIRONMENTAL AND COST IMPACTS OF THE
MACT STANDARDS 20-50
REFERENCES 20-53
IX
-------
This page intentionally left blank.
X
-------
LIST OF TABLES
Table Page
1-1 LIST OF COMMENTERS ON THE DECEMBER 17, 1993
PROPOSED RULE 1-9
1-2 LIST OF COMMENTERS ON MARCH 8, 1997
SUPPLEMENTAL NOTICE 1-25
1-3 LIST OF COMMENTERS ON MACT III SOURCES 1-27
1-4 NATIONAL COST IMPACTS FOR NESHAP 1-47
20-1 DISTRIBUTION OF CHEMICAL AND SEMI-CHEMICAL PULP
PROCESSES IN THE UNITED STATES 20-5
20-2 EMISSIONS FROM AFFECTED SOURCES FOR AN EXAMPLE MILL . 20-13
20-3 SUMMARY OF ESTIMATED NATIONAL BASELINE EMISSIONS . . . 20-14
20-4 CONTROL OPTIONS FOR EXISTING PULPING SOURCES 20-21
20-5 CONTROL OPTIONS FOR NEW PULPING SOURCES 20-21
20-6 PRIMARY AIR IMPACTS FOR AN EXAMPLE MILL 20-24
20-7 NATIONAL EMISSIONS AND REDUCTIONS FROM THE MACT, AFTER
APPLYING OW OPTION A 20-25
20-8 NATIONAL EMISSIONS AND REDUCTIONS FROM THE MACT, AFTER
APPLYING OW OPTION B 20-27
20-9 SUMMARY OF TYPES OF ENERGY REQUIREMENTS OF MACT
AND OW STANDARDS 20-29
20-10 ENERGY IMPACTS FOR AN EXAMPLE MILL 20-31
20-11 NATIONAL ENERGY IMPACTS AFTER MACT
AND OW OPTIONS 20-32
XI
-------
LIST OF TABLES
Table Page
20-12 SECONDARY AIR IMPACTS FOR AN EXAMPLE MILL 20-35
20-13 SECONDARY AIR EMISSIONS AND ENERGY IMPACTS FROM APPLYING
THE MACT AND OW OPTION A TO THE CURRENT BASELINE . 20-36
20-14 SECONDARY AIR EMISSIONS AND ENERGY IMPACTS FROM APPLYING
THE MACT AND OW OPTION B TO THE CURRENT BASELINE . . 20-38
20-15 SUMMARY OF NATIONAL COST AND COST EFFECTIVENESS OF MACT
CONTROL OPTIONS AFTER OW OPTION A 20-44
20-16 SUMMARY OF NATIONAL COST AND COST EFFECTIVENESS
OF MACT CONTROL OPTIONS AFTER OW OPTION A 20-45
20-17 TOTAL MILLS AND EMISSIONS; PULP AND PAPER
NATIONWIDE ENVIRONMENTAL AND COST IMPACTS 20-51
XI1
-------
1.0 SUMMARY
1.1 INTRODUCTION
On December 17, 1993 (58 FR 66078), the U.S. Environmental
Protection Agency (EPA) jointly published proposed National
Emission Standard for Hazardous Air Pollutants (NESHAP) and
Effluent Guidelines (December 17, 1993 proposed rule) for the
pulp and paper industry. The rule proposed standards for MACT I
sources, which include kraft, soda, sulfite, and semi-chemical
pulping mills. On March 17, 1994 (59 FR 12567) EPA published a
correction notice to the proposed NESHAP and Effluent Guidelines
(March 17, 1994 correction notice) . On February 22, 1995
(60 FR 9813) EPA published a Notice of Data Availability (NODA)
that would be considered for developing the promulgated NESHAP.
On March 8, 1996 (61 FR 9383), EPA published a supplemental
notice (March 8, 1996 supplemental notice) that presented EPA's
assessment of the additional data and information obtained after
proposal and announced potential changes to the proposed rule.
As part of the March 8, 1996 supplemental notice, EPA also
proposed standards for MACT III sources (papermaking systems,
mechanical pulping mills, secondary fiber pulping [deinked and
non-deinked] mills, and non-wood mills, and asked for additional
information on these mills). MACT II sources (combustion
sources) are covered under a separate rulemaking.
1-1
-------
In the March 8, 1996 supplemental notice, EPA solicited
additional data and comments on proposed changes to the
December 17, 1993 proposed rule. Data added to Air
Docket A-92-40 since the March 8, 1996 supplemental notice are
located in section IV of the docket. These items include
additional information on sulfite mills (IV-D1-98, IV-D1-100)
comments on definitions (IV-D1-97, IV-D1-99, IV-D1-104), comments
on the emission factor document (IV-D1-102), clarification of the
1992 MACT survey responses (IV-D1-101) , and other information.
The public comment period for the December 17, 1993 proposed
rule was from December 17, 1993 to March 17, 1994. Approximately
155 comment letters were received on the December 17, 1993
proposed rule. The public comment period for the supplemental
notice was from March 8, 1996 to April 8, 1996. Approximately
33 comment letters were received on the March 8, 1996
supplemental notice, including letters received on the MACT III
sources. Comments were provided by industry representatives,
governmental entities, environmental groups, and private
citizens.
1.2 ORGANIZATION OF THIS DOCUMENT
This introduction includes the list of commenters on the
December 17, 1993 proposed rule and notices. In order to present
the comments in a logical manner, the comments and EPA's
responses have been divided into 18 categories. The categories
and respective chapter numbers in this background information
document are as follows:
2.0 INDUSTRY CHARACTERIZATION
3.0 SUBCATEGORIZATION
4.0 BASIS OF STANDARDS
5.0 PULPING AREA
6.0 BLEACHING AREA
1-2
-------
7.0 PROCESS WASTEWATER AREA
8.0 MONITORING
9.0 TEST METHODS AND PROCEDURES
10.0 RECORDKEEPING AND REPORTING
11.0 COST/ECONOMIC IMPACTS
12.0 BENEFITS
13.0 EMISSIONS AVERAGING
14.0 DEFINITIONS
15.0 CLUSTER RULE INTERACTION
16.0 INTERACTION WITH OTHER RULES
17.0 SCHEDULE ISSUES
18.0 MACT III
19.0 MISCELLANEOUS COMMENTS
20.0 ENVIRONMENTAL IMPACT STATEMENT
The environmental impact statement is chapter 20 of this
document.
The following section, section 1.3, includes tables listing
the commenters, their affiliation, and assigned comment number.
Numerous acronyms appear throughout this document. The
following is provided for reference.
Acronyms Used in this Document
Act Clean Air Act
ADP Air-dried pulp
ADTP Air-dried ton of pulp
AF&PA American Forest and Paper Association (formerly
the American Paper Institute)
AP-42 Compilation of Air Pollutant Emission Factors,
5th edition, Volume 1: Stationary Point and Area
Sources
API American Paper Institute
AQRV Air quality related value
1-3
-------
BAT Best available technology (under the Effluent
Limitation Guidelines and Standards of the Clean
Water Act)
BACT Best available control technology
BID Background information document
BIF Boilers and industrial furnaces
BLO Black liquor oxidation
BMP Best management practices
BOD Biochemical oxygen demand
BODs Biochemical oxygen demand 5-day test
BPT Best practicable control technology
Btu British thermal unit
GEMS Continuous Emissions Monitoring Systems
CCA Clean condensate alternative
CFR Code of Federal Regulations
C1C>2 Chlorine dioxide
CO Carbon monoxide
Q02 Carbon dioxide
CTG Control Technology Guidance
CWA Clean Water Act
EA Economic Analysis for the National Emission
Standards for Hazardous Air Pollutants for Source
Category: Pulp and Paper Production; Effluent
Limitations and Guidelines, Pretreatment Standards
and New Source Performance Standards; Pulp, Paper,
and Paperboard Category - Phase I
1-4
-------
EPA U.S. Environmental Protection Agency
FOIA Freedom of Information Act
FLM Federal land manager
FR Federal Register
GC/FID Gas chromatography/flame ionization detection
GRM Gallons per minute
HAP Hazardous air pollutant
HC1 Hydrogen chloride
HON Hazardous organic NESHAP
HVLC High volume, low concentration collection system
IRIS Integrated Risk Information System
Ib/ODTP Pound per oven-dried ton of pulp
kg Kilograms
LAER Lowest achievable emission rate
LVHC Low volume, high concentration collection system.
MACT Maximum achievable control technology.
Mg Megagram
NAA Non-attainment Area
NAAQS National Ambient Air Quality Standards
NCG Noncondensible gas
NCASI National Council of the Paper Industry for Air and
Stream Improvement
1-5
-------
NESHAP
NIOSH
NOCEPM
NODA
NOX
NPDES
NSPS
NSR
OAQPS
OCCM
ODP
ODTP
OMB
ORD
OSHA
OTR
OW
PCP
PM
ppmv
National Emission Standard for Hazardous Air
Pollutants
National Institute Occupational Safety and Health
NCASI Organic Compound Elimination Pathway Model
Notice of Data Availability
Nitrogen oxides
National Pollutant Discharge Elimination System
New Source Performance Standards
New Source Review
Office of Air Quality, Planning and Standards
OAQPS Control Cost Manual
Oven-dried pulp
Oven-dried ton of pulp
Office of Management and Budget
Office of Research and Development
Occupational Safety and Health Administration
Ozone Transport Region
Office of Water
Pollution control project
Particulate matter
Particulate matter less than 10 microns mean
aerodynamic diameter
parts per million by volume
1-6
-------
ppmw parts per million by weight
PSD Prevention of Significant Deterioration
PSES Pretreatment Standards for Existing Sources
PSNS Pretreatment Standards for New Sources
RACT Reasonable Attainable Control Technology
RCRA Resource Conservation and Recovery Act
RIA Regulatory Impacts Analysis
SCR Selective catalytic reduction
SIC Standard Industrial Classification
SIP State Implementation Plan
SNCR Selective non-catalytic reduction
302 Sulfur dioxide
SOCMI Synthetic Organic Chemical Manufacturing Industry
TCF Totally chlorine free
SCF Secondarily chlorine free
TRE Total resource effectiveness
TRS Total reduced sulfur
VOC Volatile organic compounds
1.3 SUMMARY OF PUBLIC COMMENTS
Approximately 188 written comments were received on the
proposed standards and subsequent notices. A list of the
commenters on the December 17, 1993 proposed rule, their
affiliations, and the EPA docket number assigned to their
correspondence is given in table 1-1. Table 1-2 lists all
1-7
-------
persons submitting general written comments on the March 8, 1996
supplemental notice, their affiliations, and the docket item
number assigned to each correspondence. Table 1-3 lists
separately all persons submitting written comments on the
MACT III sources (papermaking systems, mechanical pulping mills,
secondary fiber pulping mills, and non-wood mills), their
affiliation, and the docket item number assigned to each
correspondence.
1.4 SUMMARY OF CHANGES SINCE PROPOSAL
In response to comments received on the proposed standards,
several changes were made to the final rule. While some of these
changes are clarifications designed to make the Agency's intent
clearer, a number of them are significant changes to the proposed
standard requirements. A summary of the substantive comments and
changes made since the proposal are described in this section.
Detailed Agency responses to public comments are presented in
chapters 2.0 through 19.0 of this document. The revised analyses
for the final rule are in the public docket.
1.4.1 Definition of Source
The December 17, 1993 proposed rule presented a broad source
definition which included pulping processes, bleaching processes,
and pulping and bleaching process condensates. The Agency
specifically requested comment on the source definition (i.e.,
broad versus narrow) in the proposal. After considering
comments, EPA adopted the broad definition in the final rule.
The EPA determined that the affected source is all emission
points in the pulping and the bleaching systems (including oxygen
delignification and the pulping condensate system). The final
rule explicitly defines the new source MACT applicability by
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE
Office of Water
docket control
number
Commenter and affiliation
20,000
20,001
20,002
Attachment
20,002A1
20,003
20,004
20,005
20,006
20,007
20,008
20,009
Tom Burgess
Chemetics International, Inc.
Vancouver, British Columbia, Canada
Wayne E. Glenn
United Paperworkers International Union
Nashville, Tennessee
Gordon D. Strickland
Chemical Manufacturers Association
Washington, DC
Richard M. Harvey
Florida Department of Environmental
Protection
Tallahassee, Florida
David W. Schmutzler
Niagara of Wisconsin Paper Corporation
Niagara, Wisconsin
David A.C. Carroll
Maryland Department of the Environment
Baltimore, Maryland
Citizen
Nat Hendricks
Putney, Virginia
Carl W. Ehmann
RJ Reynolds Tobacco Company
Winston Salem, North Carolina
Robert C. Steidel
Environmental Manager
City of Hopewell
Hopewell, Virginia
Dianne M. Reid
State of North Carolina Department of
Environment, Health, and Natural Resources
Raleigh, North Carolina
1-9
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,010
20,011
20,012
Attachment
20,012A1
20,013
20,014
Attachment
20,014A1
20,015
20,016
Attachments
20,016A1 -
20,016A6
20,017
20,018
Attachments
20,018A1 -
20,018A2
20,019
John W. Walton
State of Tennessee Department of Environment
and Conservation
Nashville, Tennessee
Gregory J. Hollod
Riverwood International
Atlanta, Georgia
David J. Lutrick
Simpson Paper Company
Anderson, California
Paul A. Walker
Hollingsworth & Vose Company
East Walpole, Massachusetts
A.D. Whitford
Longview Fibre Company
Longview, Washington
Thomas R. Hewitt
CRS Sirrene Environmental
Raleigh, North Carolina
Kathy E. Gill
Northwest Pulp & Paper
Bellevue, Washington
Seattle Audubon Society
Seattle, Washington
Kimberly A. Hughes
Weyerhaeuser
Tacoma, Washington
J.R. Nein
Chesapeake Paper Products Company
West Point, Virginia
1-10
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,020
Attachment
20,020A1
20,021
20,022
20,023
20,024
20,025
Attachments
20,025A1 -
20,025A10
20,026
Attachments
20,026A1 -
20,026A51
20,027
Attachments
20,027A1 -
20,027A32
20,028
20,029
Robert G. Smerko
The Chlorine Institute
Washington, DC
Kenneth L. Wendell
Westvaco
Luke, Maryland
Marianne Dugan
Western Environmental Law Center, Inc.
Eugene, Oregon
Herbert C. Scribner
Van Leer Packaging
Keyes Fibre Company
Waterville, Maine
Josephine S. Cooper
American Forest and Paper Association
Washington, DC
Josephine S. Cooper
American Forest and Paper Association
Washington, DC
American Forest and Paper Association
Washington, DC
American Forest and Paper Association
Washington, DC
Guy R. Griffin
Potlatch Corporation
San Francisco, California
Michael J. Wax
Institute of Clean Air Companies
Washington, DC
1-11
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,030
20,031
20,032
20,033
20,034
Attachments
20,034A1 -
20,034A6
20,035
20,036
Attachment
20,036A1
20,037
20,038
20,039
William Robert Neff
The Upper Potomac River Commission
Westernport, Maryland
C.L. Missimer
P.H. Glatfelter Company
Spring Grove, Pennsylvania
Douglas C. Pryke
Alliance for Environmental Technology
Ontario, Canada
George A. Schmitt
3M Industrial & Consumer Sector
St. Paul, Minnesota
Robert B. Burns Jr.
Albert H. Toma III
Fort Howard Corporation
Green Bay, Wisconsin
C.F. Bledsoe
Alabama Pulp and Paper Council
Montgomery, Alabama
Nicholas J. Lardieri
Scott Paper Company
Philadelphia, Pennsylvania
Leslie Ritts
Counsel for American Forest and Paper
Association
Chadbourne & Parke
Washington, DC
Wilson Blackburn
Lake Superior Paper Industries
Duluth, Minnesota
J. Carter Fox
President and CEO
Chesapeake
Richmond, Virginia
1-12
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,040
20,041
20,042
20,043
20,044
Attachment
20,044A1
20,045
20,046
Attachments
20,046A1 -
20,046A2
20,047
Attachment
20,047A1
20,048
Attachment
20,048A1
Douglas A. Hall
Minnesota Pollution Control Agency
St. Paul, Minnesota
Raymond J. Connor
Technical Director
Manufacturers of Emission Controls
Association
Washington, DC
L.J. Achee, Jr.
Jackson City Port Authority
Pascagoula, Mississippi
Robert J. Sistko, PhD, Sr.
Environmental Specialist
Rayonier
Shelton, Washington
Kurt N.W. Soderberg
Western Lake Superior Sanitary District
Duluth, Minnesota
M.T. Fisher
Proctor & Gamble
Cincinnati, Ohio
R.E. Cannon
Buckeye Florida
Perry, Florida
R.E. Cannon
Buckeye Cellulose Corporation
Memphis, Tennessee
Kenneth T. Hood
Simpson Paper Company
Anderson, California
1-13
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,049
Attachments
20,049A1 -
20,049A2
20,050
20,051
20,052
Attachment
20,052A1
20,053
Attachment
20,053A1
20,054
Attachments
20,054A1 -
20,054A5
20,055
20,056
Attachments
20,056A1 -
20,056A3
20,057
Attachments
20,057A1 -
20,057A22
20,058
Jessica C. Landman
Natural Resources Defense Council
Washington DC
Lindsay M. Lancaster
International Paper Company
Mobile, Alabama
Luigi Terziotti
Alabama River Pulp Company
Perdue Hill, Alabama
Luigi Terziotti
Alabama River Pulp Company
Perdue Hill, Alabama
James Miller
Louisiana-Pacific Corporation
Samoa, California
Richard Diforio
Champion International
Stamford, Connecticut
Erick Tokar
Rayonier
Shelton, Washington
Duane Marshall
Union Camp
Savannah, Georgia
Thomas Jorling
International Paper Company
Purchase, New York
Catherine Marshall
American Forest and Paper Association
Washington, DC
1-14
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,059
20,060
20,061
20,062
20,063
20,064
Attachments
20,064A1 -
20,064A3
20,065
20,066
Attachments
20,066A1 -
20,066A4
20,067
20,068
Attachments
20,068A1 -
20,068A7
Deborah A. Sheiman
Natural Resources Defense Council
Washington, DC
Jerry Pardilla
Penobscot Indian Nation
Old Town, Maine
David Lutrick
Simpson Paper Company
Seattle, Washington
Robert Collez
Augusta Newsprint
Augusta, Georgia
Peter Washburn
Natural Resources Council of Maine
Augusta, Maine
Greenpeace
Washington, DC
Joe Thornton
Greenpeace
Washington, DC
Charles Ackel
Stone Container Corporation
Tucker, Georgia
Roger Stone
Stone Container Corporation
Chicago, Illinois
Dana Dolloff
Rayonier
Stamford, Connecticut
1-15
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,069
Attachments
20,069A1 -
20,069A10
20,070
Attachments
20,070A1 -
20,070A15
20,071
Attachments
20,071A1 -
20,071A13
20,072
Attachments
20,072A1 -
20,072A11
20,073
20,074
20,075
amends
20,057
20,076
amends
20,045
20,077
Attachments
20,077A1 -
20,077A7
Douglas Walsh
Lincoln Pulp & Paper Company
Lincoln, Maine
Kathleen Bennett
James River Corporation
Richmond, Virginia
Dale Phenicie
Georgia-Pacific
Atlanta, Georgia
Jerome Tatar
Mead
Chillicorne, Ohio
Russell Frye
Chadbourne & Parke
Washington, DC
Steve Mason
Mead
Dayton, Ohio
Alan Lindsay
International Paper
Memphis, Tennessee
Henry Clifford
Proctor & Gamble
Cincinnati, Ohio
Corrine Goldstein
Covington & Burling
Washington, DC
1-16
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,078
20,079
Attachments
20,079A1 -
20,079A4
20,080
20,081
Attachments
20,081A1 -
20,081A2
20,082
20,083
20,084
20,085
20,086
20,087
Greg Sorlie
Washington Department of Ecology
Olympia, Washington
Charles Bridges
Van Leer Packaging
Waterville, Maine
Mary O'Brien
Environmental Research Foundation
Annapolis, Maryland
Rick Montanari
Ecotech
St. Petersburg, Florida
Martin Visnosky
Erie County Environmental Coalition
Erie, Pennsylvania
Steve Kilpatrick
Dow
Midland, Michigan
Gordon Strickland
Chemical Manufacturers Association
Washington, DC
Norman Anderson
American Lung Association of Maine
Augusta, Maine
Darrell Jeffries
Wausau Papers
Brokaw, Wisconsin
Brian Benson
Roy F. Weston, Inc.
Auburn, Alabama
1-17
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
on n Q Q
20, 0 o o
Attachment
20,088A1
20,089
20,090
20,091
Attachments
20,091A1 -
20,091A5
20,092
20,093
20,094
20,095
Attachments
20,095A1 -
20,095A9
Kenneth Gilbreath
Chesapeake Paper Products
West Point, Virginia
David Buente
Sidley & Austin
Washington, DC
Ted Strong
Columbia River Inter-Tribal Fish Commission
Portland, Oregon
Donna Hayes
Dickinson Citizens for Clean Air
Norway, Michigan
David Driesen
Natural Resources Defense Council
Washington, DC
Bharat Shah
Wisconsin Tissue
Menasha, Wisconsin
Art Vosburg
Pope & Talbot
Halsey, Oregon
Reid A. Miner
National Council of Paper Industry for Air
and Stream Improvement
New York, New York
20,096
Attachments
2 0, 0 9 6A1
20,097
Dick Brown
Gulf Coast Waste Disposal Authority
Houston, Texas
Joe Mayhew
Chemical Manufacturers Association
l-U
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,098
Also A-92-40
IV-D1-18
20,099
Attachments
2 0,0 9 9A1 -
20,09 9A2
20,100
20,101
20,102
20,103
20,104
20,105
20,106
20,107
John Pinkerton
National Council of Paper Industry for Air
and Stream Improvement
New York, New York
G.W. Zielinski
City of St. Helens, Oregon
attachments not sent to OAQPS
Stewart Thomas
Newsprint South Inc.
Grenada, Mississippi
Stacy Palamatary
Oxychem
Dallas, Texas
Robert Colby
STAPPA/ALAPCO
Washington, DC
Susan Sylvester
Wisconsin Department of Natural Resources
Madison, Wisconsin
William Nicholson
No company affiliation disclosed
Ross, California
Nicholas J. Lardieri
Scott Paper Company
Philadelphia, Pennsylvania
Corinne Goldstein
Covington & Burling on behalf of
Finch Pruyn & Company
Glensfalls, New York
Reid A. Miner
National Council of Paper Industry for Air
and Stream Improvement
New York, New York
1-19
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,10?
20,109
20,110
20,111
Attachment
20,111A1
20,112
Attachments
20,112A1 -
20,112A11
20,113
Attachment
20,113A1
20,114
Attachment
20,114A1
20,115
Attachments
20,115A1 -
20,115A5
20,116
Attachments
20,116A1 -
20,116A2
20,117
Attachment
20,117A1
Guy Griffin
Potlatch Corporation
San Francisco, California
Paul Wiegard
National Council of Paper Industry for Air
and Stream Improvement
New York, New York
Peter Baljet
American Lung Association
Washington, DC
Dennis Keschl
Maine Department of Environmental Protection
Augusta, Maine
John Festa
American Forest and Paper Association
Washington, DC
Terry Cole
St. Joe Forest Products Comapny
City of Port St. Joe, Florida
Kenneth A. Strassner
Kimberly-Clark
Washington, DC
James Season
Appleton Papers Inc.
Appleton, Wisconsin
John Festa
American Forest and Paper Association
Washington, DC
John Millican
Florida Pulp & Paper Association
Tallahassee, Florida
1-20
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,11!
20,119
20,120
Attachments
20,120A1 -
20,120A88
20,121
20,122
Attachments
20,122A1 -
20,122A7
20,123
Attachments
20,123A1 -
20,123A7
20,124
20,125
20,126
20,127
Attachment
20,127A1
Anthony Gammie
Bowater, Inc.
Greenville, South Carolina
Robert C. Kaufmann
American Forest and Paper Association
Washington, DC
Josephine S. Cooper
American Forest and Paper Association
Washington, DC
Washington Toxics Coalition and
41 other Environmental Organizations in the
Pacific Northwest
Prepared by Carol Dansereau, J.D.
Director of Washington Toxics Coalition
Seattle, Washington
Dale Phenicie
Georgia-Pacific Corporation
Atlanta, Georgia
Frank Pate
City of Port St. Joe, Florida
Catherine Marshall
American Forest and Paper Association
Washington, DC
Jessica C. Landman, Senior Attorney;
Diane M. Cameron, Environmental Engineer;
Brian L. Doster, Legal Associate
Natural Resources Defense Council
New York
Joy Cummings
HOPE in Taylor Company
Perry, Florida
1-21
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,12!
20,129
20,130
Attachments
20,130A1 -
20,130A5
20,131
20,132
20,133
20,134
20,135
Attachment
20,135A1
20,136
Attachments
20,136A1 -
20,136A14
20,137
Frank Molen
Commonwealth of Virginia Senate
New Hope, Virginia
Edward Sullivan
New York State Department of Environment and
Conservation
Albany, New York
David Lutrick
Simpson Paper Company
Anderson, California
Duane Marshall
Union Camp
Savannah, Georgia
Mr. and Mrs. James J. Sloan
Salinas, California
Phillir Chaudoir
Green Bay, Wisconsin
Albert Toma
Fort Howard
Green Bay, Wisconsin
Kathleen M. Bennett
James River Co.
Richmond, Virginia
Mark Haley
City of Hopewell, Virginia
Alan D. Whitford
Longview Fibre Company
Longview, Washington
1-22
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,138
20,139
20,140
20,141
20,142
20,143
20, 144
20,145
20,146
20,147
20, 148
Chester Williams
FUSE, Inc.
Texarkana, Arkansas-Texas
James W. Riley
Cumberland, Maryland
Jim Anders
Anders Real Estate & Timber Co., Inc.
Blountstown, Florida
Karey Shaw
Columbia River United
Hood River, Oregon
Stuart I. Gansell
Pennsylvania Department of
Environmental Resources
Joy Huber
Rivers Council of Washington
Seattle, Washington
Robert H. Collom, Jr.
Georgia Department of Natural Resources
Atlanta, Georgia
Randy Thurman
Arkansas Environmental
Little Rock, Arkansas
Federation
J.D. Weinbauer
Consolidated Papers, Inc.
Wisconsin Rapids, Wisconsin
Bruce W. Beckstrom
A.H. Lundberg Associates, Inc.
Bellevue, Washington
Richard A. Samp
Washington Legal Foundation
Washington, DC
1-23
-------
TABLE 1-1. LIST OF COMMENTERS ON THE
DECEMBER 17, 1993 PROPOSED RULE (Continued)
Office of Water
docket control
number
Commenter and affiliation
20,149
20,150
20,151
20,152
20,153
20,154
20,155
20,156
25,538
Paul Gerbec
Minnesota Pollution Control Agency
St. Paul, Minnesota
Bob Jackman and Frank Ossiander
Citizens for a Clean Columbia
Kettle Falls, Washington
Randal S. Telesz
Michigan Department of Natural Resources
Lansing, Michigan
Stephen B. Letendre
Tennessee Department of Environment
and Conservation
Nashville, Tennessee
Samuel N. Penney
Nez Perce Tribal Executive Committee
Lapwai, Idaho
Paul C. Martyn
Los Angeles County Sanitation Districts
Whittier, California
Yogesh M. Mehta
Brown & Root U.S.A., Inc.
Houston, Texas
Edward Mudd, Jr.
Birmingham, Alabama
Josephine S. Cooper
American Forest & Paper Association
Washington, DC
1-24
-------
TABLE 1-2. LIST OF COMMENTERS ON MARCH 8, 1997
SUPPLEMENTAL NOTICE
Item number in
Docket A-92-40 Commenter and affiliation
IV-D2-2 Keith M. Bentley
Georgia Pacific
Savannah, Georgia
IV-D2-3 Duane W. Marshall
Union Camp
Atlanta, Georgia
IV-D2-4 Donald F. Theiler
STAPPA/ALAPCO
Washington, DC
IV-D2-5 William 0. Dameworth
Pope & Talbot
Halsey, Oregon
IV-D2-6 Robert J. Sistko
Rayonier
Shelton, Washington
IV-D2-7 Gregory J. Hollod
Riverwood International
Atlanta, Georgia
IV-D2-8 Kathleen M. Bennett
James River Corporation
Richmond, Virginia
IV-D2-9 K.E. Lewis
Proctor & Gamble
Cincinnati, Ohio
IV-D2-10 Dan Pearson
Texas Natural Resource Defense Council
Austin, Texas
IV-D2-11 Thomas C. Jorling
International Paper
Purchase, New York
IV-D2-12 Phillip J. Arthur
Finch Pruyn & Co., Inc.
Glen Falls, New York
1-25
-------
TABLE 1-2. LIST OF COMMENTERS ON MARCH 8, 1997
SUPPLEMENTAL NOTICE (Continued)
Item number in
Docket A-92-40 Commenter and affiliation
IV-D2-13 Josephine Cooper
American Forest and Paper Association
Washington, DC
IV-D2-14 Sara S. Kendall
Weyerhaeuser
Tacoma, Washington
IV-D2-15 Robert C. Kaufmann
American Forest and Paper Association
Washington, DC
IV-D2-16 Richard C. Abrams
Kimberly-Clark
Everett, Washington
IV-D2-17 Dana B. Dolloff
Rayonier
Stamford, Connecticut
IV-D2-18 R.E. Cannon
Buckeye Florida
Perry, Florida
IV-D2-19 Larry Tenth
Chemitics
Vancouver, British Columbia, Canada
IV-D2-20 Duane W. Marshall
Union Camp
Savannah, Georgia
1-26
-------
TABLE 1-3. LIST OF COMMENTERS ON MACT III SOURCES3
Item number in
Docket A-95-31 Commenter and affiliation
IV-D-1 K.E. Lewis
The Proctor & Gamble Company
Cincinnati, Ohio
IV-D-2 J. Grumet and W. Cass
Northeast States for Coordinated
Air Use Management
Boston, Massachusetts
IV-D-3 R.H. Colby and D.F. Theiler
State and Territorial Air Pollution
Programs Administrators/Association of
Local Air Pollution Control Officials
Washington, DC
IV-D-4 K.M. Bennet
James River Corporation
Richmond, Virginia
IV-D-5 G.J. Hollod
Riverwood International
Atlanta, Georgia
IV-D-6 J. Brooks
State of Maine
Augusta, Maine
IV-D-7 J.S. Cooper
American Forest & Paper Association
Washington, DC
IV-D-8 R.C. Kaufmann
American Forest & Paper Association
Washington, DC
IV-D-9 R.E. Cannon
Buckeye Cellulose Corporation
Memphis, Tennessee
IV-D-10 T. Mattson
Environmental Technology - Air
Fort Howard
Green Bay, Wisconsin
1-27
-------
TABLE 1-3. LIST OF COMMENTERS ON MACT LLL SOURCES3
(Continued)
Item number in
Docket A-95-31 Commenter and affiliation
IV-D-11 C. Ackel
Stone Container Corp.
Tucker, Georgia
IV-D-12 S.S. Kendall
Weyerhauser Corp.
Taucoma, Washington
IV-D-13 R.A. Ellis, J.H. Lewis, R.J. Hampson,
L.J. Barry (et al.), P.J. Luciano,
W.J. Schulz, L. Gill, and
R.J. Ellithorpe. Letters to
Elaine Manning (OAQPS/EPA)
aMACT III sources include papermaking systems, mechanical pulping
mills, secondary fiber pulping mills, and non-wood mills.
1-2!
-------
specifying the control requirements for (1) greenfield sites,
(2) the addition of new equipment at existing sources, and
(3) changes to existing equipment that could trigger
reconstruction. By designating the exact equipment to be
controlled at new and existing sources, the rule reduces
confusion and misinterpretation over what actions trigger new
source requirements. This approach preserves the advantages of a
broad source definition for compliance by existing sources while
ensuring that new and reconstructed equipment are regulated as
new sources consistent with Section 112(a) and 112(d) of the
Clean Air Act (Act) .
The final rule also provides for an alternative definition
of source for use with the clean condensate alternative (CCA).
For mills using the CCA (see section 1.4.4.2) to comply with the
standards, the broad definition includes all the pulping,
bleaching, causticizing, and paper making systems. These
additions were made to the definition of affected source to
encourage pollution prevention since the paper making and
causticizing systems typically receive recycled or reused
condensates.
1.4.2 Subcategories
In the proposed rule, no distinction was made between the
different types of pulping processes. The standards for control
of hazardous air pollutant (HAP) emissions from vents and
wastewater sources (i.e., pulping process condensates) were the
same for kraft, semi-chemical, soda, and sulfite pulping
processes. After evaluating public comments and data received
following proposal, EPA established separate subcategories for
1-29
-------
kraft, semi-chemical, soda, and sulfite pulping processes due to
differences in process emissions and applicable control
techniques. As in the proposed rule, the final standards for
kraft, semi-chemical, and soda pulping processes in the final
rule are based on combustion. For sulfite pulping processes, the
final rule is based on absorption technologies.
1.4.3 Control Applicabilitv Determination
The proposed rule prescribed applicability cutoff values
(i.e., volumetric flow rate and mass flow rate) as a way to
specify the vent and condensate streams that would be required to
meet the rule. Since proposal, the pulp and paper industry
submitted additional data that allowed EPA to better characterize
the vent and condensate streams that should be controlled.
In the final rule, the applicability cutoff values contained
in the December 17, 1993 proposed rule have been replaced in
favor of specifically naming the vent and condensate streams that
would be required to meet the rule for each subcategory, with the
exception of decker, knotter, and screen systems at kraft pulping
mills. For these systems, the rule specifies applicability
cutoffs in the form of emission limits (knotter and screen
systems) and HAP concentration in process condensates (decker
systems) to identify the systems that should be controlled at new
and existing mills.
The different approach used in the final rule does not
significantly change the stringency or scope of the
December 17, 1993 proposed rule. The emission points and
condensate streams that are being controlled in the final rule
are fundamentally the same emission sources that EPA intended to
be controlled in the December 17, 1993 proposed rule. The
1-30
-------
revised approach is easier and less costly to implement, for both
the affected industry and the enforcement officials, since
extensive emission source testing is not required to identify the
vent and condensate streams to be controlled.
1.4.4 Kraft Standards
1.4.4.1 Applicabilitv for Existing Pulping Kraft Sources.
In the December 17, 1993 proposal all pulping vent emission
points were, with some exceptions, reguired to be enclosed and
vented to a closed-vent system and routed to a control device
that achieves 98 percent destruction. The exceptions were for
deckers and screens at existing mills and small vents below
specified volumetric and mass flow rates. Pulping wastewater
streams with HAP concentrations below 500 parts per million by
weight (ppmw) and flow rates below 1.0 liter per minute did not
reguire control.
In the final rule, specific vent and condensate streams
are reguired to be controlled. For existing sources, the vent
emission sources include: the low volume, high concentration
collection (LVHC) system, pulp washing system, decker system,
oxygen delignification system, knotter system, and screening
system. The EPA based its decisions to reguire these systems to
be controlled on information presented in responses to industry
surveys used to characterize controls that are installed at
existing mills and in comments to the proposed rule.
Based on analysis of additional information provided by
industry, the final rule does not reguire the control of existing
weak black liguor storage tanks or control of decker systems that
use clean water or process water from papermaking systems ("white
1-31
-------
water"). These types of process water are defined as streams
with HAP concentrations less than or equal to 400 ppmw.
Also, in the final rule, existing sources are required to
control knotters with mass emission rates greater than 0.1 pounds
of HAP per oven dried ton of production (Ib/ODTP) (0.05 kilograms
per megagram), screens with mass emission rates greater than
0.2 Ib/ODTP (0.10 kilograms per megagram), or combined knotter
and screen systems with emissions greater than 0.3 Ib/ODTP
(0.15 kilograms per megagram). New sources are required to
control all decker, knotter, and screen systems and weak black
liquor storage tanks.
Condensate Segregation. The proposed standards for process
wastewater required that all pulping wastewaters that met the
applicability criteria had to be treated by one of several
specified control options. Comments and data submitted to EPA
indicated that kraft mills typically steam strip the condensates
from the digester, turpentine recovery, LVHC, and high volume,
low concentration collection (HVLC) system, and certain
evaporator system condensates. The data also indicated that
mills that use steam strippers also practice varying degrees of
condensate segregation in order to reduce treatment costs by
minimizing the flow rate and maximizing the methanol mass of
streams sent to treatment.
To allow this cost saving option, the final rule requires
that the entire volume of condensate generated from the named
pulping process equipment must be treated unless the condensates
from the digester system, turpentine recovery system, and the
weak liquor feed stages in the evaporator system are segregated.
If these condensates are adequately segregated, only the high-HAP
1-32
-------
fraction stream from these systems, along with the condensates
from the LVHC and HVLC collection systems, must be sent to
treatment.
The final rule contains two options for demonstrating
compliance with the segregation requirements. The first option
is to isolate 65 percent of the total HAP mass that was present
in the specified system condensate streams prior to segregation.
The second option specifies that a minimum HAP mass from the
digester, turpentine recovery, evaporator, LVHC collection, and
HVLC collection systems be sent to treatment.
1.4.4.2 Clean Condensate Alternative. The December 17,
1993 proposed rule did not contain any provisions for emissions
averaging. Industry comments on the proposal indicated support
for incorporating an emission averaging approach in the final
rule. After the public comment period, industry submitted a
report comparing the emission reductions that could be achieved
using the option developed by industry and emission reductions
that could be achieved using the proposed MACT standards.
Specifically, the industry option is based on comparing the HAP
emission reductions achieved by implementing the alternative
technology with the baseline HAP emission reductions that would
have been achieved by implementing the MACT standards. The
industry option formed the basis for what is referred to as the
CCA in the final rule.
The CCA is an option for compliance with kraft pulping
standards for the HVLC system. As an alternative to combustion
of HVLC vent emissions, a mill may reduce the HAP concentration
in process water that is used in the HVLC process equipment and
in other areas throughout the mill, such as the paper making and
1-33
-------
causticizing systems. By reducing the HAP loading in the process
water, less HAP will be available to be emitted to the
atmosphere. The final rule specifies that the determination of
the baseline HAP emission reductions and the reductions achieved
by the alternative strategy must be determined by emissions
testing data.
1.4.4.3 Biological Treatment. At proposal, one of the
compliance options for process wastewaters was to destroy at
least 90 percent HAP by weight by hard piping the process
wastewater streams to biological treatment. For the performance
test, owners or operators were required to measure inlet and
outlet methanol concentrations using Method 305, and determine
the mass flow rate of total HAP or methanol entering the
biological treatment system. The biological treatment system's
destruction efficiency was determined by dividing the difference
of the outlet and inlet mass flow rates by the inlet mass flow
rate and multiplying by the fraction of methanol removed in the
biological treatment system. The site-specific fraction of
methanol removed in the biological treatment system was
determined using EPA's WATER7 model.
The continuous monitoring requirements specified that total
HAP or methanol concentration be measured at the inlet and outlet
of the biological treatment system every 30 days. Additionally,
the standard required monitoring of appropriate operating
parameters as specified in the operating permit and demonstrated
to the Administrator's satisfaction.
In the final rule, biological treatment systems may still be
used to comply with the pulping process condensate standards,
however, the monitoring procedures have been revised. in the
1-34
-------
final rule, mills using a biological treatment system to treat
pulping process condensates must monitor, on a daily basis,
samples of outlet soluble biochemical oxygen demand 5-day test
(BOD5) concentration (maximum daily and monthly averages), inlet
liguid flow, mixed liquor volatile suspended solids, liquid
temperature, and the horsepower of aerator units. Additionally,
inlet and outlet grab samples from each biological treatment
system unit must be collected and stored for 5 days. These
samples must be collected and retained since some of the
monitoring parameters (e.g., soluble BODs) can not be determined
within a short period of time. These samples are to be used in
conjunction with the WATERS emissions model to demonstrate
compliance if any of the monitoring parameters (except the liquid
temperature and inlet flow) fall outside the range established
during the initial performance test. Additionally, quarterly
percent reduction tests must be performed using the WATERS model
and site-specific inputs. The first quarter test must be
performed for total HAP while the remaining quarterly tests may
be performed for methanol only.
1.4.5 Sulfite Standards
At proposal, all pulping vent streams from sulfite processes
were required to be enclosed and routed to a control device
achieving 98 percent reduction in emissions. In the March 8,
1996 supplemental notice, the Agency discussed in detail its
determination that the sulfite standards should be based on
absorption technology and apply to the total emissions from
specific vents and any wastewater emissions associated with HAP
emission control devices. The specific vents are associated with
the digester, evaporator, and pulp washing systems.
1-35
-------
Several commenters objected that the proposed emission
limits were not appropriate because they were based on limited
data that did not reflect the variability of emission from
sulfite pulping processes. The commenters provided the Agency
with emissions test data that illustrated fluctuations in the
methanol mass emissions over an extended time period due to
variations in products and process conditions.
The Agency evaluated the information provided by the
commenters and subsequently agreed with the commenters regarding
process variability at sulfite mills. For sodium- and calcium-
based sulfite pulping processes, the final emission limit is
0.89 Ib/ODTP. For ammonium- and magnesium-based sulfite pulping
processes, the final emission limit is 2.2 Ib/ODTP. Because the
emission limits were statistically derived to reflect process
variability, these emission limits and corresponding monitoring
parameters are never-to-be-exceeded values.
1.4.6 Soda and Semi-Chemical Mill Standards
The proposed standards required the owners or operators of
new or existing semi-chemical and soda mills to comply with the
same pulping standards as kraft mills. As a consequence of
subcategorizing the pulp and paper industry by pulping type,
different MACT control requirements were developed for soda and
semi-chemical mills. The final rule requires existing soda and
semi-chemical mills to control the digester and evaporator
systems (LVHC system). New soda and semi-chemical mills are
required to control the LVHC and the pulp washing systems.
1.4.7 Bleaching System Standards
In the December 17, 1993 proposed rule, all HAP emissions
from bleach plants were required to be reduced by 99 percent
1-36
-------
using a caustic scrubber. The Agency proposed to control
chlorinated HAP emissions only, using chlorine as a surrogate for
chlorinated HAP. As an alternative to the percent reduction
standard, the Agency proposed a 10 parts per million by
volume (ppmv) HAP caustic scrubber outlet concentration (measured
as chlorine). The Agency also proposed that chloroform emissions
be controlled by using 100 percent chlorine dioxide (C102)
substitution and eliminating hypochlorite use or by complying
with the requirements of the Effluent Limitation Guidelines and
Standards of the Clean Water Act (CWA). In addition, the Agency
proposed different control requirements for paper-grade and
dissolving-grade bleaching systems. The Agency also solicited
comments on providing a mass emission limit alternative to the
percent reduction and the outlet concentration standards.
The final rule continues to require chlorinated HAP
emissions (not including chloroform) to be reduced by 99 percent
(based on caustic scrubbing). As an alternative, bleach plants
can achieve an outlet concentration limit of 10 ppmv total
chlorinated HAP or a mass emission limit of 0.001 kg of total
chlorinated HAP (not including chloroform) per Mg ODP produced
(0.002 Ib/ODTP) (not including chloroform) for the following
bleaching systems: systems that use chlorine; systems at kraft,
sulfite, or soda pulping processes that use any chlorinated
compounds; and systems that use C102 to bleach pulp from
mechanical wood pulping processes or from any process using
secondary or non-wood fibers. A bleaching system that does not
use any chlorine or chlorinated compounds is exempt from
controls. The mill may use chlorine as a surrogate for
chlorinated HAP other than chloroform.
1-37
-------
All bleaching systems are also required to control
chloroform emissions by using 100 percent C1C>2 substitution and
eliminating hypochlorite use or by complying with the effluent
limitation guidelines and standards. For dissolving-grade
bleaching systems, the effective date of compliance with all the
bleaching standards has been stayed until the effluent limitation
guidelines and standards for dissolving-grade mills are
promulgated.
1.4.8 Compliance Schedule
In the December 17, 1993 proposed rule, the compliance
schedule for all pulping and bleaching processes was 3 years.
The final rule allows a total of 8 years to comply with the HVLC
vent standards at kraft pulp mills. Since the industry will be
implementing both water and air rules essentially at the same
time, the extended compliance schedule was adopted to allow the
necessary time to fully consider all pollution control options
including pollution prevention. Given the engineering
requirements, permitting requirements, and resources necessary to
implement the standards, the Agency decided that additional
compliance time for kraft HVLC sources is appropriate. The
3-year compliance schedule is retained for semi-chemical,
sulfite, and soda pulping processes the LVHC kraft pulping vent
standards, and bleaching systems at paper-grade mills. Standards
for the pulping process condensates apply to streams that are
typically not recycled or reused in the pulping process without
prior treatment. Therefore, the Agency did not consider it
necessary to extend the additional compliance time to pulping
wastewater streams. Dissolving-grade mills are required to
comply with the bleaching system standards no later than 3 years
1-3!
-------
after promulgation of the effluent limitation guidelines and
standards for dissolving-grade mills under 40 CFR 430, subpart D.
In addition, the final rule sets out a two-phased standard
for paper-grade bleach plants at a limited number of mills which
elect to control wastewater discharges to levels surpassing the
Advanced Technology Incentives Program in the effluent limitation
guidelines and standards portion of the final rule. The first
phase for existing source MACT requires no increase in the
existing HAP emission levels from the paper-grade bleaching
system (i.e., no backsliding) during the interim period when the
mill is working toward meeting their advanced technology (Best
available technology (under the Effluent Limitation Guidelines
and Standards of the Clean Water Act) (BAT) requirements. The
effective date of the first phase requirements is 60 days from
the date of publication in the Federal Register of the final
rule. The second phase requires compliance with revised MACT
based on baseline BAT requirements for all parameters, or
100 percent C1C>2 substitution and elimination of hypochlorite,
for bleached paper-grade kraft and soda mills. The compliance
date of the second phase of existing source MACT would be 6 years
after publication of the standards in the Federal Register.
The final rule also includes requirements for kraft mills to
submit a non-binding control strategy report along with the
initial notification. The purpose of the control strategy report
is to provide the Agency and the permitting authority with a
means for measuring a mill's progress towards compliance. The
control strategy report contains information such as a
description of the emission controls or process modifications
selected for compliance with the control requirements and
1-39
-------
compliance schedule. The information in the control strategy
report must be revised or updated every two years until the mill
is in compliance with the standards of § 63.443.
1.4.9 Test Methods
At proposal, the Agency required that Methods 308 and 26A be
used to test for compliance with the provisions of the rule.
Method 308 is used to measure methanol in vent streams.
Method 26A is used to measure chlorine in vent streams.
Method 305 is used to measure methanol in wastewater streams.
Since proposal, Method 308 has been validated using
Method 301 validation criteria. Method 308 has also been revised
to incorporate the technical comments received after proposal.
The Agency evaluated the commenter's claims regarding the
appropriateness of Method 26A and agrees that C1C>2 is a potential
interferant to the method. In the final rule the Agency decided
to incorporate modifications to Method 26A, based on the industry
chlorine test method.
In March of 1997, industry communicated to EPA that
Method 305 was not used by National Council of the Paper Industry
for Air and Stream Improvement (NCASI) to obtain the data used to
evaluate steam stripper system performance. Consequently,
industry asserted that Method 305 should not be specified in the
final rule for determining compliance with the pulping process
condensate standards. However, the method originally used by
NCASI has not been validated using the Method 301 procedures.
The Agency has considered the industry argument and has
decided to proceed with specifying Method 305 in the final rule.
However, EPA may amend the rule with a supplemental Federal
Register notice to allow this method to be used as either an
1-40
-------
alternative or a replacement for Method 305 pending satisfactory
completion of the Method 301 validation procedures.
1.4.10 Control Device Downtime
At proposal, emission limits were required to be met at all
times, except during startup, shutdown, or malfunction. No
allowance for control devices or collection system downtime was
specified in the rule.
The EPA re-evaluated the need to incorporate downtime or
excess emissions allowances for LVHC, HVLC, and steam stripper
systems. Based on the information collected in the 1992
voluntary MACT survey (A-92-40, IV-B-8) EPA has concluded that
100 percent compliance is not achievable at a well-designed and
operated system in this industry. The data indicate that some
allowance for excess emissions is part of the MACT floor level of
control. In the final rule, EPA established excess emissions
allowances to approximate the level of downtime and number of
backup control devices that exist at the best-performing mills.
The excess emissions allowances are designed to account for
periods when the control device is inoperable and when the
operating parameter values established during the initial
performance test cannot be maintained due to problems with the
process.
The excess emissions allowance for LVHC system control
devices is 1 percent of the operating hours on a quarterly basis.
For the HVLC system control devices or for control devices that
reduce both LVHC and HVLC system vent gases, the excess emissions
allowance is 4 percent. For LVHC and HVLC systems, the excess
emissions allowances do not include scheduled maintenance
activities malfunctions, startups, and shutdowns. Malfunctions,
1-41
-------
startups, and shutdowns must comply with the part 63 general
provisions.
The excess emissions allowance for steam stripper systems is
10 percent. This downtime allowance includes all periods when
the stripper systems are inoperable including scheduled
maintenance.
1.4.11 Equipment Enclosures, Closed-Vent Svstems, and Control
Equipment
1.4.11.1 Requirements for Closed-Vent Systems. At
proposal, the Agency required specific standards and monitoring
requirements for closed-vent systems. The standards required:
(1) maintaining a negative pressure at each opening, (2) ensuring
enclosure openings that were closed during the performance test
be closed during normal operation, (3) designing and operating
closed-vent systems to have no detectable leaks, (4) installing
flow indicators for bypass lines, and (5) securing bypass line
valves. Monitoring requirements included visually inspecting
seal/closure mechanisms and closed-vent systems and demonstrating
no detectable leaks in the closed-vent system.
The Agency evaluated comments on these provisions and made
several changes to the closed-vent system requirements. The
Agency agreed with the commenters that most closed-vent systems
will be under negative pressure. Any leaks, therefore, would
pull air into the collection system rather than release HAP's to
the atmosphere. Therefore, the Agency revised the requirement
for demonstration of negative pressure and no detectable
emissions to apply only to enclosures/hoods and portions of the
closed-vent system operated under positive pressure. The Agency
also agreed that requiring a lock and key type seal on bypass
1-42
-------
lines would be burdensome and could potentially pose a safety
hazard. The intention of the requirements was to prevent
circumvention of the control device by venting directly to the
atmosphere. The Agency believes that this assurance can be
achieved using car-seals or seals that could easily be broken, to
indicate when a valve has been turned. The Agency revised the
bypass line requirements to allow the use of car-seals but
require log entries recording valve position, flow rate, and
other parameters. The Agency has modified the enclosure
requirements to allow for short-term openings for pulp sampling
and maintenance.
The final rule retains the visual monitoring requirements.
These requirements can be conducted at a reasonable cost and are
necessary to ensure proper operation of collection systems.
1.4.11.2 Concentration Limit for Combustion Devices and
Design Incinerator Operating Parameters. At proposal, the rule
required vent streams to be controlled in a combustion device
that achieves 98 percent reduction of HAP's or a thermal oxidizer
that achieves an outlet HAP emission concentration of 20 ppmv
corrected to 3 percent oxygen. Alternatively, mills could comply
with the control requirements by routing vent streams to a design
incinerator operating at 1,600 °F with a residence time of
0.75 seconds or to a boiler, lime kiln, or recovery furnace. In
the final rule, EPA maintained the design incinerator operating
requirements.
The EPA re-evaluated the 3 percent correction factor in
order to ensure that it is appropriate for the pulp and paper
industry. Based on industry data and thermodynamic models, EPA
decided to revise the oxygen correction factor to 10 percent in
1-43
-------
the final rule. Therefore, the final rule allows thermal
oxidizers to be in compliance if they reduce HAP concentrations
to 20 ppmv corrected to 10 percent oxygen.
1.4.12 Interaction With The Resource Conservation and Recovery
Act (RCRA)
Following proposal, industry presented an approach for
recovering the energy contained in steam stripper condensates.
The condensates exhibit characteristics that would lead to its
classification as a hazardous waste under the Resource
Conservation and Recovery Act (RCRA).
After review of the characteristics of the condensate, the
Agency concluded that no additional control under RCRA is
warranted since combustion of these condensates will not increase
environmental risk, would reduce secondary impacts, and would
provide a cost savings. Therefore, the final rule contains a
direct final notice that amends RCRA to allow the on-site
combustion of condensates derived from steam stripping systems
used to comply with the pulping process condensate standards.
1.5 SUMMARY OF IMPACT OF FINAL STANDARDS
This section summarizes the emissions, energy, cost, and
economic impacts for the final NESHAP.
1.5.1 Emission Impacts
This NESHAP will reduce nationwide emissions of HAP from
pulp and paper mills by 139,000 Mg/yr (154,000 tpy), which
represents a 67 percent reduction by 2005 compared to the
emissions that would result in the absence of standards.
Emissions of volatile organic compounds (VOC) will be reduced by
409,000 Mg/yr (450,000 tpy), which represents a 49 percent
reduction by 2005 compared to emissions that would result in the
1-44
-------
absence of standards. Emissions of total reduced sulfur (TRS)
compounds will be reduced by 78,500 Mg/yr (86,500 tpy), which
represents a 54 percent reduction by 2005 compared to the
emissions that would result in the absence of the standards.
1.5.2 Energy Impacts
The national energy usage required to comply with the NESHAP
is expected to increase by 33 x lO-^ British thermal units
(Btu's) per year. The additional energy includes electricity
required to power fans and blowers to transport vent streams to
an emission control device, additional steam required for steam
stripping of pulping condensates, and auxiliary fuel required for
incineration of pulping area vent streams.
1.5.3 Secondary Environmental Impacts
Secondary environmental impacts of the NESHAP include
increased emissions of carbon monoxide (CO), nitrogen oxide
(NOX), sulfur dioxide (802), and particulate matter (PM) .
Secondary impacts are generated from combustion of fuel used to
power pollution control equipment and as a by-product of the
destruction of HAP's in combustion devices. Sulfur dioxide
emissions are expected to increase by approximately 94,500 Mg
annually. Sulfur dioxide emissions are generated primarily from
the combustion of sulfur-containing compounds (such as TRS) in
the vent streams at kraft mills. The CO emissions are expected
to increase by approximately 8,660 Mg annually. The NOx
emissions are expected to increase by approximately 5,230 Mg
annually. The PM emissions are expected to increase by
approximately 83 Mg annually.
1-45
-------
1.5.4 Cost Impacts
The implementation of this NESHAP is expected to result in
an annualized national cost of $130 million/year. This estimate
includes a cost of $123 million/year for air pollution control
devices and operational changes, and a monitoring, recordkeeping,
and reporting cost of $7 million/year. Table 1-4 presents the
national control cost impacts for the NESHAP at mills that pulp
wood using the kraft, semi-chemical, soda and sulfite processes.
No significant costs from mills that mechanically pulp wood, pulp
secondary fibers or non-wood are anticipated.
1.5.5 Economic Impacts and Benefits
Utilizing the estimated annualized cost of this NESHAP, an
evaluation of the economic impacts and distributional effects to
the pulp and paper industry is performed. The final rule when
evaluated independently of other regulatory requirements for air
and water pollution, is not expected to have a substantial impact
on the industry. Estimated price increases are less than
0.5 percent for bleached paper-grade kraft and sulfite,
dissolving-grade kraft and sulfite, and semi-chemical pulp and
paper products, while unbleached kraft pulp is estimated to have
a price increase of almost 5 percent. The costs imposed on
affected facilities do not result in any mill or firm closures,
thus, the rule assessed individually is not expected to alter
employment, shipments, or exports for the industry by appreciable
amounts.
Implementation of the final rule is expected to reduce
emissions of HAP's, VOC, and TRS, but increase emissions of PM,
S02, CO, and NOx. The benefits (damages) that accrue as a result
of the standard result from changes in human health effects
1-46
-------
TABLE 1-4. NATIONAL COST IMPACTS FOR NESHAPa
Cost category
Total capital Total annualized
investment costb
(million $) (million $/yr)
Control Equipment Costs
Kraft
Sulfite
Semi -chemical
Soda
Recordkeeping and
Reporting Costs
Total
452
23
11
<•)
8
496
117
5
1
0.2
1
130
almpacts are for controlling air emissions after the CWA
effluent guidelines are implemented.
bAmortized capital costs plus operation and maintenance costs,
1-47
-------
associated with inhalation of the above pollutants, as well as,
changes in welfare effects, such as: visibility and crop yields,
materials soiling and corrosion. The EPA is not able to place a
monetary value on all of the benefits achieved by the rule.
Values are obtained for changes in VOC, PM, and S02 emissions
only. Total benefits for these pollutants range in value from
($1,040) million to $1,054 million for the NESHAP, and ($727)
million to $1,493 million for the entire cluster rule.
1-4!
-------
2.0 INDUSTRY CHARACTERIZATION
2 . 1 ADEQUACY OF DATA FOR STANDARDS DEVELOPMENT
Comment: One commenter (20,059) contended that the data
base was unrepresentative of actual control levels because it did
not take into account best available control technology (BACT) or
lowest achievable emission rate (LAER) determinations. The
commenter (20,059) asserted that the Act requires EPA to include
all BACT determinations and all but the most recent LAER
determinations in determining the floor levels of control for
MACT standards. The commenter (20,059) asserted that Congress
intended for EPA to gather actual emissions data from a sample
likely to represent the top performers in order to determine the
floor level of control, and that a data gathering program must be
sufficient to ensure that EPA does not miss any sources that have
superior levels of emissions control. The commenter (20,059)
contended that EPA claimed to have reviewed BACT and LAER in the
BID (A-92-40, II-A-35) but had not provided any data or analyses
in the BID. Therefore, the commenter (20,059) concluded that EPA
had not collected or evaluated the data needed to identify the
average emission limitation achieved by the best-performing
12 percent of sources as required by the statute.
Response: BACT is the level of control required in
attainment areas undergoing Prevention of Significant
2-1
-------
Deterioration (PSD) review. LAER is the control level required
in nonattainment areas undergoing New Source Review. Both are
determined on a case-by-case basis. In establishing the MACT
level of control, EPA evaluated existing controls at all mills,
including the BACT/LAER controls in place as a result of PSD/NSR
review (although in some circumstances, EPA may not consider LAER
level of control to be MACT; see Act section 112 (d) (3) (A) ) . The
BACT and LAER determinations are accounted for in the control
devices that were reported in the MACT survey responses. These
controls were used to calculate the baseline emissions and
baseline level of control.
2.2 EMISSION FACTORS
2.2.1 Data and Approach Used
Comment: Several commenters on the proposed rule (20,011,
20,043, 20,054, 20,056, 20,071, 20,102, 20,103, 20,115) expressed
concerns over the use of general models and liquid-based emission
factors for a mill-wide characterization because: (1) it was
unreasonable and insupportable to base decisions in the
rulemaking on emission factors for vent streams developed from
models and from liquid stream concentrations, and (2) actual
measured data should have been used to develop the emission
factors.
Two commenters (20,011, 20,027) asserted that EPA models did
not have the capability to accurately predict emissions. One of
the commenters (20,027) asserted that: (1) the emission factors
used in the model process units were based on several assumptions
for which they could find no scientific or technical basis (the
commenter provided several examples of erroneous assumptions),
(2) the models for estimating air emissions from HAP
2-2
-------
concentrations tested in process wastewater incorrectly assumed
that equilibrium was reached between the water and air
components, and (3) they were not aware of data pertaining to
liquid-phase HAP concentrations entering or exiting bleach plant
equipment that would be sufficient for developing reliable air
emission rates for standards development.
Two commenters (20,102, 20,129) recommended that EPA develop
more specific emission factors for various emissions from the
pulping area. One of the commenters (20,129) indicated that EPA
should summarize the air emission data from NCASI Technical
Bulletin No. 650 into emission factors dealing with the MACT
regulation and for Compilation of Air Pollutant Emission Factors,
5th edition, Volume 1: Stationary Point and Area Sources (AP-42)
emission inventory purposes. After reviewing the industry data,
one commenter (20,054A2) pointed out that there appear to be
differences in various process emissions, which if properly
understood, may provide options for less expensive controls.
One commenter (20,071) concluded that the approach that EPA
used to characterize the HAP emissions of over 160 diverse
chemical pulping mills (including bleached kraft, unbleached
kraft, sulfite, and semi-chemical processes) was inadequate for a
regulation with such significant financial impact on the
industry. The commenter (20,071) stated that EPA should not have
relied on limited data and the extrapolation of these data
through mathematical models to develop emission factors.
Response: At proposal, EPA developed emission factors for
each type of individual emission point typically found at pulp
and paper mills. The emission factors were developed from
measured air emissions at process vents and from air emissions
2-3
-------
estimates extricated from liquid stream data, assuming
equilibrium conditions. This information was the best data
available to EPA.
Based on test data received after proposal, EPA changed the
approach from individual emission point factors to emission
factors based on mill systems. Availability of these data was
announced in the Federal Register on February 22, 1995 and
proposed changes to emission factors were announced in the
March 8, 1996 Federal Register supplemental notice.
The EPA concluded that the system approach is the best
approach because it provides a more objective comparison of mills
and lessens the problems associated with the nomenclature
assigned to individual process components. The EPA believes that
the revised system emission factors provide the best data to
characterize emissions from the pulp and paper industry.
The EPA concluded that the liquid-based model used at
proposal provided an adequate estimate of emissions when compared
to the actual test data. However, the system emission factors
used in the final rule were not based on the liquid-based
equilibrium models, but on actual data received from industry
after proposal. Since most of the standards in the final rule
remain at the MACT floor level, the significance of emission
factors is somewhat reduced.
Comment: One commenter (20,122) expressed concern about the
chloroform releases that occur even with C1C>2 substitution. To
support their concern, the commenter (20,122) reported an
estimated chloroform emission factor of 0.22 tons per 1,000 tons
of pulp for market bleached kraft mills using C1C>2 substitution.
2-4
-------
Response: Emission information submitted to EPA from
several facilities provides an emission factor of 0.39 Ib
chloroform per oven-dried ton of pulp (ODTP) for bleaching
systems operating with 100 percent C1C>2 substitution and a
hypochlorite bleaching stage. This emission factor reduces to
0.012 Ib of chloroform per ODTP for bleaching systems operating
with 100 percent C1C>2 substitution and no hypochlorite bleaching
stage. The emission factor submitted by the commenter (20,122)
for chloroform of 0.22 tons of chloroform per 1,000 tons of pulp
converts to 0.44 Ib of chloroform per ton of pulp, which is
comparable to the emission factor of 0.39 Ib chloroform per ODTP
for bleaching systems operating with 100 percent C102
substitution and a hypochlorite bleaching stage.
The MACT floor level of control at bleaching systems is
caustic scrubbing and process modifications (100 percent
substitution and no hypochlorite use). The effluent limitation
guidelines and standards requirements for paper-grade bleaching
are also 100 percent substitution of C102 and no hypochlorite
(EPA is evaluating requirements for dissolving-grade bleach
mills). The EPA considers the effluent limitation guidelines and
standards requirements to be at least as stringent as the floor-
level process modifications. Therefore, the final rule requires
compliance with the effluent limitation guidelines and standards
requirements to control chloroform in the bleaching system or
certification that no hypochlorite or chlorine is used for
bleaching. This requirement will significantly reduce chloroform
emissions from bleaching systems because chloroform emissions are
related to using hypochlorite as a bleaching agent.
2-5
-------
2.2.2 Emission Factor Document
Comment: After evaluating the draft Chemical Pulping
Emission Factor Development Document (A-92-40, IV-A-6), one
commenter (IV-D1-102) provided suggestions and comment on the
development of emission factors. In particular, the commenter
disagreed with the HAP ratio procedure used. The commenter
(IV-D1-102) perceived two flaws with the approach: (1) EPA
ignored results where a compound was tested but the results were
below the detection limit, and (2) EPA assumed, in the absence of
data, that the ratios between specific HAP compounds and methanol
were constant for a given type of source (e.g., brownstock
washers or weak black liquor tanks).
Response: Generally, the scope of the emissions tests were
limited to a select group of compounds. In the draft emission
factor document, the method used to estimate emission factors was
based on the assumption that the ratio of a compound's
concentration in a vent to the concentration of methanol in the
same vent is similar to the ratio in vents of similar systems.
Based on industry comments, the data were re-evaluated and a
system-unit approach to estimating emission factors was adopted
in place of the previous HAP-ratio approach.
The system-unit approach consists of sorting the data for
each HAP species into the same mill-system groupings used to
develop the methanol emission factor, as described in the revised
Chemical Pulping Emission Factor Development Document (A-92-40,
IV-A-8) . Where sufficient data to characterize a HAP compound by
mill system were not available, the unit approach was used for
that compound. In the unit approach, equipment-specific emission
factors were developed. Then, mill system equivalent emission
2-6
-------
factors were generated for these compounds by assuming that mill
systems are typically made up of certain equipment configurations
(e.g., the typical pulp washing system consists of three washer
hood vents and one filtrate tank vent).
The system-unit approach accounted for results below the
detection limit. For compounds for which a detection limit was
reported, one half of the detection limit was used. If no
detection limit was recorded in the test report, that test was
not used. The EPA believes that the system-unit approach to
analyzing the industry test data provides an accurate
characterization of emissions by incorporating results below the
detection limit and avoiding assumptions of constant ratios
across different sources.
Comment: One commenter (IV-D1-102) disagreed with EPA's
contention in the draft emission factor document (A-92-40,
IV-A-6) that the summary of results presented in the NCASI
Technical Bulletin No. 701 (IV-J-31) fall within the ranges
developed in the draft emission factor document and will not
significantly alter the results. The commenter (IV-D1-102)
agreed that the methanol results would not be significantly
different but argued that the summary of HAP emissions would need
to be revised upon incorporating the Technical Bulletin
No. 701 data.
Response: The data presented in NCASI Technical
Bulletin No. 701 are a summary of the same data that EPA used to
revise the emission factors; they are not new or separate test
data. The EPA agrees that the methanol emissions would not be
significantly different by including of the NCASI Technical
Bulletin No. 701 data. The EPA also agrees that the approach to
2-7
-------
determining total HAP used in NCASI Technical Bulletin No. 701
would result in significantly lower total HAP emissions for the
mill systems. The approach used to develop speciated HAP
emission factors for the final rule has been revised and is more
consistent with the results in the Technical Bulletin No. 701
summary. (See previous discussion regarding the mill system-unit
approach.)
Comment: One commenter (IV-D1-102) noted two problems with
the appropriateness of the model plant approach: the lack of
neutralization units in some models, and the lack of a diffused
aeration model plant. The commenter (IV-D1-102) also questioned
the number of mills assigned to the model plants.
Response: Because neutralization occurs in units besides
strict "neutralization basins," EPA does not believe that the
absence of explicit neutralization units in some models is
inconsistent with the models having neutralization units. Also,
the available data do not support creating a diffused aeration
model or making changes to the mill assignments. Since no new
data were provided, EPA maintains that the model plants and mill
assignments used are an accurate representation of the industry.
Comment: One commenter (IV-D1-102) disagreed with three
aspects of the WATERS modules: the model plant B settling pond,
the assumption that in the model plant C that neutralization
occurs in stabilization basins following the clarifier, and the
model plant D non-aerated basin. The commenter (IV-D1-102)
expressed concern over the length of settling basin residence
time in model plant B. The commenter (IV-D1-102) also objected
to having neutralization follow the model plant C clarifier and
to having the model plant D polishing basin as non-aerated.
-------
Response: The residence time for the model plant B settling
basin is based on a settling pond at one kraft paper mill. The
WATERS outputs for that mill (based on the assumed residence
time) agree with the test data.
In model plant C, no neutralization was assumed to occur in
the basins following the clarifier. Neutralization was assumed
to occur between the bar screen and the clarifier.
The non-aerated basin in model plant D was determined to be
improperly labeled as aerated. The current basin is a composite
of the ten model plant D mills with non-aerated systems and the
11 mills with aerated systems. The EPA does not believe that
revising the WATERS inputs would yield a significant change, but
the labels and documentation of the approach have been updated.
Comment: Eight specific WATERS input parameters (e.g.,
concentrations, temperatures) were rated by one commenter
(IV-D1-102) as inconsistent, inaccurate, or unreasonable.
Response: The EPA evaluated the commenter's concerns and
data characterizing the industry. The results of the evaluation
show that the parameters used in the WATERS model accurately
reflect the industry based on comparison with industry data.
Based on EPA's analysis of the commenter's concerns and
suggestions, EPA maintains that only minor changes would result
from altering the input parameters as suggested by the commenter
(A-92-40, IV-B-101) .
Comment: One commenter (IV-D1-102) asserted that the WATERS
primary clarifier module overestimates emissions.
Response: Validation of the WATERS primary clarifier module
is documented in a memorandum included in the revised emission
2-9
-------
factor document (A-92-40, IV-A-8). The EPA maintains that the
WATERS primary clarifier module adequately estimates emissions.
Comment: One commenter (IV-D1-102) supplied WATERS outputs
using modified input parameters, and suggested updating the
emission factor document to reflect emission estimates based on
the modified input parameters.
Response: The EPA reviewed the industry-derived emission
factors for methanol and chloroform. Although the revised
methanol results were lower, the revised chloroform results were
higher than those obtained by the Agency's model. The EPA
believes that the current model adequately characterizes
emissions from wastewater treatment and did not incorporate the
industry-derived emission factors since the factors had little
effect on overall HAP emission estimates.
Comment: One commenter (IV-D1-102) noted the following
specific concerns about using kraft mill system emissions as
defaults for non-kraft mill systems where data were not
available:
Oxygen delignification systems at sulfite mills are
configured differently from kraft oxygen
delignification systems and should have a different
emission factor based on typical equipment.
Stand-alone semi-chemical mills should have refiners
rather than knotters and screens, and the causticizing
area at a stand-alone semi-chemical mill would not have
a lime kiln, lime mud washer, or slaker.
Kraft digester and evaporator numbers should not be
used for other types of pulping because of the
different liquor characteristics and cooking
conditions.
Semi-chemical pulping wastewater has considerably lower
methanol concentrations than kraft pulp mill
2-10
-------
wastewater, so it does not seem appropriate to use
average kraft mill values.
Response: In response to these specific concerns, EPA made
the following revisions to the emission factor document:
No data were supplied in relation to oxygen
delignification systems at sulfite mills; therefore no
revisions were made.
The knotter and screening systems at stand-alone semi-
chemical mills were correctly identified by the
"refiner" terminology; however, no data were available
to suggest that the emissions from the pre-washing
screening area are different at semi-chemical mills.
The equipment that is not present at a stand-alone
semi-chemical mill causticizing area were removed.
Kraft digester and evaporator numbers were not used for
semichemical and sulfite mills. The HAP emissions at
soda mills are expected to be similar to the non-TRS
HAP emissions from kraft mills. Therefore, the soda
numbers were based on kraft emissions.
Semi-chemical pulping wastewater emission
characteristics were developed separately from the
kraft characteristics.
The revised emission factor document (A-92-40, IV-A-8)
contains more detail regarding emission factor development,
assumptions, and applications.
Comment: One commenter (IV-D1-102) asserted that the
boiler methanol emission factor in the draft emission factor
document (A-92-40, IV-A-6) is too high and that the Agency has
not adequately documented why 0.5 Ib methanol/ODTP from a boiler,
especially one without a wet scrubber, is typical of the
industry.
2-11
-------
Response: Emissions from boilers are not addressed under
this rule and the boiler methanol emission factor does not impact
the final rulemaking. Since it was not relevant to this
standard, EPA did not revise the emission factor for the NESHAP.
The boiler emission factor is discussed and evaluated in further
detail in the revised emission factor document (A-92-40, IV-A-8).
Comment: One commenter (IV-D1-102) stated that the reason
for the difference in chloroform generation in the bleach plants
is largely a function of chlorine use. The commenter (IV-D1-102)
stated that the difference should not be attributed to the
presence or absence of oxygen delignification.
Response: The EPA agrees with the commenter that other
parameters besides the presence of oxygen delignification have
greater impact on chloroform emissions from the bleach plants.
The presence of a hypochlorite stage in the bleach sequence and
the degree of C1C>2 substitution have both been determined to
significantly affect bleach plant chloroform emissions. Greater
detail and data analysis pertaining to this issue are presented
in the revised emission factor document (A-92-40, IV-A-8).
2.3 MODEL PROCESS UNITS
Comment: Two commenters (20,027, 20,086) stated that EPA
should not have used model process units to evaluate the range of
possible control options and the ability to achieve the proposed
MACT standards; they should have evaluated these things on a
"real world" mill-by-mill basis. Another commenter (20,011)
argued that invalid process models led to mischaracterized pulp
mill and wastewater emissions.
One commenter (20,027) cautioned that the models used for
semi-chemical mills and sulfite mills were incorrect. The
2-12
-------
commenter (20,027) contended that for semi-chemical mills, EPA
incorrectly assumed that there were digester relief gases and
digester blow evaporators. The commenter (20,027) recommended
that mill Q from the industry test program be used as the basis
for the semi-chemical model mill rather than EPA's models P9
and P10. The commenter (20,027) also claimed that the sulfite
pulping model mill developed by EPA did not accurately reflect
the process emissions points because several of the emission
points in the sulfite model (P7) were inappropriately taken from
the kraft model (P2). The commenter (20,027) did not provide
alternative points.
Another commenter (20,072) indicated that their model
developed for a soda mill was more effective at estimating the
effects of process changes at soda mills than a model which uses
kraft TRS control technology. The commenter (20,072) provided a
report on this mathematical model.
Response: Based on comments and data received after
proposal, EPA has re-evaluated the methodology used to estimate
national impacts for the pulp and paper industry. The impacts
estimated in the final rule were determined for each mill using
mill-specific data provided by industry after proposal.
Therefore, the MACT floor analyses and impacts analyses were
based on actual processes and controls at each of the mills and
were not based on models.
Where information was missing, average characteristics of
mills with similar pulping types were used to complete the data
base. Through this revised analysis, EPA has more accurately
estimated emissions from non-kraft mills. (Kraft models are not
assigned to semi-chemical, sulfite, or soda mills because actual
2-13
-------
mill-specific data was used in place of model mills.) The
revisions to the national impacts analyses are discussed in
detail in chapter 20 of this document.
2-14
-------
3.0 SUBCATEGORIZATION
Comment: Several commenters to the December 17, 1993
proposed rule (20,001, 20,011, 20,018, 20,027, 20,054A2,
20,072A8, 20,086) requested that EPA subcategorize mills by pulp
type because different pulp types have different emission
characteristics, baseline controls, and retrofit costs. One
commenter (IV-D2-15) on the March 8, 1996 supplemental notice
supported the decision to subcategorize by pulping type.
Two commenters (20,027, 20,072A8) supported their argument
to subcategorize by noting that section 112 (c) (1) of the Act
requires MACT standards to be consistent with the list of source
categories established in section 111. The commenters (20,027,
20,072A8) contended that section 111 New Source Performance
Standards (NSPS) source categories only include kraft pulping
mills, and do not include semi-chemical, soda, or sulfite pulping
mills. Therefore, the commenters concluded that EPA should have
treated kraft mills separately from other mill types. The
commenters also suggested that sources other than kraft mills not
be included in the regulation.
Two commenters (20,059, 20,103) recommended no
subcategorization, agreeing with the consolidation of
subcategories proposed by EPA and suggesting no further division
or combination of subcategories. One commenter (20,059) on the
3-1
-------
December 17, 1993 proposed rule indicated that the industry has
not provided data to support subcategorization. Another
commenter (20,011) contended that the lack of air and wastewater
emissions data for sulfite, soda, and semi-chemical mills
prevented a balanced assessment of the need for subcategories.
Response: Section 112 of the Act requires NESHAP for
categories of major sources of HAP. On July 16, 1992, EPA
published a list of source categories for the 189 listed HAP' s
(57 FR 31576). Pulp and paper production was listed as a major
source of HAP emissions. Soda, semi-chemical, and sulfite mills
are major HAP sources and, therefore, are being regulated as a
part of this source category. The final standards are based on
evaluation of all available data for potential controls and the
best opportunity for integration with effluent guidelines.
In the March 8, 1996 supplemental notice, EPA presented the
rationale for establishing separate subcategories based on
pulping type (kraft, soda, sulfite, or semi-chemical). The
establishment of the subcategories was based on comments received
and review of additional emissions information submitted after
proposal of the standards.
The information obtained after proposal indicated that as a
result of the differences in digestion methods, mills utilizing
different types of pulping systems produce different emissions,
and as a result, achieve different degrees of control with
different applicable control technologies. At proposal, EPA
understood that the four types of mills differ in the way they
digest wood to make pulp, but did not have the data to determine
the extent to which these differences influence potential
emission control strategies. The information received after
3-2
-------
proposal indicated the significant extent of these differences.
The commenters are referred to the March 8, 1996 supplemental
notice for a detailed discussion of the differences between the
pulping types and rationale for the decision to establish each of
the subcategories.
Where two or more subcategories are located at the same mill
site and share a piece of equipment, that piece of equipment
would be considered a part of the subcategory with the more
stringent MACT requirements for that piece of equipment. For
example, the pulping process condensates from an evaporation set
processing both kraft weak black liquor and spent liquor from a
semi-chemical process would have to comply with the kraft
subcategory requirements for pulping process condensates. This
more stringent requirement is appropriate because there is no
viable way to isolate the emissions for each pulping source to
determine compliance separately.
Comment: One commenter (20,043) indicated that separate
subcategories should be established for dissolving-grade and non-
dissolving (paper-grade) sulfite mills based on significant
differences that exist between dissolving- and paper-grade
sulfite mills. The commenter (20,043) urged EPA to accurately
characterize the emissions, control technology, and the costs of
controlling emissions at dissolving-grade sulfite mills.
Response: The EPA believes the commenter's point is valid
for the bleaching systems at all mills, not just sulfite mills.
In characterizing the bleaching system, there are greater
differences between the paper-grade bleaching process and
dissolving-grade bleaching process than between the type of pulp
mill that proceeds the bleaching systems. The EPA evaluated the
3-3
-------
differences between paper-grade and dissolving-grade bleaching
systems and determined the appropriate MACT requirements for
each.
The average emission limitation of the best-controlled
paper-grade and dissolving-grade mills is control of chlorinated
HAP' s using a caustic scrubber and control of chloroform using
process modifications. For paper-grade mills, the effluent
limitation guidelines and standards were determined to be at
least as stringent as the process modifications evaluated
(100 percent C1C>2 substitution and no hypochlorite use) .
Therefore, the MACT requirements for paper-grade bleaching
systems is caustic scrubbing for control of chlorinated HAP,
other than chloroform, and compliance with the effluent
limitation guidelines and standards or certification that no
hypochlorite or chlorine is used for bleaching for control of
chloroform.
The EPA at present lacks sufficient information to establish
effluent limitation guidelines and standards at dissolving-grade
mills, and also lacks information to reliably ascertain what a
MACT floor standard for chloroform air emissions would be for
this unit operation. The EPA is continuing to evaluate potential
limitations for dissolving-grade mills and is deferring
establishing MACT standards for chloroform until effluent
limitation guidelines and standards are established. Therefore,
dissolving-grade mills are required to comply with the bleaching
system chloroform standards no later than 3 years after
publication of the wastewater effluent limitation guidelines and
standards under 40 CFR 430, subparts A and D.
3-4
-------
In a related action, EPA is also delaying MACT requirements
for chlorinated HAP's other than chloroform from dissolving-grade
bleaching operations until 3 years after publication of the
wastewater effluent limitation guidelines and standards under
40 CFR 430, subparts A and D. The Agency is doing so in order to
avoid imposition of Act requirements that would be inconsistent
with, or superseded by, forthcoming CWA regulations. A more
detailed discussion of bleaching system compliance times is
presented in section 17.2.
3-5
-------
-------
4.0 BASIS OF STANDARDS
4.1 STATUTORY INTERPRETATION
4.1.1 88th percentile vs. 94th percentile Interpretation
Comment: Several commenters (20,027, 20,037, 20,046,
20,046A2, 20,056, 20,070A1, 20,083, 20,089, 20,092,
20,102, 20,103, IV-D2-15, IV-D2-7) objected to EPA's
interpretation that the 94th percentile represents "the average
emission limitation of the best-performing 12 percent of existing
sources." These commenters contended that: (1) the MACT floor
level of control should be set at the 88th percentile rather than
the 94th percentile, and (2) the 94th percentile interpretation
was impractical; irrational; not allowed by the Act;
significantly more costly than the 88th percentile interpretation
[Case law cited: Chevron. U.S.A. v. NRDC. 467 U.S. 837 (1984)];
and likely more stringent than that achieved by any existing
source. Several commenters (20,027, 20,054A2, 20,057A2,
20,070A1) advised that using the 88th percentile interpretation
would have significant consequences regarding which pieces of
equipment must be controlled, asserting that brownstock washers,
oxygen delignification units, and weak black liquor storage tanks
are not enclosed or controlled at the 88th percentile.
One commenter (20,054A2) indicated that the method for
determining the floor level of control was applied inconsistently
4-1
-------
between components that were controlled at greater than
12 percent and those controlled at less than 12 percent. The
commenter (20,054A2) stated that if the component was not
controlled by at least 12 percent of mills, EPA determined what
the top 6 percent of mills were doing.
One commenter (20,122) supported EPA's interpretation that
the MACT floor level of control effectively equaled the
94th percentile control technology, and one commenter (20,059)
indicated that the interpretation was immoral because it was set
below the minimum legal stringency for protecting human health.
Several industry commenters (20,057A2, 20,059, 20,092,
20,102, 20,149) agreed with the interpretation of averaging the
top 12 percent, but they did not agree with using the median to
represent the average performance of these sources. Some
commenters (20,057A2, 20,059) indicated that an arithmetic mean
should be used regardless of corresponding control technologies,
while others (20,092, 20,102) indicated that the floor should be
set at the next most stringent corresponding technology.
One commenter (20,103) expressed support for EPA's use of
the median of the top 12 percent, stating that the arithmetic
mean may place too much emphasis on either the best or worst
performing of the top 12 percent.
Response: In the June 6, 1994 Federal Register
(59 FR 29196) EPA presented its final decision regarding the
interpretation of section 112(d) (3) (A) of the Act for purposes of
the Hazardous Organic NESHAP (HON) . As presented in 59 FR 29196,
EPA concluded that section 112 (d) (3) (A) is best interpreted to
require EPA to first determine the emission limitation achieved
by sources within the best-performing 12 percent, and then
4-2
-------
average these limitations. This interpretation of the statute
has been referred to as the "Higher Floor Interpretation." The
Agency adheres to that interpretation in this rule. The Agency
notes, however, that while the interpretation presented in
59 FR 29196 sets a precedent, it is not binding since EPA
believes the Agency retains discretion in establishing floors for
MACT standards depending on the circumstances of each source
category.
The EPA has the discretion to use its best engineering
judgment in collecting and analyzing the data, and in assessing
the data's comprehensiveness, accuracy, and variability in order
to determine which sources achieve the best emission reductions.
The EPA fully considered all comments regarding the proper
interpretation of section 112 (d) (3) (A) of the Act in the context
of the pulp and paper rulemaking. For this rulemaking, EPA held
to the "Higher Floor Interpretation" (average of the
best-performing 12 percent).
Commenters on the December 17, 1993 proposal provided
additional emissions and control information to be evaluated in
determining the floor levels of control and characterizing the
industry. After review of the data, EPA revised several aspects
of the proposal. These changes are discussed in the
March 8, 1996 supplemental notice. Specific changes to the rule
and associated rationale are presented in the notice. In
general, EPA determined that it was appropriate to subcategorize
the pulp and paper industry based on pulping type (e.g., kraft,
soda, semi-chemical, sulfite) . Revisions were made to
recalculate the floor level of control for regulated emission
points within each subcategory. Additionally, emission points
4-3
-------
were grouped together to form mill systems in order to better
characterize emissions from this industry. In determining the
best-performing sources from which to compute the floor level of
control, EPA calculated the emission controls for emission points
within each system. In most cases, EPA relied on the arithmetic
average of the best-performing sources. Whenever the resulting
value did not correspond to an emission limitation that was
achievable by any particular technology, the median of the best-
performing sources was used in order to develop a standard in
fact reflecting achievable performance (see section 112(d)(2)).
The EPA believes that the changes to the rule presented in
the March 8, 1996 supplemental notice result in floor
determinations that are appropriate and reasonable for all mills
within each subcategory.
4.1.2 MACT Floor on a "Per Unit" vs. "Whole Mill" Basis
Comment: Several commenters (20,027, 20,045, 20,051,
20,057A2, 20,066A3, 20,114, 20,118, 20,145, IV-D2-7) disagreed
with EPA's use of the "best-performing individual emission units"
to determine the MACT floor level of control, rather than
considering the integrated mill performance. The commenters
(20,027, 20,045, 20,051, 20,057A2, 20,066A3, 20,114, 20,118,
IV-D2-7) suggested that EPA overstated the MACT floor level of
control, because it did not consider the interrelationship of the
different processes used to produce bleached and unbleached kraft
pulp, and the commenters indicated that this interpretation led
to a floor level of control that exists at less than the top
six percent of mills. The commenters (20,027, 20,045, 20,057A2,
20,066A3, 20,118, 20,145, IV-D2-7) asserted that this approach
4-4
-------
proved that EPA did not use a correct interpretation of
section 112 of the Act.
Response: The Act does not define "source." A source may
be a facility, a kind of emission point, or a collection of
emission points. The definition chosen for each MACT standard is
dependent on the characteristics of the source category being
regulated, and the information available to characterize
emissions. The EPA has chosen to define a source in the pulp and
paper rule as a collection of emission points (i.e., pulping
system, bleaching system, pulping process condensates). The
floor level of control was then determined for each emission
point. This method is referred to as the "per unit" approach.
The approach the Agency used to determine the floor level of
control based on emission points was the most appropriate because
this approach represents the best use of the data available. The
data available at proposal consisted of responses from a 1992
voluntary MACT survey, a field test program of air and liquid
samples from four kraft mills and one sulfite mill (the "EPA
5-mi11 study") and some limited industry data used to supplement
the EPA 5-mi 11 study. Based on comments and data received after
proposal (A-92-40, IV-D1-29, IV-Dl-29a, IV-D1-31, IV-D1-33,
IV-D1-34, IV-D1-35, IV-D1-38, IV-D1-39, and IV-D1-41), EPA re-
evaluated the approach and established the MACT floor by
subcategories (i.e., kraft, soda, semi-chemical, sulfite). The
EPA, however, retained the "per unit" approach to setting the
MACT floor by adopting MACT standards for specifically defined
equipment systems (pulping, bleaching) and associated wastewater
streams within each subcategory.
4-5
-------
The EPA elected to determine MACT floors on a per unit basis
because sufficient information was not available to determine the
MACT floor on a mill basis. Due to the differences in control
technologies used in processes and systems in a mill, the MACT
floor for a whole mill could not be based on a type of control
technology, but would need to be based on emissions or percent
reductions from the mills. Computing emission levels or percent
reductions of the whole mill would require accurate site-specific
knowledge of the emission levels of each process at each mill
being regulated (i.e., emission source tests). At proposal EPA
did not have sufficient data to establish a mass emission limit
or a mass emission reduction percentage across each mill.
Since proposal, EPA obtained site-specific information that
was used to develop emission factors for various systems at a
mill. However, these emission factors represent average or
typical systems and are not specific to each mill. While EPA
believes such information may be used to estimate national
impacts, it is not adequate to determine the MACT floor level of
control (i.e., the factors are not representative of the actual
emissions at each mill but may be used to represent typical
emissions from all mills). Actual mass emission levels or mass
emission reductions would still be required. Information on the
controls for various systems at each mill was available to EPA.
Therefore, EPA decided to develop MACT floors on a unit
(i.e., system) basis.
Additionally, the day-to-day variability of the pulp and
paper processes would preclude establishing mill-wide emission
or percent reduction limits. These process variabilities include
swings in production depending on the wood species available and
4-6
-------
products being produced, as well as other variables associated
with using a natural feedstock such as wood.
4.1.3 Legal Requirement to Base MACT Floor on Actual Data
Comment: Several commenters (20,027, 20,061, 20,146)
maintained that EPA is legally required to base the MACT floor on
what is actually achieved by sources or sources technically
similar. The levels achieved must be determined based on
reliable data and analyses rather than on predictions or
projections. One commenter (20,027) asserted that EPA must
redetermine the MACT floor based on actual data.
Response: Section 112 (d) (3) (A) of the Act requires that the
maximum degree of reduction in emissions be calculated from "the
average emission limitation achieved by the best-performing
12 percent of the existing sources (for which the Administrator
has emissions information) . . ." (emphasis added) . The EPA
agrees with the commenters that the MACT standards should be
based on the best data available to the Administrator and EPA
contends that the data available at proposal was used properly.
The EPA made all reasonable efforts to gather available data
using literature, State regulations, previous studies, sampling
tests, and a voluntary industry survey. Additionally, EPA worked
with the pulp and paper industry to gather data and used data the
industry submitted. Where information was lacking, average
values from the data base were used to fill in gaps.
Also, in the proposal of December 17, 1993, EPA acknowledged
that industry had air emissions sampling and data collection
underway. However, the data results were not expected to be
available until after proposal. Therefore, the proposal stated
that EPA would analyze and any data that became available before
4-7
-------
promulgation of the NESHAP. After proposal, commenters and
industry representatives submitted additional data including
results of sampling tests to EPA (A-92-40, IV-D1-29, IV-Dl-29a,
IV-D1-31, IV-D1-33, IV-D1-34, IV-D1-35, IV-D1-38, IV-D1-39, and
IV-D1-41). This information was evaluated and, where
appropriate, changes were made to the proposed rule accordingly.
A detailed discussion of EPA's evaluation and these proposed
changes as well as EPA's plans to address some other concerns
raised by the commenters are presented in the March 8, 1996
supplemental notice.
Comment: One commenter (20,011) indicated that the use of a
model pulp mill improperly extended the MACT floor beyond the
statutory definitions.
Response: Model pulp mills were not used by EPA to
determine the MACT floor. Rather, the MACT floor was based on
data collected in the 1992 voluntary MACT survey of the industry.
At proposal, model mills were used to estimate emissions and
other regulatory impacts corresponding to the specific control
options considered by the Agency. For the final rule, data from
each mill were used to estimate emissions and regulatory impacts.
4.1.4 Legal Requirement to Re-propose
Comment: One commenter (20,027) contended that EPA is
legally required to re-propose the standards after actual data is
obtained in order to give the public the opportunity to comment
on the new data and EPA's method for making the MACT
determination. [Case law cited: National Lime Assoc. v. EPA,
627 F.2d 418, 433, 452-53 (B.C. Cir. 1980); Weyerhauser Co. v.
Costle, 590 F.2d 1011, 1030 (B.C. Cir. 1978); Portland Cement
Assoc. v. Ruckelhaus, 486 F.2d 375, 392-93 (B.C. Cir. 1973);
-------
Sierra Club v. Costle, 657 F.2d 298, 334 (B.C. Cir. 1981); Sol i te
Core v. EPAr 952 F.2d 473, 484 (B.C. Cir. 1991) quoting
Connecticut Light & Power v. NRCf 673 F.2d 525, 530-31
(D.C. Cir. 1982); Fertilizer Institute v. EPA, 935 F.2d 1303,
1311 (B.C. Cir. 1991) .]
One commenter (20,057A2) stated that the MACT standards
should be re-proposed with EPA's proposed combustion source MACT
standards. (A-92-40, II-I-13 and 11-1-18. Bata provided:
Appendix MACT 6.)
Response: The EPA's position is that re-proposal is not
required because notices of data availability for data received
after the original proposal, EPA's assessment of the data, and
proposed changes to the original proposal were published in the
Federal Register February 22, 1995 (60 FR 9813) and on
March 8, 1996. The March 8, 1996 supplemental notice provided
the public the opportunity to comment on the new information and
on the approach under consideration by EPA in developing the
final standards.
These subsequent notices provided ample notice and
opportunity to comment on all key elements of the standard,
including data, potential floor levels of control, and potential
standards. In addition, EPA notes that it has provided actual
notice and opportunity to comment to many key parties to the
proceeding, including the pulp and paper industry and key
environmental groups (A-92-40, section E) . This ongoing dialogue
again fully satisfies notice and comment obligations as to all
persons having actual notice. The EPA cites Small Lead Refiners
Phase Bown Task Force v. EPA. 705 F.2d 506, 549 (B.C. M 83).
4-9
-------
The combustion MACT is a separate action and was proposed at
promulgation of the pulp and paper mill rule. The EPA
acknowledges that there are interrelations between this rule and
the combustion MACT rulemaking. The EPA evaluated those
interrelationships for the final rule (see chapter 16) . The EPA
maintains it is unnecessary to incorporate the combustion sources
in this pulp and paper rule.
4.1.5 Control Devices Were Not Installed to Reduce HAP
Comment: Several commenters (20,011, 20,027, 20,043,
20,118) listed control devices (steam strippers and scrubbers)
that should not have been considered MACT floor level of control
technologies because they were not installed for the reduction of
HAP. Several commenters (20,027, 20,043, 20,118) pointed out
that because steam strippers have never been calibrated or
operated for the continuous emission reduction of HAP's, they
should not have been considered a floor technology for HAP
reduction from process wastewater. One commenter (20,027) also
noted that because scrubbers have not been installed in bleach
plants to control methanol or total HAP's other than chlorine,
they should not have been considered as the floor technology, as
defined in section 112 (d) (3) (A) of the Act.
Response: Any technology that achieves HAP emission
reduction can be considered a potential MACT control option
regardless of whether or not the technology was installed for the
purpose of HAP reduction. There is no language in
section 112(d)(3) even suggesting that intent (i.e., the purpose
for installing air pollution control devices) is relevant for
purposes of establishing MACT floors. All that matters is the
"emission limitation achieved." In addition, as stated in
4-10
-------
section 112 (d) (2) of the Act, "Emission standards . . . shall
require the maximum degree of reduction in emissions of the
hazardous air pollutants that the Administrator, taking into
consideration the cost of achieving such emission reduction, and
any non-air quality health and environmental impacts and energy
requirements, determines is achievable for new or existing
sources . . . through application of measures, processes,
methods, systems, or techniques . . .'I Again, there is no
suggestion that the purpose for which existing controls were
installed is of any relevance.
4.1.6 Authority to Regulate Process Wa.stewater
Comment: Two commenters (20,027, 20,146) stated that the
Act does not give EPA the authority to determine applicability
for process wastewater provisions of the rule at the point of
generation. Rather, emissions may only be regulated at the first
air/water interface. The commenters (20,027, 20,146) contended
that EPA has an obligation to state its theory on this issue, and
support with data any arguments made to indicate that the HAP
content in wastewater is indicative of air emissions that warrant
regulation. Another commenter (20,011) stated that the
regulation of process wastewater at "point of generation" is
illegal unless EPA can demonstrate that it is infeasible to set
an emission limit, and also prove that concentration-based limits
are work practice standards under section 112(h) of the Act. One
commenter (20,146) indicated EPA's conclusions that the control
of HAP's from process wastewater was either a "floor" industry
practice or was needed to protect public health or welfare was
incorrect.
4-11
-------
Response: The Act does not place any restrictions on the
Administrator as to where within the affected source the
applicability determinations are made or where the controls are
applied to achieve the desired emission reductions. (Indeed,
controls can even be based on process changes, i.e., before a
point of wastewater generation (see Act 112 (d)(2) (A)). Regarding
air emissions from process wastewater, EPA's position on this
issue has been presented in several places including the proposal
BID (A-92-40, II-A-35) and the March 8, 1996 draft Chemical
Pulping Emission Factor Development Document (draft emission
factor document) (A-92-40, IV-A-6). Additionally, the methodology
of estimating air emissions associated with volatilization of
compounds from process wastewater has been well documented in
models such as EPA's WATERS. In general, EPA believes that
pollutants volatize from wastewater upon contact with the
atmosphere. This is consistent with standard laws of physics.
Therefore, wastewater streams need to be controlled at the point
of generation (i.e., at the first air-water interface) if HAP
emissions from wastewater are to be adequately controlled.
Additionally, EPA has found that the best controlled mills reduce
the pollutant loading in the process wastewater streams prior to
being recycled to process equipment or sent to subsequent
treatment.
4.2 DEFINITION OF SOURCE
4.2.1 Plant-wide Definition of Source
Comment: Several commenters (20,027, 20,049A3, 20,054A2,
20,056, 20,057A2, 20,059, 20,086, 20,089, 20,102, 20,103,
IV-D2-15) provided input on the definition of source.
4-12
-------
Several commenters (20,027, 203-210, 20,054A2, 20,056,
20,057A2, 20,086, 20,089, 20,103, 20,146, IV-D2-14, IV-D2-3)
supported a broad definition, stating that EPA should adopt a
plant-wide definition of source to allow for integrated
compliance with the proposed rule and to best comply with Act
section 112(g) provisions. Additionally, another commenter
(IV-D2-15) agreed with the broad single source definition which
includes the pulping processes, the bleaching processes, the
pulping and bleaching wastewater streams, paper machines, and
causticizing equipment. This definition, according to the
commenter (IV-D2-15) , reflects the physical realities of pulp and
paper mills which consist entirely of technically and
economically interdependent activities.
One commenter (20,056) suggested a narrow definition would
cause too many sources to become subject to the rule for minor
modifications, which would cause continual tinkering with the
emission control systems. Three commenters (20,049A3, 20,059,
20,102) indicated that the definition of source used by EPA was
lax and would enable facilities to undertake major modernization
projects without having to comply with the proposed standards.
One commenter (20,059) stated that a broad source definition
would delay by two years the date that new and reconstructed
sources would have to comply with new source MACT.
One commenter (20,092) supported the proposed narrow source
definition. The commenter (20,092) stated that the final rule
should clarify that the narrow source definition applies to new
and modified area sources. Additionally, one commenter (20,102)
suggested that the proposed definition be modified such that the
4-13
-------
source is defined as each of the process lines: pulping
processes, bleaching processes, and wastewater processes.
Response: The definition of affected source is used to
distinguish: (1) the collection of equipment or groups of
equipment that is subject to the emission limitations in the
rule; (2) equipment that is subject to the new source MACT
requirements; and (3) equipment considered in determining
reconstructed sources.
At proposal, EPA defined a single broad source for both
existing and new source MACT. That single source included the
pulping processes, the bleaching processes, and the pulping and
bleaching process wastewater streams at a pulp and paper mill.
The EPA also considered and solicited comments on the concept of
multiple smaller sources that would be subject to the existing
and new source MACT requirements.
In defining the source at proposal, EPA considered the
impact of the definition on mills making changes to existing
facilities. In general, the narrower the definition of source,
the more likely it is that changes to existing facilities would
be deemed "new sources" under the Act. With limited exceptions,
these new sources must be in compliance with new source standards
on the date of startup (or date the standards are promulgated,
whichever is later) . However, the Act and the CWA differ
regarding applicability requirements and compliance deadlines for
new sources. As such, EPA was concerned that a pulp and paper
mill planning to construct or reconstruct a source of HAP' s
between proposal and promulgation of these integrated regulations
would find it necessary to plan for compliance with the rule
(required on the date of promulgation) without knowing the
4-14
-------
requirements of the effluent guidelines for the industry. This
situation appeared to be inconsistent with one objective of the
integrated rulemaking: allowing facilities to do integrated
compliance planning. The EPA thus determined that the best
solution to these concerns was to define a single broad source at
proposal in order to reduce the applicability of new source MACT.
In the March 8, 1996 supplemental notice, EPA indicated a
continuing inclination for a broad, single source definition.
The EPA also discussed broadening the source definition further
to include paper making systems and causticizing equipment and
solicited comments on these additions. The EPA's reason for
considering the addition of these two equipment systems was to
facilitate implementation of the clean condensate alternative for
kraft mills. Commenters on the December 17, 1993 proposal and on
the March 8, 1996 supplemental notice largely agreed with the
broad, single source definition.
In considering how best to define the source, EPA did not
want to define it so narrowly that changes to or additions of
individual pieces of equipment would be subject to new source
MACT and be required to be in compliance with new source MACT at
startup. In fact, EPA was concerned that to do so could
discourage mills from implementing pollution prevention changes
as soon as practicable after promulgation of the proposed rule.
Such changes might include replacing an existing rotary vacuum
washer system with a low-flow washer system or installing an
oxygen delignification system, both of which if subject to
existing source requirements, would get the 8-year compliance
time (see chapter 16) . Once mills are complying with the
existing source MACT requirements, it also did not seem
4-15
-------
reasonable that they should have to tear out and rebuild that
vent collection system to accommodate small equipment changes in
the future unless those changes occurred along with other
substantial changes that would justify rebuilding the vent
collection system.
However, EPA also agrees with the commenter that at some
point, changes to an existing mill are substantial enough that
new source MACT should apply.
For the final regulation, EPA is defining the affected
source to which existing MACT requirements apply to include the
total of all HAP emission points in the pulping and bleaching
systems (including pulping condensates). In considering how
mills might engineer their vent collection systems and control
devices, EPA has concluded that the following construction
actions occurring after proposal are substantial enough that new
source MACT requirements will apply:
A pulping or bleaching system at an existing mill is
constructed or reconstructed; or
A new pulping line or bleaching line is added to an
existing mill.
The proposal date for mills that chemically pulp wood is
December 17, 1993. The proposal date for mills that mechanically
pulp wood, pulp secondary fibers, or pulp non-wood materials is
March 8, 1996. In selecting these actions, EPA determined that
the costs of complying at startup are reasonable and will not
discourage mills from implementing pollution prevention options
to comply with the proposed rule.
The final rule resolves the concerns of possible
circumvention of new source MACT applicability by specifying the
4-16
-------
control requirements for (1) greenfield sites, (2) the addition
of new equipment at existing sources, and (3) changes to existing
equipment that could trigger reconstruction. By designating the
exact equipment to be controlled at new and existing sources, the
rule reduces confusion and misinterpretation over what actions
trigger new source requirements. This approach preserves the
advantages of the broadest source definition for compliance by
existing sources while ensuring that new and reconstructed
equipment are regulated as new sources consistent with
section 112 (a) and 112 (d) of the Act. For example, under the
final rule a weak black liquor storage tank is not regulated at
an existing source. Nor would replacement of an existing tank be
regulated. But a new tank would be regulated at a greenfield
site or at an existing site if the new tank was installed
contemporaneously with the construction or reconstruction of a
new pulping system or an additional pulping line.
The final regulation also provides for an alternative
definition of source to facilitate implementation of the clean
condensate alternative. For mills using the alternative to
comply with the kraft pulping standards, the final regulation
defines a single broad source that includes the total of all
pulping, bleaching, causticizing, and paper making systems.
These additions were made to the definition of affected source to
allow for the application of advanced technologies to paper-
making and causticizing systems that typically receive recycled
or reused condensates. This broader definition allows increased
compliance flexibility while ensuring an equivalent level of HAP
control on a mill-wide basis.
4-17
-------
4.2.2 Definition Should Be Limited Within Each Subcategory
Comment: One commenter (IV-D2-14) cautioned that the
definition of source should be limited within each subcategory.
For example, a major change to a kraft mill should not draw a co-
located thermomechanical pulping mill into the new source MACT
standards. The commenter (IV-D2-14) suggested redefining source
by using the major subcategory rather than the artificial
divisions created within MACT I, MACT II, and MACT III.
Response: The proposed rule defined the affected source as
all pulping, bleaching, and wastewater components at a mill, in
combination. The final rule has been restructured to define the
affected source within each of six subcategories. The
subcategories are kraft, soda, sulfite, semi-chemical, mechanical
(wood) , and secondary or non-wood fiber pulping. The MACT new
source provisions would be applied within each of these
subcategories independently. For example, an affected source
would be all the emission points in the pulping and bleaching
systems of a kraft pulping system. If a sulfite-based pulping
system was co-located at the same mill, then a second affected
source would be all the pulping and bleaching emission points
within the sulfite process. Under these definitions, no
construction activities at the kraft system would affect the
applicability of new or reconstructed source provisions to the
sulfite system (and vice versa).
The proposed MACT II rule covers the chemical recovery
section of a pulping mill and would always be co-located with a
MACT I or MACT III source. The MACT II affected source covers
different equipment than the pulping and bleaching system
standards under MACT I and III. The affected source definitions
4-18
-------
do not overlap, and new source provisions of both rules,
therefore, apply independently. For example, no construction
activities on a kraft pulping or bleaching system would affect
the applicability of new or reconstructed source provisions
within the chemical recovery section of a kraft pulp mill (and
vice versa).
4.2.3 Woodpiles, Power Systems, and (Methanol) Recovery
Comment: One commenter (IV-D2-3) suggested including the
wood handling, power, and recovery components in the definition
of source to ensure they are not included in 112 (g) . Another
commenter (IV-D2-16) agreed with EPA's decision to exclude
woodpiles from the definition of source because woodpiles are not
significant HAP emission sources, emission controls are not
currently practiced, and collection schemes would be totally cost
prohibitive.
Response: The EPA contends that it is unnecessary to
include combustion devices in the source definition for this
NESHAP because they are covered under a separate standard.
The EPA agrees with the commenters that certain emission
points which are excluded from the definition of affected source
in today's rule, or are subject to a determination that MACT for
these operations is no control, should not be required to undergo
Act section 112 (g) review. The sources that have been so
identified are wood yard operations (including wood piles), tall
oil recovery systems, pulping systems at mechanical, secondary
fiber, and non-wood fiber pulping mills, and paper making
systems. With regard to wood yard operations, tall oil recovery
systems, and pulping systems at mechanical, secondary fiber, and
non-wood fiber pulping mills, EPA has determined that these
4-19
-------
sources do not emit significant quantities of HAP' s and EPA is
not aware of any reasonable technologies for controlling HAP's
from these sources. For paper making systems, EPA has not
identified any reasonable control technology, other than the
clean condensate alternative, that can reduce HAP emissions
attributable to HAP's present in the pulp arriving from the
pulping and bleaching systems. Additionally, EPA has determined
that the use of paper making systems additives and solvents do
not result in significant emissions of HAP's (A-95-31, IV-B-5).
Therefore, based on the applicability requirements of
section 112 (g) [40 CFR 63 part B, 63.40(b)], wood yard
operations; tall oil recovery systems; pulping systems at
mechanical, secondary fiber, and non-wood fiber mills; and paper
making systems would not be required to undergo section 112(g)
review. Any emission points that are specifically excluded from
control in a section 112(d) standard would not be required to
undergo section 112 (g) or 112 (j) (5) case-by-case MACT
determinations. To qualify for this exclusion does not require
that emission points be included in the affected source
definition. It is sufficient that they are specifically
addressed in the preamble or public record supporting the rule.
4.3 MACT
4.3.1 MACT Floor Level of Control Technologies
4.3.1.1 General Comments.
Inadequate data used to determine the MACT floor level of
control technologies.
Comment: Two commenters (20,018A1, 20,122) disagreed about
the information used to determine the MACT floor level of control
technologies and the stringency of the resulting MACT floor. One
4-20
-------
commenter (20,018A1) indicated that incomplete industry and
technology data and an abbreviated evaluation process for
existing technologies resulted in the selection of MACT floor
technologies that are more stringent than those currently used at
any single pulp and paper facility. The commenter (20,018A1)
indicated that some existing mills using proposed MACT technology
on certain vents in the mill would be unable to meet the
standards for all emission points included in the proposed rule;
the control technologies that EPA selected do not perform at the
levels which EPA has set at the mills where they are currently
installed. The commenter (20,018A1) suggested that EPA re-
evaluate the proposed rule based on new industry data to better
characterize control technology capabilities.
Response: At proposal, EPA delineated vent streams and
pulping wastewater streams controlled at the floor level of
control from those not controlled at the floor level of control
with numerical applicability cutoffs. The EPA used data
available at proposal, along with engineering evaluation
calculations to determine the performance capabilities of the
control equipment on which the floor level of control was based.
The EPA solicited comments and additional data on applicability
determinations and on control technologies and performance.
Since proposal, additional tests and studies were conducted
by the pulp and paper industry to provide these data. The
industry data received since proposal (A-92-40, IV-D1-29,
IV-Dl-29a, IV-D1-31, IV-D1-33, IV-D1-34, IV-D1-35, IV-D1-38,
IV-D1-39, and IV-D1-41) was considered by EPA and the Agency re-
evaluated the MACT floor by subcategories. As a result of this
data consideration, EPA has replaced the numerical cutoffs from
4-21
-------
proposal with specifically defined equipment systems and
associated named vents and pulping condensates. The EPA believes
these named streams more accurately identify the vents and
condensates being controlled at the best controlled sources. The
EPA has also used this additional data to evaluate the
performance capabilities of the controls on which the floor level
of control is based. A detailed discussion of many changes
related to determining applicability of the MACT standard and
control technology requirements are presented in the
March 8, 1996 supplemental notice. Additional changes are
discussed in this document. (Commenters on the March 8, 1996
supplemental notice supported EPA's decision to subcategorize,
which resulted in different MACT floor determinations.) In
instances where the commenters disagreed with the notice, in
particular, for new subcategories or provided additional data,
EPA re-evaluated the MACT floor level of control if new data were
received. The EPA maintains that the MACT floor level of control
determination is based on the best data available.
Cost-effectiveness is an improper criteria.
Comment: One commenter (20,059) stated that EPA used cost-
effectiveness as the primary criteria for selecting control
technology options, which they contended was improper.
Response: Cost-effectiveness was not the primary criteria
used to develop the MACT level of control. In developing the
MACT standard, EPA first determined the floor level of control as
defined in section 112 (d) of the Act. Costs were not considered
in developing the MACT floor level of control. For knotter and
screen systems, limited data were available to characterize
emissions. Cost-effectiveness was used as a means of supporting
4-22
-------
the emission limits developed from the emissions information.
Options more stringent than the floor level of control were then
identified. In evaluating the options more stringent than the
floor level of control, EPA considered a range of factors
including cost, emission reduction, energy impacts, and other
environmental impacts. Cost-effectiveness was only one of the
factors considered. This is precisely the type of evaluation
required by section 112(d)(2) and (3).
HAP-specific Effectiveness of Control Technology.
Comment: One commenter (20,059) stated that the prescribed
control technologies in the proposed regulations are not
effective for controlling all HAP's emitted from pulp and paper
processes. The commenter (20,059) also suggested that the
implications of control options on individual HAP's was not
evaluated by EPA. As an example, the commenter (20,059) reported
that bleach plant scrubbing works well for methanol but does not
provide control for other pollutants, such as chloroform,
formaldehyde, and carbon tetrachloride.
Response: The EPA recognizes that control devices may not
reduce emissions of every compound equally. However, EPA asserts
that the MACT standards for pulping vents and pulping wastewater
streams (combustion and steam stripping followed by combustion,
respectively) and bleaching vents (C1C>2 substitution, elimination
of hypochlorite use, and use of a caustic scrubber) represent the
maximum achievable control for the mixture of HAP's at pulp and
paper mills. The EPA evaluated other control technologies (such
as incineration of bleaching vent streams) and determined that
although some of the technologies could obtain better control of
some pollutants they would get worse control of others. The EPA
4-23
-------
also determined that it was not cost feasible to require these
other technologies in place of or in addition to the floor levels
of control (e.g., incineration of bleaching vents could achieve
greater reduction of non-chlorinated HAP emissions, but applying
a second technology in series with other controls would be
cost-prohibitive). Detailed discussions of this issue are
presented in section X.E. of the proposal, in the March 8, 1996
supplemental notice, and in chapter 20 of this document. The EPA
made some changes in the promulgated rule for bleaching system
control requirements. These changes were discussed in the
March 8, 1996 supplemental notice and in the preamble to the
final rule.
General comment on new vs. existing floor level of control
technology.
Comment: One commenter (20,027) stated that data do not
support any distinction between the floor level of control
technologies for new and existing sources. On the other hand,
one commenter (20,059) argued that it was not credible for EPA to
establish standards for new sources that were virtually identical
to existing source standards. New source MACT should reflect the
emissions limitation achieved by the best-performing similar
source.
Response: A discussion of the analysis used to determine
new source MACT is contained in section F of the December 1993
proposal preamble. As presented in the preamble, EPA considered
whether there were controls applicable to new sources beyond the
floor level of control MACT standard but concluded at proposal
that more stringent controls were not reasonable. The Act does
not require new MACT standards to be significantly more stringent
4-24
-------
than existing standards; the two standards can be virtually equal
if the best controlled source or similar source is virtually
identical to the average of the best-controlled 12 percent of
existing sources.
Since proposal, EPA has based MACT floor decisions on
specific named vents and wastewater streams controlled at each
pulping subcategory. A discussion of changes in the existing and
new source floor level of control determination since proposal is
presented in the March 8, 1996 supplemental notice. Additional
changes are discussed in section 4.3.1.2 of this document.
In the final rule, new source MACT differs from existing
source MACT. New sources are required to control additional
vents including: knotter and screening systems with mass
emission rates less than 0.05 kilograms of HAP per megagram of
ODP produced and 0.10 kilograms HAP per megagram ODP produced,
respectively (or less than 0.15 kilograms HAP per megagram ODP
produced combined), decker systems using process water other than
fresh water or Whitewater from paper machines or water with HAP
concentrations less than 400 ppmw, and weak black liquor storage
tanks at kraft mills; weak liquor tanks, strong liquor tanks, and
acid condensate tanks at sulfite mills; and pulp washing systems
at soda and semi-chemical mills.
4.3.1.2 Pulping Area.
MACT floor level of control needs to be determined by
subcategories.
Comment: One commenter (20,027) agreed with the EPA's
proposed floor level of 98 percent HAP control for pulping
emissions from kraft mills. Several commenters (20,027, 20,071,
20,072, 20,073) disagreed with the floor level of control
4-25
-------
technologies established for mills other than kraft. Two
commenters (20,027, 20,073) argued that the Act requires EPA to
consider different technologies inherent in each process when
determining the MACT floor level of control. The commenters
(20,027, 20,073) noted that failure to subcategorize mills by
pulping type and failure to recognize the different control
technologies and efficiencies for different pulping types caused
EPA to improperly establish the MACT floor level of control for
each pulping type.
Response: Information available at proposal did not
indicate a need to subcategorize the pulp and paper industry for
the purpose of setting MACT standards. As a result of new
operation, steam characterization, and control technology data
received after proposal (A-92-40, IV-D1-29, IV-Dl-29a, IV-D1-31,
IV-D1-33, IV-D1-34, IV-D1-35, IV-D1-38, IV-D1-39, and IV-D1-41),
EPA established subcategories for mills according to pulping
process (kraft, sulfite, semi- chemical, and soda). The data
indicated that sufficient differences exist between kraft and
sulfite, soda, and semi-chemical processes to warrant
subcategorization. Accordingly, EPA revised the MACT floor level
of control and MACT determinations for each subcategory. A
discussion of development of pulping subcategories and respective
floor level of control determinations is presented in the
March 8, 1996 supplemental notice and received essentially
unanimous support by commenters.
Kraft pulping MACT floor control technology.
Knotter and Screening Systems.
Comment: Several commenters (IV-D2-8, IV-D2-7, IV-D2-15)
requested that knotter and screening vents preceding brownstock
4-26
-------
washing should not be controlled by the rule. One commenter
(IV-D2-15) explained that based on an erroneous interpretation of
the 1992 MACT survey data, EPA has incorrectly concluded that
knotter and screening systems are controlled by 7 percent of the
systems and are therefore part of the floor level of control.
The commenter (IV-D2-15) stated that following proposal NCASI
contacted those mills which had indicated that: (1) their
knotters were not vented or (2) vent gases from the knotters were
collected and incinerated. The commenter (IV-D2-15) asserted
that only 4 percent controlled knotter system vents. There were
no "not vented" systems. Therefore, the commenter (IV-D2-15)
stated that the pre-washer knotting and screening systems are not
controlled at the floor level.
Response: The EPA has reviewed available data on knotter
and screen systems and has concluded that these systems are
controlled sufficiently to establish a MACT floor level of
control, and also that control more stringent than the floor
level of control is not warranted. Data used to reach this
conclusion include survey responses from the 1992 voluntary
survey, follow-up telephone surveys conducted by the NCASI, and
emissions data from the NCASI 16-mills study. Although the data
indicates that many of these systems are currently controlled to
some degree, the survey responses were not detailed enough in
their equipment system descriptions and the test data were too
limited for the Agency to use these two sources of information
alone to develop the MACT control requirements. Because
equipment designs, nomenclature, and control configurations vary
across the industry, the Agency decided that a HAP emissions
limit would be the best way for mills to determine which systems
4-27
-------
would require control. The EPA lacks sufficient data, however,
to pinpoint any single value that represents the MACT floor level
of control. Rather, based on the survey and test data, there are
a range of values from which EPA could choose. The EPA further
considered the costs of control in choosing from this zone of
reasonable values.
Of the 171 knotter systems reported in the 1992 voluntary
survey, 12 knotter systems at 5 mills were reported as controlled
and ducted into the noncondensible gas (NCG) collection system
and another 49 knotter systems at 23 mills were reported as
having no vents. NCASI followed up by telephone surveys with
these 28 mills (A-92-40, IV-D1-112, IV-D1-114). The follow-up
surveys indicated a moderate amount of misreporting at these
28 mills. NCASI did not resurvey all 171 knotter systems.
Therefore, the following knotter system floor determination
assumes that the mills not resurveyed that originally reported no
knotter system controls did not control any vents.
From the 28 mills resurveyed, it was determined that
six knotter systems or 3.6 percent (6/171) route all vents into
the NCG collection system; another two knotter systems or
1.2 percent (2/171) route all knotter hood vents into the NCG
collection system; another eight knotter systems or 4.7 percent
(8/171) use only pressure knotters; and another two knotter
systems or 1.2 percent (2/171) route all vents to the smelt
dissolving tank scrubber. Industry collected data at seven
pressure/open (also referred to as pressure/vibrating) knotter
systems and found the methanol emissions to range from
0.005-0.07 kilograms per megagram of ODP produced, and collected
data at one pressure knotter system and found the methanol
4-2S
-------
emissions to be 0.0034 kilograms per megagram ODP produced.
Emissions data are summarized in the Chemical Pulping Emission
Factor Document (A-92-40, IV-A-8). Because the pressure knotter
system emissions were lower than the emissions at the
pressure/open systems, pressure systems can be considered a type
of controlled system. Therefore, 18 or 10.5 percent (18/171) of
the knotter systems have some level of emissions control. The
Agency believes this estimate of number of knotter systems
controlled may be somewhat low because it is uncertain how many
of the mills not resurveyed may have had the lower emitting
pressure systems.
The 1992 voluntary MACT survey responses indicated that
96 screening systems out of the 199 reported are not vented.
NCASI resurveyed by telephone 41 of these 96 mills. Assuming
that the 55 mills not resurveyed look similar to the 41, the
follow-up survey determined that 7 percent (6/41 x 96/199) route
their vents to the NCG collection system and 41 percent (35/41 x
96/199) have closed screens that vent through auxiliary tanks.
Therefore, 48 percent of the screening systems have some level of
control.
Industry collected data at one closed screen system and one
open screen system. The closed screen system tested had methanol
emissions of 0.004 kilograms per megagram of ODP produced. The
open screen system tested had methanol emissions of
0.22 kilograms per megagram of ODP produced.
The Agency considered how best to characterize the average
emissions limitation achieved by the best controlled 12 percent
of the knotter systems and screen systems given the wide variety
of control scenarios present in the industry. Either collecting
4-29
-------
and controlling vents on an open system or using closed equipment
result in lower air emissions. The Agency decided to select the
emissions limitation using the test data from the closed and open
equipment systems. The Agency's decision is due in part to the
fact that the effluent limitation guidelines and standards being
promulgated will require that screening areas be closed for water
discharge, which will require mills to move toward wider use of
the lower air emitting pressure systems.
Because there is only one test data point for the pressure
knotter systems and that emissions value is similar to the low
end of the range of data points for the pressure/open knotter
systems, the Agency did not believe it would be appropriate to
set the emission limit equal to the one pressure knotter system.
Similarly, because there is only one test data point for closed
screens, the Agency did not believe it would be appropriate to
use that single data point to set the emission limit for
screening systems. The Agency could have selected any emission
limit within the range of all available data for knotters (i.e.,
0.0042 to 0.07 kilograms per megagram of ODP produced) and
screens (i.e., 0.004 to 0.22 kilograms per megagram of ODP
produced). However, recognizing the limited data available, the
Agency also considered the cost-effectiveness of controlling
these systems to aid in setting the emission limits within the
range of reasonable values (A-92-40, IV-B-21).
Based on consideration of all available data, the final rule
requires that existing kraft sources are required to control all
knotter systems with total mass emission rates greater than or
equal to 0.05 kilograms of HAP per megagram ODP produced.
Existing kraft sources are required to control all screening
4-30
-------
systems with total mass emission rates greater than or equal to
0.10 kilograms of HAP per megagram ODP produced. Since it is
often difficult to distinguish between the knotter system and
screening system at mills, a mill may also choose to meet a total
emissions limit of 0.15 kilograms per megagram ODP produced
across the knotting and screening combined system. New sources
are required to control all knotter and screen systems,
regardless of emissions level.
Browns took Washers.
Comment: Several commenters (20,027, 20,054A2, 20,066A3,
20,070A1) indicated that EPA's determination of floor level of
control technology for brownstock washer control was erroneous
because it failed to recognize the distinctions among types of
washers. One commenter (20,066A3) indicated that EPA included
the following equipment in one group: red stock washers at
sulfite mills, which are enclosed but routed to a scrubber; low
emitting washers (such as diffusion washers that cannot be used
with batch digesters); and drum washers which require a major
expense to enclose because they emit high-volume
low-concentration streams.
One commenter (20,054A2) stated that EPA failed to recognize
the significant differences between vacuum, pressure, and
diffusion washers when establishing the floor level of control
for washers. One commenter (20,070A1) suggested that a
distinction should be made in the regulation between newer washer
systems and the older vacuum drum washers. One commenter
(20,027) also noted that diffusion washers are excluded from the
kraft NSPS due to low emission rates. The commenter (20,027)
added that new washers are expected to be non-rotary vacuum
4-31
-------
design. One commenter (IV-D2-7) noted that because of the low
level of emissions from improved washer systems and because of
the lack of existing controls on such units (2 of 21 operational
chemi-washers), the MACT floor for improved washers (pressure
washers, diffusion washers, and horizontal belt washers) should
be no control.
Response: The EPA recognizes the difference in emissions
and flow characteristics among the different types of pulp
washers. However, information collected in the 1992 voluntary
MACT survey showed that greater than 25 percent of low flow
washer systems (diffusion, pressure, chemi-) are controlled
(A-92-40, IV-B-8). Based on these data EPA determined that the
control of pulp washers was part of the floor level of control
for all types of washers at kraft and sulfite mills. As
discussed in the March 8, 1996 supplemental notice, EPA
encouraged the use of low flow washer systems because of
significant pollution prevention advantages and environmental
benefits. The EPA has extended compliance with the kraft pulping
standards for HVLC systems by 5 years in order to promote the use
of low flow washer systems, as part of the strategy to encourage
water pollution controls more stringent than BAT, and to provide
sufficient time to design and construct these systems.
Deckers.
Comment: One commenter (IV-D2-15) stated that control of
the decker was beyond the floor level of control. The commenter
(IV-D2-15) said that all the deckers which were reported in the
1992 voluntary MACT survey as not being vented actually had vents
somewhere in the system, either for the hood, for the filtrate
tank, or for both. Vent gases from nine of the decker systems
4-32
-------
were collected and incinerated. Nine deckers represent less than
5 percent of the reported systems. Therefore, the commenter
(IV-D2-15) asserted that decker vents are not controlled at the
floor level.
The commenter (IV-D2-15) stated that industry will collect
more information about the existing decker systems for the
purpose of evaluating the cost-effectiveness of collecting and
incinerating these vent gases. However, the commenter (IV-D2-15)
speculated that it is not cost-effective to control emissions
from the decker system.
Response: The EPA evaluated the information submitted by
the commenter and concluded that there were 170 decker systems in
mills responding to EPA's industry survey questionnaires. All
the decker systems are associated with bleached mills. Of the
170 decker systems, 14 are controlled (8 percent) (A-92-40,
IV-B-22). Therefore, control of decker systems is in the MACT
floor.
The majority of decker systems controlled at the floor level
of control (10 systems) are associated with oxygen
delignification systems or are being used as an additional stage
of pulp washing. The Agency believes that these types of decker
systems are operated similarly to and have similar emissions as
pulp washers. Decker systems used in this manner receive
contaminated condensates or filtrates that may be recycled from
other processes, such as the oxygen delignification system or
combined condensate tanks. The process water may have a HAP
concentration that would release significant amounts of HAP to
the air from the air-water interface. The Agency characterized
the emissions from this source to identify the types of decker
4-33
-------
systems with high emissions. Information supplied in NCASI
Technical Bulletin No. 678 provided a relationship between air
emissions and methanol concentrations in process water used in
rotary vacuum drums. The EPA evaluated this relationship and
determined that decker controls and higher HAP emission rates
were associated with deckers that used process water with HAP
concentrations greater than or equal to 400 ppmw, or that did not
use fresh water or "Whitewater" from paper making systems
(A-92-40, IV-B-22) .
Therefore, the Agency has determined that it is appropriate
to make a distinction among types of decker systems at existing
sources for the purpose of setting the MACT standard. Decker
systems at existing sources using fresh water or "Whitewater"
from paper making systems, or using process water with HAP
concentrations less than 400 ppmw, are not required to be
controlled. Decker systems at new sources are required to be
controlled regardless of the HAP concentration in the process
water introduced into the decker.
Oxygen Delignification Systems.
Comment: Two commenters (20,027, 20,054A2) asserted that
EPA improperly determined that the floor level of control
includes control of oxygen delignification systems. One
commenter (20,027) stated that less than 6 percent of these
systems are controlled even if mills with oxygen delignification
systems are considered their own subcategory. Additionally, the
commenter (20,027) contended that oxygen delignification systems
should be considered part of the bleach plant since oxygen acts
as a bleaching agent similar to chlorine or C102 -
4-34
-------
One commenter (IV-D2-5) asserted that requiring incineration
of vent gases from oxygen delignification units is punitive to
mills that have expended the extra effort to install these
environmentally beneficial systems. The commenter (IV-D2-5)
claimed that molecular oxygen is very effective at oxidizing
pollutants such as TRS and organics such as methanol. The
commenter (IV-D2-5) included data from their mill showing a low
methanol emissions rate from their oxygen delignification system.
Response: The EPA disagrees with the commenters and has
determined that control of oxygen delignification systems is part
of the floor level of control (A-92-40, IV-B-16) based on the
data collected in the 1992 voluntary MACT survey. An evaluation
of the number of oxygen delignification systems constructed after
proposal and their controls (A-92-40, IV-D1-29, IV-B-16)
indicates that greater than 6 percent of the oxygen
delignification systems are controlled. Therefore, the MACT
floor level of control is control of oxygen delignification
systems.
With regard to the commenter's contention that oxygen
delignification systems should be considered part of the bleach
plant, information submitted to the Agency following proposal
(A-92-40, IV-D1-97, IV-D1-104) indicated that several commenters
from the industry have revised their position and recommend that
oxygen delignification systems be considered part of the pulping
process. The basis for this recommendation is that process
waters from oxygen delignification systems are typically recycled
or reused in other pieces of pulping equipment. Conversely,
process waters originating in the bleach plant cannot be used in
the pulping process without extensive treatment due to
4-35
-------
interferences in the chemical recovery process caused by the
presence of chlorine and chlorinated compounds.
Regarding the issue raised by the commenter that it is
punitive to control air emissions from the environmentally
beneficial oxygen delignification systems, EPA recognizes that
some mills have already committed to using oxygen delignification
systems. But, based on industry emissions data submitted after
proposal (A-92-40, IV-D1-29 and IV-B-16) oxygen delignification
systems appear to be a significant source of HAP emissions, and
greater than 6 percent of the existing oxygen delignification
systems are controlled. A floor level of control is mandated by
the Act. The EPA does not think that MACT compliance (which will
apply to all sources with oxygen delignification) will discourage
introduction of the technology. In fact, to encourage the pulp
and paper industry to consider the benefits of oxygen
delignification, as discussed in the March 8, 1996 supplemental
notice, the final rule grants kraft mills a compliance extension
of 5 years. There are also significant incentives provided in
the effluent guidelines portion of the final rule to encourage
use of oxygen delignification (or superior) technology.
Weak Black Liquor Storage.
Comment: Several commenters (IV-D2-8, IV-D2-7, IV-D2-15)
maintained that the MACT floor level of control for weak black
liquor storage tanks is no control, noting that the cost of
controlling these tanks far outweighs the environmental benefits
and that add-on controls would threaten the structural integrity
of these units. One commenter (IV-D2-15) explained that the
NCASI survey provided ambiguous responses, which misrepresented
the extent of control at existing weak black liquor storage tanks
4-36
-------
and that they are not part of the floor level of control. In
order to resolve this ambiguity, NCASI contacted all the mills
that had reported that vent gases from their weak liquor storage
tanks were collected and incinerated. NCASI also sent a single
page survey to 121 kraft mills in the industry and received
117 responses. From this information, the commenter (IV-D2-15)
asserted that only 5 percent of weak black liquor storage tanks
were controlled. Therefore, the commenter (IV-D2-15) asserted
that weak black liquor storage tanks are not controlled at the
floor level.
The commenter (IV-D2-15) stated that control of the weak
black liquor storage tanks beyond the floor level of control
would not be cost effective, and the emissions from these tanks
based on NCASI's latest tests indicated that the quantity of
emission is negligible. One commenter (20,027) asserted that
EPA's assumption that tanks could withstand a vacuum for routing
vents to a control device was inappropriate for older tanks
because the vacuum may cause the tanks to collapse. One
commenter (IV-D2-4) suggested that it is appropriate that any
weak black liquor storage tank strong enough to withstand
sufficient vacuum (based on engineering analysis or the age of
the tank) should be subject to control. One commenter (IV-D2-10)
agreed with EPA's position (as outlined in the March 8, 1996
supplemental notice) that the age and, therefore, the structural
integrity of the weak black liquor storage tanks should be
considered as one parameter for determining control applicability
of tanks. Several commenters (IV-D2-14, IV-D2-8), however,
disagreed that the age of a tank is a good parameter for
determining the control applicability for tanks.
4-37
-------
Response: The EPA evaluated the supplemental information
submitted by NCASI (A-92-40, IV-D1-101) and concurs that control
of weak black liquor storage tanks is not in the existing source
floor level of control. Information submitted by the commenters
indicated that of the 597 weak liquor storage tanks in the survey
only 28 (4.7 percent) actually had emissions routed to a control
device (A-92-40, IV-D-106). Some respondents had previously
included other types of controlled tanks, such as washer filtrate
tanks, in their totals because EPA's original survey did not
provide a definition of weak liquor storage tanks. The Agency,
therefore, has concluded that the MACT floor level of control for
weak liquor storage tanks at existing sources is no control.
While some tanks are controlled, available information does not
support the supposition that age is a good parameter for
distinguishing structural integrity. No other parameter could be
identified for distinguishing between controlled and uncontrolled
tanks. Therefore, no basis for controlling existing sources was
determined. In addition, the Agency evaluated the cost of going
beyond the floor level of control to control weak liquor tanks.
The results of EPA's analysis indicated that a significant cost
would be incurred for a limited emission reduction. This
analysis is presented in chapter 20 of the background information
document for the promulgated rule. The EPA concurs with the
comments that older tanks could not handle the vacuum caused by
the closed-vent collection system without collapsing.
Additionally, sweep-air systems that could be used to alleviate
the vacuum problem are cost prohibitive for the amount of
emissions reduction achieved by controlling the tanks.
4-38
-------
Therefore, the Agency agrees with the commenters that control
beyond the floor is not justified.
Although weak black liquor storage tanks are not controlled
at the floor, the available data does indicate that some tanks
are being controlled at some mills. Therefore, EPA has concluded
that these tanks should be controlled at new sources since new
tanks could be designed to withstand the slight vacuum associated
with the collection system at a reasonable cost.
Comments on sulfite pulping MACT floor level of control
technology.
Comment: One commenter (20,027) stated that EPA incorrectly
concluded that control technologies common in mills with one
pulping process would be applicable to other pulping processes.
The commenter contended that incineration is the common control
technology for total reduced sulfur compounds. However, 14 of
the 15 sulfite mills do not practice any form of vent gas
incineration. The commenter stated that scrubbing and SC>2
recovery should be considered as the floor level of control
technology for pulping vents at sulfite mills. The commenter
(20,027) stated that the sulfite process generates sulfur dioxide
emissions which are typically recovered using scrubbers, for
reuse as cooking acid makeup. The commenter (20,027) reasoned
that since any HAP generated in a sulfite digester may pass
through multiple process devices, each one capable of altering
the HAP emission rate, the final HAP emission rate is a function
of all of the equipment that exists between the point of
generation and the eventual emission point. The commenter
(20,027) contended that to ascribe a removal or control
efficiency to the final scrubber would be to ignore the entire
4-39
-------
control process. The commenter (20,027) concluded that for this
reason EPA should establish the pulping component standard for
sulfite pulping at the exit of the control device.
Another commenter (20,151) suggested that the proposed rule
should address emission control and discharge requirements on
C1O2 generating equipment and other replacement types of
bleaching, such as hydrogen peroxide and ozone. The commenter
(20,151) indicated that consideration should be given to
appropriate controls and discharge requirements on strong waste
ponds, cooling towers, and aeration ponds.
Response: Since proposal, EPA has established a separate
subcategory for the sulfite process and has re-evaluated the
floor determination for the sulfite process. The EPA agrees with
the commenter that the floor level technology is scrubbing and
S02 recovery. A discussion of the analysis for determining the
level of the standard for the sulfite process is presented in the
March 8, 1996 supplemental notice. In the March 8 notice, EPA
provided a control efficiency requirement and an emission limit
requirement. For calcium-based sulfite pulping processes, the
emission limit presented was 0.65 Ib methanol/ODTP and
the percent reduction was 92 percent. For ammonium- and
magnesium-based sulfite pulping processes, the emission limit was
1.10 Ib methanol/ODTP, and the percent HAP reduction was
87 percent. The Agency developed applicability cutoffs based on
methanol because only methanol emissions data were obtained for
all of the equipment systems and wastewater streams considered
for control at sulfite mills. The test data from sulfite mills
also indicated that for the equipment systems tested for other
HAP's, methanol comprised the majority of HAP emissions.
4-40
-------
Therefore, the Agency believes that the maximum control of HAP
emissions will be achieved by controlling methanol as a
surrogate.
Since the supplemental notice was published, EPA has further
evaluated the industry's data regarding process variability. The
data indicates that methanol emissions from individual process
vents varied significantly over time (A-92-40, IV-B-20). The
industry data that were used to develop the initial emission
limits cited in the March 8, 1996 supplemental notice were based
on limited information that did not account for this process
variability. One of the compliance options for sulfite mills is
a numerical emissions limit. The EPA determined that the
appropriate limit should incorporate the process variability
inherent in normal operation. The EPA determined the amount of
variability associated with a 99.9 percent confidence level in
the data supplied by the industry. This amount of variability
(confidence interval), therefore, was applied to the average
emission limits from the best controlled mills to develop the
final emission limit. After the close of the March 8, 1996,
Federal Register supplemental notice comment period, additional
information was provided to the Agency that indicated that the
sodium-based sulfite pulping process is in use at some mills
(A-92-40, IV-E-86, IV-E-94). No emissions information was
available for this process. However, the Agency determined that
due to the similarities in processes between calcium- and sodium-
based sulfite pulping processes, the same limit developed for
calcium-based mills would be applicable to sodium-based mills.
For sodium- and calcium-based sulfite pulping processes, the
final emission limit is 0.44 kilograms of methanol per megagram
4-41
-------
of ODP produced. For ammonium- and magnesium-based sulfite
pulping processes, the final emission limit is 1.1 kilogram of
methanol per megagram of ODP produced. Because the variability
is included into the emission limits, these emission limits and
corresponding monitoring parameters are never to be exceeded
values.
Comment: One commenter (IV-E-91) indicated that a mill they
are working for uses a sodium-based sulfite process and the
March 8, 1997 Federal Register supplemental notice does not
appear to address this specific process.
Response: This comment was submitted to EPA after the close
of the comment period for the March 8 notice and shortly before
promulgation; however, EPA has reviewed and evaluated the
commenter's assertions. Based on a review of the information
contained in the 1996 Lockwood-Post's Directory (A-92-40,
IV-J-87) and discussions with mill operators (A-92-40, IV-E-94),
EPA has decided that the pulping process used at the mill meets
the definition of sulfite pulping proposed in the
December 17, 1993 Federal Register notice (see 58 FR 66176) . The
March 8, 1996 supplemental MACT notice did not propose to amend
the sulfite definition.
Although EPA does not have data specific to the sodium-based
sulfite pulping process, EPA believes it is reasonable to group
this process with the calcium-based sulfite pulping process for
purposes of the MACT standard. This decision was made since the
calcium- and sodium-based pulping process have similar equipment
such as an acid making system and, unlike the ammonium- and
magnesium-based sulfite processes, neither the calcium- nor the
sodium-based sulfite process utilizes recovery furnaces.
4-42
-------
Therefore, the final rule specifies that the calcium-based
sulfite process methanol emission limits and percent reductions
are applicable to the sodium-based sulfite pulping process.
The EPA believes that this is the only mill currently using
a high-yield sodium-based sulfite pulping process, based on a
review of the Lockwood-Post information. However, the
information available to EPA also indicates that this mill
utilizes some degree of mechanical refining in the pulping
process and has a single peroxide bleaching stage. Based on this
information, the Agency has assigned this mill to the semi-
chemical pulping subcategory for purposes of the effluent
limitation guidelines and standards, but is not setting revised
effluent limits at this time for this subcategory in this
promulgation.
The EPA believes that grouping this pulping process into
separate subcategories within the MACT and effluent limitation
guidelines and standards is consistent with the regulatory intent
of the two EPA programs since the high-yield sodium-based sulfite
pulping process has characteristics of both the sulfite and semi-
chemical pulping processes (sulfite from an emissions standpoint
and semi-chemical from a liquid discharge standpoint).
Comments on semi-chemical pulping MACT floor control
technology.
Comment: Two commenters to the proposal (20,027, 20,071)
maintained that the MACT floor level of control for pulping vent
control at stand-alone semi-chemical mills should be no control,
because none of the existing stand-alone semi-chemical mills are
controlled.
4-43
-------
Regarding the information contained in the March 8, 1996
supplemental notice, in which the Agency discussed separate MACT
standards under consideration for semi-chemical mills, one
commenter (IV-D2-15) agreed with EPA that the collection and
control of LVHC vents is a cost-effective control option and
represents the floor for existing stand-alone and co-located
semi-chemical mills. The commenter (IV-D2-15) also agreed that
new source MACT for semi-chemical mills should be the control of
the LVHC system plus the control of emissions from the pulp
washing system. The commenter (IV-D2-15) agreed that the MACT
for semi-chemical wastewater is no control.
Response: Information provided by industry in survey
responses and after proposal (A-92-40, IV-D1-41, IV-D1-80,
IV-D1-86, IV-D1-89, IV-D1-90, and IV-D1-93) confirmed that the
MACT floor level of control at semi-chemical mills is collecting
and controlling LVHC vents. The Agency determined that it was
not reasonable to control other emission points at existing
semi-chemical mills (A-92-40, IV-B-12). New source MACT is based
on the best controlled at sources. Data indicate that the
best-controlled semi-chemical mills combust LVHC emissions and
emissions from pulp washers. New sources, therefore, are
required to control LVHC vents and emissions from the pulp
washers. A detailed discussion of the level of the standards for
semi-chemical processes is contained in the March 8, 1996
supplemental notice and in the docket (A-92-40, IV-B-12).
Comments on soda pulping MACT floor level of control
technology.
Comment: Regarding the December 1993 proposal, one
commenter (20,072) argued that since EPA has historically
4-44
-------
regulated kraft mills differently than soda mills, the MACT floor
level of control would be significantly different between kraft
and soda mills. The commenter (20,072) stated that by combining
the two existing soda mills with over 100 kraft mills, the
difference in the actual MACT floor was lost.
With regard to the information contained in the
March 8, 1996 supplemental notice, in which the Agency discussed
soda mill requirements, one commenter (IV-D2-15) agreed with EPA
that both the existing and new source MACT for the soda mill
subcategory is the collection and control of LVHC vents. The
commenter agreed that the MACT for wastewater at soda mills is no
control.
Response: Data available to EPA indicate that soda mills do
not currently control any of the equipment that is subject to the
MACT requirements for kraft mills. Therefore the floor level of
control is no control. However, EPA has determined that the
emissions from soda mills are similar to kraft mills (with the
exception of TRS compounds) and control of LVHC vents is
technically feasible and can be achieved at a reasonable cost.
The EPA also determined that controlling additional vents beyond
the LVHC vents at existing soda mills could not be achieved at a
reasonable cost. However, controlling the pulp washing system at
new soda mills could be achieved at a reasonable cost.
Therefore, the final rule requires existing soda mills to collect
and control LVHC vent streams and new mills to control LVHC vents
and pulp washing system. The commenters are referred to the
March 8, 1996 supplemental notice and (A-92-40, IV-B-12) for a
detailed discussion of EPA's determination of soda mill
requirements.
4-45
-------
Closed-vent collection system.
Comment: Two commenters (20,027, 20,054A2) indicated that
EPA did not evaluate the operational requirements and performance
of the average of the best-performing 12 percent of sources when
developing the proposed closed-vent collection system
requirements. One commenter (IV-D2-15) requested that EPA re-
evaluate the calculation of the floor level of control technology
for enclosure of pulping component sources and associated gas
conveyance systems. Another commenter (IV-D2-7) asserted that no
total enclosures and closed-vent systems are currently in place
in the industry, and that their addition would not be cost
effective in any existing facility.
Two commenters (IV-D2-7, IV-D2-15) requested that EPA scale
back the visual inspection and leak detection requirements for
the gas conveyance systems since the proposed regime does not
represent the floor. In addition, since most conveyance systems
operate at negative pressure, there is no need for leak
detection. One commenter (20,027) stated that the proposed
visual inspections for closed-vent collection systems were
unnecessary due to the design (limited use of flanges) and type
of materials of construction (stainless steel) used in LVHC and
HVLC collection systems. Another commenter (20,057A2) stated
that there were no data to support the inclusion of no detectable
leaks from pulping and bleaching process vent collection systems
as part of the MACT floor.
One commenter (20,027) asserted that the best-performing
12 percent of sources do not seal or lock bypass vents.
Response: The EPA proposed requirements that the Agency
deemed reasonable to ensure that the closed-vent collection
4-46
-------
systems are properly operated and that the affected vent
emissions are conveyed to the control devices. The EPA has
evaluated the comments submitted after the December 1993 proposal
and the March 1996 supplemental notice, and several revisions
were made to the closed-vent collection system requirements.
The requirement for demonstration of negative pressure has
been revised to apply only to enclosures and hoods. The
requirement for demonstration of no detectable leaks has been
revised to apply only to positive pressure systems or portions of
systems. The EPA concluded that the leak detection requirements
are necessary to verify that enclosures are collecting all
emissions from applicable emission points in these systems. The
EPA agrees with the commenters that leak detection for negative
pressure systems is not useful since any leaks in the collection
system will draw air into these systems.
The bypass line requirements were also revised. The proposed
rule language requiring lock-and-key type seals was replaced with
language specifying car-seals or seals that can easily be broken
in case of emergencies, yet still indicate when the bypass valve
position has changed. Additionally, the final rule specifies the
use of log entries to record valve position.
The EPA disagrees with the commenters that the visual
inspections are not necessary. No changes were made to the
proposed visual inspection requirements for closed-vent
collection systems since the intention in the rule was to inspect
for bypass valve position, clogged drains, broken fan belts, etc.
These problems are not necessarily affected by the design or
material of construction of the system. A related issue on
downtime and back-up controls is discussed in section 4.3.4.
4-47
-------
4.3.1.3 Bleaching.
Suggested alternative MACT floors for bleaching.
Comment: Several commenters (20,018A1, 20,027, 20,036A1,
20,045, 20,051, 20,056, 20,057, 20,115A2, IV-D1-4, IV-D1-8,
IV-D1-15, IV-D1-16) discussed the floor level of control for
chlorine, chlorinated HAP's, and non-chlorinated HAP's .
With regard to the December 1993 proposal, several
commenters (20,027, 20,056) agreed that scrubbing is the correct
control technology for bleach plants but that EPA did not
correctly establish control efficiencies or pollutants
controlled. Several commenters (20,018A1, 20,027, 20,036A1,
20,045, 20,051, 20,056) reasoned that because existing bleach
plant scrubbers are not effective on any HAP except chlorine, the
floor level of control for methanol and HAP's other than chlorine
should be no control. One commenter (20,027) concluded that to
go beyond the floor of no control, the cost-effectiveness should
be evaluated. (Data provided: Table 10 p. MACT-190, Table 11
p. MACT-191, Table 12 p. MACT-194, and Appendix MACT 22.) Two
commenters (20,056, 20,070A1) indicated that the control
efficiency selected had not been shown to be achieved in practice
by the best-performing 12 percent of mills. One commenter
(20,115A2) argued that reduction of chlorine and methanol by
99 percent using a scrubber, as specified in the proposed
regulations, would not be possible in the pulp and paper industry
or any other industry. One commenter (20,057) stated that EPA
failed to consider methanol generation rates or scrubber removal
efficiencies when they established the proposed MACT standards
for bleach plant vents.
-------
Several commenters to the March 8, 1996 supplemental notice
(IV-D2-15, IV-D2-16, IV-D2-8) agreed with EPA's conclusion that
the data provided by NCASI supports the control of chlorinated
HAP only in the bleaching component and the MACT floor should be
control of only chlorinated HAP's.
Several commenters (20,018A1, 20,027, 20,080, 20,149,
IV-D1-3, IV-D1-4, IV-D1-8, IV-D1-10, IV-D1-15, IV-D1-16,
IV-D1-17, IV-D1-18) discussed the MACT floor based on process
changes and effluent guidelines BAT requirements.
With regard to comments solicited on the MACT control
technology basis for bleach plants in the December 1993 proposal,
one commenter (20,018A1) indicated that complete C1C>2
substitution should be MACT for the bleaching component because
the effluent guidelines required complete substitution of
chlorine with C102 for kraft bleaching, and C1C>2 bleaching has
lower emissions of chlorine and C102- Two commenters (20,027,
20,059) supported emission limits for chlorinated organic
compounds set based on the reductions obtained by process
changes. One commenter (20,071) suggested compliance for both
chlorine and chloroform could be demonstrated by elimination of
hypochlorite and complete C102 substitution. Several commenters
(20,049A2, 20,091, 20,102, 20,103, 20,127, 20,129) stated that
chloroform emissions should be minimized by using process changes
or through the use of advanced innovative technologies, such as
biofiltration following gas scrubbing. One commenter (20,149)
indicated that the proposed rules did not control chloroform
emissions which they state posed a significant health risk. One
commenter (20,118) suggested deferring the requirement to control
chloroform emissions until the impact of 100 percent C102
4-49
-------
substitution and elimination of hypochlorite stages has been
evaluated. Another commenter (20,027) reported that reliable
data are not available regarding the effect of process changes on
HAP emissions from bleach plant vents; consequently, it would not
be appropriate to use a combination of process changes and gas
scrubbing to set the MACT floor level of control for bleach plant
vents. Several commenters (20,091, 20,138, 20,141, 20,143,
20,156) indicated that the rules should be written as a guide for
a complete phaseout of processes using chlorine or CICH compounds
to bleach pulp. One commenter (20,091) indicated that if a
complete phaseout is not possible, they prefer maximum ciOo
substitution to alternatives that do less to reduce the formation
of organochlorines.
Two commenters (20,027, 20,115A2) stated that a chloroform
emissions limitation is not needed for the bleach plant because
process modifications will reduce chloroform.
One commenter (20,031) suggested, however, that EPA rewrite
the bleaching component standards to allow the continued use of
hypochlorite as a bleaching agent on a site-specific basis to
address other environmental concerns.
Regarding the information presented in the March 8, 1996
supplemental notice, several commenters (IV-D2-15, IV-D2-17,
IV-D2-16, IV-D2-8) agreed with EPA's intent to consider
compliance with the effluent guidelines BAT option equivalent to
MACT compliance for chloroform. One commenter (IV-D2-4)
supported the EPA Office of Water's requirements to eliminate
hypochlorite bleaching through C102 substitution for all paper
grades where it is technically feasible.
4-50
-------
Several commenters (IV-D2-18, IV-D2-10, IV-D2-3, IV-D2-8)
agreed with the MACT floor level of control as outlined in the
March 8, 1996 supplemental notice. In response to EPA's
solicitation of comments on whether an alternative equivalent
numerical limit for chloroform is needed, the commenters asserted
that compliance with the BAT water standards will virtually
eliminate chloroform emissions and that a numerical limit is
neither needed nor desired.
One commenter (IV-D2-4) expressed concern over EPA's
decision to move away from using methanol as a surrogate for
organic HAP from bleaching processes and whether organic HAP' s
will be adequately controlled from the bleaching process. The
commenter (IV-D1-4) indicated that most of the organic HAP
emissions that would remain after the elimination of hypochlorite
originate from the use of dirty wash water and that organic HAP
emissions could be reduced by limiting the organic HAP content of
the wash water. The commenter urged EPA to clearly define how
the chosen approach will control the organic HAP and TRS
emissions from the bleaching process.
Response: In the March 8, 1996 supplemental notice, the
Agency revised the proposal for the bleaching system requirements
based on information and comments received after proposal. The
new data indicated that caustic scrubbing reduces emissions of
chlorinated HAP compounds (except chloroform), but does not
control non-chlorinated HAP emissions. The Agency determined
that no other option was feasible to control non-chlorinated
HAP'S. Although chloroform emissions could not reasonably be
reduced by use of one add-on air pollution control technology,
4-51
-------
chloroform emissions can be reduced using process modifications,
such as C102 substitution.
In the March 8, 1996 supplemental notice, the Agency
proposed to require chlorinated HAP emissions other than
chloroform to be controlled by 99 percent using a caustic
scrubber (with chlorine as a surrogate for chlorinated HAP) . As
an alternative to the percent reduction standard, the Agency also
proposed an emission limit of 10 ppmv HAP at the caustic scrubber
outlet (with chlorine as a surrogate for chlorinated HAP) . The
Agency also solicited comments on providing a mass emission limit
alternative to the percent reduction and the outlet concentration
standards.
Commenters on the March 8, 1996 supplemental notice
supported the changes to the scrubber requirements in the
proposed rule. Several commenters also supported development of
a mass emission limit alternative for the scrubber outlet. The
Agency evaluated data supplied by the commenters and data in
sampling tests. The results of the evaluation indicated that
sufficient data exist to develop an appropriate mass emission
limit (A-92-40, IV-B-29). Therefore, the final rule includes a
mass emission limit alternative of 0.001 kg of total chlorinated
HAP (not including chloroform) per Mg of ODP produced for
bleaching system scrubbers.
After proposal, the Agency also evaluated the effect of
process modifications on chloroform emissions. The results of
this analysis indicated that the technology basis for MACT
control of chloroform was 100 percent C102 substitution and
reduction of hypochlorite as a bleaching agent. These process
modifications were determined to reduce chloroform emissions
4-52
-------
significantly. The technology basis for BAT under the effluent
limitation guidelines and standards also require 100 percent
C102
substitution and elimination of hypochlorite. Since BAT and MACT
are essentially the same, EPA therefore proposed in the March 8
notice that chloroform emissions be controlled by complying with
the BAT requirements. No adverse comments were received to this
proposal.
In the March 8, 1996 supplemental notice, the Agency
solicited comments on whether an alternative numerical limit for
chloroform (i.e., besides complying with BAT) was needed. Some
commenters contended that a numerical limit for chloroform would
be unnecessary because the BAT requirements would achieve the
requisite reductions. The Agency did not receive any indication
of any benefit from a numerical limit for chloroform.
Additionally the Agency did not have sufficient data and did not
receive any further data after the March 8 notice to develop a
numerical limit. Therefore, the final rule does not include a
numerical limit for chloroform.
Consequently, EPA has concluded that the existing and new
source floor level of control for chlorinated HAP's is caustic
scrubbing with 100 percent ClC»2 substitution and elimination of
hypochlorite use. Compliance with the effluent guidelines BAT
option is at least as stringent as the MACT floor level of
control. Therefore, the final rule requires mills to comply with
the BAT requirements in the effluent limitation guidelines and
standards, or eliminate the use of hypochlorite and chlorine.
For non-chlorinated HAP's, the existing source and new source
floor is no control since no emission reduction from the current
4-53
-------
baseline for non-chlorinated HAP's is being achieved with the
bleach plant control technology.
Because MACT for new sources is equivalent to MACT for
existing sources, the new source MACT standards for bleaching
systems require compliance with BAT/Pretreatment Standards for
Existing Sources (PSES) requirements (or implementation of 100
percent substitution and elimination of hypochlorite). This
requirement applies even if the mill or bleaching system also
meets the definition of new source under the effluent guidelines
limitations and standards, and thus is required to meet the more
stringent new source effluent requirements of NSPS/Pretreatment
Standards for New Sources (PSNS) . Although the NSPS/PSNS may
require installation of technologies that reduce effluent loading
beyond what is achieved by 100 percent substitution and
elimination of hypochlorite, EPA is not aware that these advanced
technologies will provide air emission reductions beyond what the
BAT/PSES requirements will achieve.
The EPA evaluated the cost of going beyond the floor to
control non-chlorinated HAP's (see proposal preamble) using a
scrubber or a combustion device but determined that these options
had a substantial cost and environmental impact for limited
emission reduction. Therefore, EPA determined these options were
not warranted.
Comment: Three commenters (20,049A2, 20,080, 20,127) argued
that zero use of chlorine compounds should be the baseline
standard for MACT. In support of this position, one commenter
(20,080) cited the fact that there is at least one place in the
world where all grades of paper are produced using chlorine-free
technology. One commenter (20,122) argued that EPA should have
4-54
-------
evaluated the top 12 percent of mills in other countries, not
just the United States. The commenter (20,122) stated that
totally chlorine free (TCF) technologies at paper-grade kraft
mills in other countries were not reflected in the proposed rule.
The commenter (20,122) contended that EPA should revise the
proposed regulation and evaluate the appropriateness of TCF
technologies.
One commenter (20,059) argued that the references in the Act
to the elimination of emissions and its emphasis on elimination
of dioxin emissions should have led EPA to select totally
chlorine free technologies as the basis for the bleaching area
MACT standard. The commenter (20,059) provided technical and
legal support for their argument in their comments on the
proposed effluent guidelines. One commenter (20,129) contended
that in addition to TCF processes EPA should have also focused on
emerging technologies, such as biofiltration, to minimize toxic
HAP emissions.
Response: The EPA interprets the Act as requiring EPA to
establish the MACT floor level of control based on available data
from the source category to be regulated (best-performing
12 percent of the existing sources for which [EPA] has emissions
information) " . Control devices or technologies in use in other
countries may be evaluated for determining control options beyond
the MACT floor level of control.
The use of TCF technologies in the U.S. is limited
(currently only one mill has implemented the TCF process and this
mill does not produce a full array of products). Therefore, it
does not constitute the MACT floor level of control for existing
sources. The use of TCF technology has not been sufficiently
4-55
-------
demonstrated to produce the wide variety of U.S. pulp and paper
products for EPA to conclude that the technology can be used
widely. Therefore, EPA determined that it was inappropriate for
TCF technologies to be the new source MACT. A detailed
discussion of TCF is presented in the effluent guidelines portion
of the promulgation preamble (VLB).
The installation and operation of the TCF bleaching process
meets all the bleaching MACT standards for paper-grade bleaching
and would constitute compliance with the final rule. Therefore,
TCF bleaching is an alternative compliance option for the bleach
plant. Also, the effluent guidelines portion of the final rule
provides incentives for mills to adopt TCF technologies.
The EPA should defer requirements for dissolving grade
mills.
Comment: One commenter (IV-D2-17) suggested that a separate
MACT floor level of control for chloroform should be developed
for dissolving-grade pulp production. The commenter (IV-D2-17)
anticipated that the dissolving-grade MACT would be similar to
EPA's requirements for paper-grade production (i.e.,ClC>2
substitution and elimination of hypochlorite use in order to
control chloroform, alkaline scrubbing for chlorinated HAP). The
commenter (IV-D2-17) also anticipated that numerical emission
limits for chlorinated HAP emissions would not be applied. Other
commenters (IV-D2-15, 25,538) recommended that EPA defer
chloroform control requirements for dissolving-grade mills until
BAT is established for those mills.
Response: The EPA has concluded that MACT for chlorinated
HAP' s is caustic scrubbing and process modifications (100 percent
substitution and elimination of hypochlorite) for paper-grade
4-56
-------
mills. The effluent limitation guidelines and standards BAT
requirements are at least as stringent as the MACT requirements.
Therefore, the final rule requires bleach plants to control
emissions using a caustic scrubber and comply with the BAT
requirements or eliminate the use of chlorine and hypochlorite.
As stated in the July 15, 1996 Federal Register notice
(61 FR 36835), EPA is evaluating new data on the technical
feasibility of reducing hypochlorite usage and implementing high
clo
levels of 2 substitution on a range of dissolving-grade pulp
products. Therefore, EPA is deferring issuing effluent
limitation guidelines and standards for dissolving-grade mills
until the comments and data can be fully evaluated. The EPA
expects to promulgate final effluent limitation guidelines and
standards for dissolving-grade subcategories at a later date.
The EPA has decided to delay establishing these MACT
standards for chloroform and for other chlorinated HAP's for
dissolving-grade bleaching operations until promulgation of
effluent limitation guidelines and standards for those
operations, for the following reasons. With respect to the MACT
standard for chloroform, first, as explained above and in the
March 8, 1996 notice, the control technology basis for the
effluent limitation guidelines and standards and the MACT
requirements will be the same. Second, at present, the Agency is
unsure what level of chlorine substitution and hypochlorite use
is achievable for dissolving-grade mills. Thus, although EPA has
a reasonably good idea what the technology basis of MACT and
effluent limitation guidelines and standards is likely to be for
dissolving-grade mills, the precise level of the standards
remains to be determined. Consequently, at present, EPA is
4-57
-------
unable to establish what the MACT floor would be for chloroform
emissions from bleaching systems at these mills, and there is no
conceivable beyond-the-floor technology to consider. The EPA
will make these determinations based on data being developed, and
then promulgate for these mills effluent limitation guidelines
and standards and, concurrently, MACT standards based on those
effluent limitation guidelines and standards. Covered mills
would therefore be required to comply with the MACT standards
reflecting performance of the effluent limitation guidelines and
standards no later than 3 years after the effective date of those
standards, pursuant to Act section 112 (i) (3) (A) .
The basis for delaying MACT requirements for chlorinated
HAP's other than chloroform (again, from dissolving-grade bleach
operations only) differs somewhat. As noted above, the
technology basis for control of these HAP's is use of a caustic
scrubber. However, when plants substitute C1O2 for chlorine and
eliminate hypochlorite (in order to control chloroform emissions
and discharges to water, as explained above), a different
scrubber will be needed that can adequately control both the C102
emissions for worker safety reasons and the emissions of
chlorinated, non-chloroform HAP's. The Agency's concern (shared
by the commenters who addressed this question) is that immediate
control of the non-chloroform chlorinated HAP's could easily
result in plants having to install and then replace a caustic
scrubber system in a few years due to promulgation of effluent
limitation guidelines and standards and MACT requirements for
chloroform. This result is an inappropriate utilization of
scarce pollution control resources.
4-5S
-------
The EPA notes that an affected bleached paper-grade mill
must comply with the MACT requirements no later than 3 years from
publication in the Federal Register,, even if the mill's existing
CWA National Pollutant Discharge Elimination System
(NPDES) permit does not yet reflect the corresponding effluent
limitation guidelines and standards because its existing terms
have not expired or it has been administratively extended. Put
another way, even if a mill's existing NPDES permit serves as a
shield (until reissuance) against imposition of new limits based
on new effluent limitation guidelines (see CWA section 402(k)),
the MACT requirement for bleached paper-grade mills to control
chloroform emissions through compliance with all parameter
requirements in the effluent limitation guidelines and standards
takes effect to satisfy the requirements of the Act. Similarly,
if a bleached paper-grade mill's NPDES permit is reissued sooner
than the expiration of the 3-year compliance schedule authorized
for the chloroform MACT requirements and calls for immediate
compliance with the BAT limitations, that deadline would prevail.
The same principles will apply when effluent limitation
guidelines and MACT standards are promulgated for dissolving-
grade mills.
Incineration followed by scrubbing.
Comment: In the preamble to the proposed rule, EPA
requested comment or data on the use of combustion followed by
scrubbing to control emissions from the bleach plant. In
response to EPA's request, one commenter (20,027) claimed they
were not aware of any mill that used a combination of
incineration followed by scrubbing to control bleach plant
emissions; therefore, it should not be considered a floor level
4-59
-------
of control technology for bleaching vents. The commenter
(20,027) further stated that combustion of certain gas streams
followed by gas scrubbing of others would not be cost effective
as a beyond-the-floor level of control technology for bleaching
vents, based on a cost analysis presented in the submitted
comments.
Response: The EPA agrees with commenters that the
combination of incineration followed by scrubbing is a technology
that is more stringent than the MACT floor. The EPA determined
that combustion followed by scrubbing could be achieved at a
cost-effectiveness greater than $8,000/Mg HAP (see proposal
preamble) . The EPA determined that the costs were not reasonable
given the level of emission reduction and the additional
environmental impacts (increased water discharge and use) from
this option. Therefore, the technology was not adopted for MACT.
MACT floor for non-kraft mill bleach plants.
Comment: Two commenters on the proposed rule (20,053A1,
20,072) suggested alternative MACT floor levels of control for
the bleaching area for mills other than kraft. Based on a
project they undertook to understand methanol generation and
control at a soda mill, one commenter (20,072) indicated that the
MACT floor level of control for the bleaching component at soda
mills should be equivalent to the proposed rules with the
exception that the performance of the treatment device should be
based on 95 percent removal of chlorine andc]_Qo rather than
99 percent removal of total HAP.
Response: Based on current data, EPA has decided not to
create subcategories for the bleaching processes based on the
type of pulping technology. The differences in the bleaching
4-60
-------
processes between mills using different pulping technologies does
not appear to be great enough to warrant separate control
requirements.
The final rule requires mills to control chlorinated HAP' s
using caustic scrubbing and by meeting the effluent limitation
guidelines and standards BAT requirements or by eliminating
hypochlorite and chlorine use. Regarding one commenter's concern
about the bleach plant scrubber requirements, data reviewed by
EPA (A-92-40, II-I-24) show that bleach plant scrubbers in pulp
and paper mills achieve 99 percent control of chlorinated HAP's.
The commenter did not provide support for requiring a
lower percent reduction (95 percent). Therefore, the bleach
plant scrubbers are required to achieve 99 percent control of
chlorinated HAP's in the final rule (excluding chloroform).
4.3.1.4 Wastewater.
Steam strippers are not appropriate as floor level of
control technology.
Comment: Several commenters (20,027, 20,051, 20,054A2,
20,146) disagreed with the conclusion that steam stripping is a
floor level of control technology for process wastewater. The
commenters (20,011, 20,027, 20,051, 20,054A2) explained that
steam strippers currently in place were used for reducing odor
and BOD loadings to the biological treatment plant, and for
generating hot water for use in other process areas, but they
were not used for HAP or VOC emissions control. The commenters
(20,027, 20,051, 20,054A2) indicated that for these reasons,
steam strippers have never been calibrated or operated for the
type of continuous emission reduction that EPA has proposed.
However, one commenter (20,059) supported EPA's proposal of steam
4-61
-------
stripping as the floor level of control technology for wastewater
treatment. The commenter (20,059) cited the fact that the
American Paper Institute (API)/NCASI survey identified 31 mills
that currently use steam strippers to control emissions from
wastewater as support for this position.
Two commenters (20,027, 20,066A4) indicated that steam
stripping for the process wastewater component should be
considered as beyond-the-floor level of control and must be cost
justified. One commenter (20,146) stated that EPA vastly
underestimated the costs of steam stripper installation and
operation. According to one commenter (20,027), the cost-
effectiveness of controlling wastewater components using steam
stripping would not be reasonable based on EPA or industry cost
and emission reduction estimates (A-92-40, ll-B-20, ll-B-28,
ll-B-43, ll-C-10, 11-1-13, and 11-1-18).
Several commenters (20,027, 20,045, 20,066A4) claimed that
because methanol is the principal HAP and biological treatment
systems typically achieve greater than 90 percent reduction of
methanol, biological treatment should be the floor level of
control technology.
One commenter (20,027) stated that there are no sulfite
mills that currently use steam strippers. However, one commenter
(20,123A6) provided data on an existing steam stripper that is
used to recover S02 at a sulfite mill. Two commenters (20,027,
20,076) declared that there are no sulfite mills that currently
capture all emissions from process wastewater collection and
treatment. One commenter (20,027) submitted that because of
this, the MACT floor level of control for the process wastewater
component at sulfite mills should be no control. Other
4-62
-------
commenters (20,045, 20,076) indicated that biological treatment
should be the appropriate MACT floor control for sulfite mills.
Two commenters (IV-D2-15, IV-D2-16) agreed with EPA's conclusion
that the floor level of control for sulfite wastewater emissions
is no control.
One commenter on the March 8, 1996 supplemental notice
(IV-D2-7) agreed with EPA's conclusion that MACT for
semi-chemical wastewater is no control. Another commenter
(IV-D2-15) agreed with EPA's conclusion that MACT for bleaching
wastewater is no control.
Response: The March 8, 1996 supplemental notice presented a
detailed discussion of revisions made to the steam stripping
requirements since proposal. The EPA concluded that steam
stripping is the floor level of control for kraft wastewater
streams. This conclusion was based on information collected in
the 1992 voluntary MACT survey and other industry data submitted
after proposal (A-92-40, IV-B-10, IV-D1-3, IV-D1-82, IV-D1-91,
and IV-J-32) . Based on this information, EPA determined that
greater than 20 percent of kraft mills practice steam stripping.
Therefore, the MACT floor level of control for kraft wastewater
is steam stripping.
For sulfite, semi-chemical and soda mills and bleaching
processes, EPA has determined that process wastewater is not
controlled at the floor level of control. Steam stripping is not
required as a beyond-the-floor option due to the high cost
required for a limited emission reduction.
With regard to the comment that steam strippers were not
installed for HAP or VOC emissions reduction, the Act requires
that EPA determine MACT based on the best-performing facilities
4-63
-------
(i.e., the facilities with the lowest achievable emissions rate;
see discussion under section 4.1, Statutory Interpretation). The
specific technologies in place at a best performing mill or the
reasons for their existence are not relevant to the MACT
determination.
The commenters' concern that steam strippers have never been
calibrated or operated for the type of continuous emission
reduction proposed by EPA is addressed in section 4.3.4 on
downtime.
Enclosure of wastewater streams should not be in the floor
level of control.
Comment: Several commenters (20,027, 20,018A1, 20,051,
20,054A2, 20,146) stated that EPA failed to demonstrate that any
sources, much less 12 percent of sources, practice universal
enclosure of all wastewater streams. One commenter (20,027) also
stated that the use of covers in the industry was not
sufficiently widespread enough to be considered a floor level of
control technology; therefore, universal enclosure should be
considered a beyond-the-floor option. In addition, the commenter
(20,027) indicated that EPA did not provide any record of the
costs or benefits associated with enclosing wastewater streams.
Two commenters (20,027, 20,146) reasoned that since surface
impoundments and clarifiers emit very small amounts of HAP's, the
requirements to cover and control emissions from these units are
totally unwarranted and should be removed from the final
regulation.
One commenter (IV-D2-7) noted that covering and venting to a
control system all the equipment and tanks associated with
wastewater treatment would be costly, impractical, and completely
4-64
-------
unreflective of current industry practice. The commenter
(IV-D2-7) believes that covering these units does not represent a
floor level of control practice and that EPA has not provided a
cost-effectiveness analysis to substantiate a "beyond-the-floor"
level of control.
Response: As discussed previously in this section, EPA
determined that steam stripping is the floor level of control for
kraft mill condensates (A-92-40, IV-B-8). The EPA determined
that a well-operated biological treatment system can achieve
equivalent control if the wastewater conveyance system is
enclosed to prevent volatilization of HAP's from the wastewater
(A-92-40, IV-D1-75). At proposal, the covering and enclosure
requirements were set forth for mills that wanted to use their
existing sewer system to convey the wastewater to the biological
treatment system. Failure to enclose conveying pipes and
trenches would vitiate the rules effectiveness, since volatile
HAP's would be released by the time wastewater reaches the
biological treatment unit. See 56 FR at 33495, 33530 (necessity
of controlling volatile wastes at the point of generation) . The
final rule requires that mills choosing to use biological
treatment must hardpipe the effluent to the treatment unit using
a condensate collection system meeting the individual drain
system requirements specified in subpart RR §§ 63.960, 63.961,
63.962, and 63.964.
4.3.2 MACT Floor Control Applicability
4.3.2.1 Named Stream Approach Versus Applicability Cutoffs.
Comment: Many commenters (20,027, 20,054A2, 20,056, 20,057,
20,057A2, 20,059, 20,070A1, 20,074, 20,118, 20,146) disagreed
with the levels chosen for the applicability criteria in the
4-65
-------
December 1993 proposal for the pulping, bleaching, and process
wastewater components. Most of the commenters (20,027, 20,054A2,
20,056, 20,057, 20,057A2, 20,070A1, 20,074, 20,118) objected that
the de minimis levels for the applicability criteria were too
low, did not represent the MACT floor level of control, and would
result in sources being subject to the MACT standards beyond the
floor level of control. One commenter (20,059) indicated the
applicability level was set too high. Three of the commenters
(20,027, 20,059, 20,146) objected that EPA lacked sufficient data
and quantitative information to support the numerical values or
cutoffs. One commenter (20,010) supported the proposal as set
forth because it required the control of all significant emission
points from the bleaching process.
One commenter (20,059) contended that EPA did not evaluate a
range of applicability levels, nor did it assess the
environmental implications of the proposed cutoff levels, or any
alternatives. The commenter (20,059) was concerned that some
LVHC concentration vents that should be controlled would not be.
The commenter (20,059) suggested a sliding scale combination of
flow and concentration to determine applicability rather than
excusing a vent because either flow or concentration were low.
Several commenters (20,027, 20,054A2, 20,056, 20,057A2, 20,066A4,
20,118) suggested specific levels for de minimis flow and
concentration rates.
Several commenters (20,027, 20,056, 20,066A4) proposed
approaches for re-determining the floor level of control
applicability levels. One commenter (20,027) strongly
recommended that EPA use the control and stream characterization
information submitted by industry to set applicability levels
4-66
-------
(A-92-40, II-B-20, II-B-21, II-F-27). Two commenters (20,027,
20,066A4) suggested that EPA establish de minimis flow and
emission rates for pulping and bleaching process vents using the
data from deckers and screens, sources for which the proposal
determined to have no control at the floor level of control. One
commenter (20,027) suggested limiting the use of de minimis
criteria to the floor level of control emission points, revising
the criteria to annual averages, and allowing for engineering
evaluations to determine source applicability. The commenter
(20,027) also suggested using a total resource effectiveness
(TRE) equation to determine applicability as cited in the HON.
The commenter's (20,027) rationale was that a TRE brings in a
third parameter (i.e., cost-effectiveness) in determining
applicability.
Several commenters (20,027, 20,046A2, 20,059, 20,070A1,
20,071, 20,074, IV-D2-15) supported naming the emission points in
the pulping component that must be controlled. One commenter
(IV-D2-15) noted that this approach would make the rule easier
for the regulated community to understand and implement,
eliminate the need for widespread testing to determine
applicability, and would guarantee the treatment of those streams
with significant HAP concentrations. The commenter (IV-D2-15)
also stated that this approach will simplify preparation, review,
and enforcement of permits for pulp and paper mills, as well as
result in significant reductions in implementation costs to mills
while resulting in emission reductions equivalent to those
proposed. One commenter (20,056) stated the only process
equipment that should have been considered for control in the
floor was as follows: (1) digester or NCG system; (2) digester
4-67
-------
relief system; (3) evaporator NCG and hotwell gases; (4) oxygen
delignification unit (blow gases and washer); (5) foam breaker or
filtrate tanks; and (6) weak black liquor storage tanks. The
commenter (20,056) indicated that dewatering devices other than
deckers should also be exempt from control.
One commenter (20,043) explained that for quality assurance
purposes, samples are collected throughout the processing of pulp
and paper and requested that the sample pots and their associated
air and water emissions be considered de minimis in the final
rule.
Response: At proposal, EPA had limited data to characterize
some of the smaller emission points and condensate streams within
the pulping component. In the absence of more specific data, the
applicability values were identified as a way to distinguish
between the emission points that were controlled at the floor
level of control and those that were not. Since proposal,
industry has submitted additional data (A-92-40, IV-D1-29,
IV-Dl-29a, IV-D1-31, IV-D1-33, IV-D1-34, IV-D1-35, IV-D1-38,
IV-D1-39, IV-D1-41, and IV-D1-41). This new information allowed
EPA to identify which vent and condensate streams are actually
controlled at the floor level of control. The EPA reanalyzed the
floor level of control based on these designations. The EPA then
revised the format of the proposed rule to account for the new
data and the results of the floor level of control analysis. The
format for the final rule names specific streams to be
controlled. The EPA also determined that applicability values
were appropriate for decker, knotter, and screen systems. The
EPA is not regulating all decker, knotter and screen systems
because control of all these streams are not in the floor level
-------
of control, only the high emitting ones. No further control is
justified. Commenters are referred to section 4.3.1.2 for a
discussion of the applicability determinations for decker,
knotter, and screen systems.
The different approach used in the final rule does not
significantly change the number of emission points controlled
from those intended to be controlled in the proposed rule. The
emission points and condensate streams that are being controlled
in the final rule are fundamentally the same emission sources
that EPA intended to be controlled in the proposed rule. The EPA
concluded that the revised approach is easier and less costly to
implement, for both the affected industry and the enforcement
officials, since extensive emission source testing is not
required to identify the vent and condensate streams to be
controlled.
MACT floor level of control applicability for condensate
streams.
Comment: Regarding the December 17, 1993 proposal, one
commenter (20,027) declared that the applicability format
selected for wastewater streams presumed that all streams in the
mill (except the bleach plant acid and caustic sewers) would
require control unless they are tested to prove they are below
the cutoff thresholds. The commenter (20,027) stressed that this
format would require unnecessary testing and evaluation in areas
of the mill where control is not warranted.
Two commenters (20,027, 20,056) suggested that EPA specify
by name or class which wastewater streams are exempt and which
streams must be sampled to prove they are de minimis. One
4-69
-------
commenter (20,027) suggested specific format changes for the
process wastewater area.
One commenter (IV-D2-3) requested that the rule specifically
allow stripped condensates or condensates not listed for control
to be reused at any location in the mill or be sewered without
additional control requirements.
Regarding EPA's approach of naming condensate streams
subject to the MACT control applicability in the March 8, 1996
supplemental notice, one commenter (IV-D2-15) agreed with EPA's
decision to name the specific pulping process wastewater streams
that will be subject to control. The commenter (IV-D2-15) stated
that this approach eliminates the need for an expensive
open-ended sampling program to show what streams did not meet the
concentration and flowrate applicability criteria. The commenter
(IV-D2-15) stated that the control of the named streams will
treat more HAP-containing water than is currently being treated;
therefore it will provide a level of control beyond the floor.
While the commenter (IV-D2-15) did not take exception to this
particular requirement, the commenter felt it should be noted.
The commenter (IV-D2-15) added that the rule should also state
explicitly that the treated condensates should be available for
reuse throughout the mill without any further restrictions.
One commenter (IV-D2-14) disagreed with EPA's decision in
the March 8, 1996 supplemental notice to define pulping process
wastewater streams requiring control as those achieving a
65 percent or greater methanol recovery. The commenter
(IV-D2-14) noted that evaporator or condenser systems that do not
currently achieve 65 percent methanol recovery cannot simply
readjust the internal configuration of the equipment and that
4-70
-------
improved recovery would require extensive modification or
replacement of equipment. The commenter (IV-D2-14) also
suggested that newly installed systems may or may not be able to
achieve the 65 percent methanol level and the commenter proposed
to redefine pulping process wastewater streams to be controlled
as those with a minimum of 50 percent methanol recovery.
Two commenters (IV-D2-3, IV-D2-20) agreed that pulping
process wastewater to be controlled should contain 65 percent of
the methanol present in the vapor from the first weak liquor feed
stage(s). The commenters (IV-D2-3, IV-D2-20) suggested that
because they aggregate pulping process wastewater on more than
the weak liquor feed stage(s), the definition needs to be
clarified or else the mill will have to cease collection of
methanol from evaporator stages other than weak liquor feed
stage(s). The commenters (IV-D2-3, IV-D2-20) recommended that
the rule make it clear that the 65 percent requirement applies to
the system as a whole and not to every individual evaporator
stage where condensate segregation is practiced.
Response: In the final rule, EPA has decided to retain the
approach of requiring named streams to be controlled. This
approach is the most efficient method of specifying applicability
of the rule and will eliminate unnecessary testing and compliance
burden on the affected industry. The final rule requires that
the entire volume of the named streams must meet MACT, expressed
as one of several treatment options.
However, the rule includes an option for reducing the volume
of condensate to be treated from specified streams. Most mills
currently practice some degree of condensate segregation on the
pulping process wastewater streams. Condensate segregation is
4-71
-------
the practice of generating, producing, or isolating a high-HAP
concentration/low flow rate condensate stream from process vent
vapors or gases in order to maximize the HAP mass and minimize
the condensate volume sent to subsequent treatment. If a mill
utilizes condensate segregation to produce the pulping process
condensate streams, only the high-HAP fraction stream must be
treated according to the options specified in the standards. If
condensate segregation is not practiced, the entire volume of the
pulping process condensate stream must be treated.
Based on the information obtained in the 1992 voluntary
NCASI survey, the floor level of control for kraft pulping
process condensates is 92 percent removal of total HAP (based on
the performance of steam stripping) from the high-HAP fraction
condensates from the digester, turpentine recovery, and
evaporator systems. However, no standard definition (e.g., HAP
concentration, flow rate, mass, etc.) exists for designating the
high-HAP fraction condensate streams from these systems.
Consequently, EPA developed the percent mass split criteria for
designating the high-HAP fraction condensate streams. As
discussed in the March 8, 1996 Federal Register supplemental
notice (61 FR 9390), EPA determined that condensate segregation
can generate a high-HAP fraction stream containing 65 percent of
the overall HAP mass present in the process vapor stream.
The 65 percent mass split was developed based on information
provided by industry during a meeting with EPA (A-92-40,
IV-E-15). The information contained example mass balances for
digester, turpentine recovery , and evaporator system condensates
before and after condensate segregation was implemented. The
before-and-after mass balances were used to estimate the typical
4-72
-------
mass split found in the high-HAP fraction condensate streams that
was achievable using segregation.
In their comments on the March 8, 1996 Federal Register
supplemental notice and in additional correspondence to EPA
(A-92-4, IV-D1-97), industry indicated their support for the
definition of the high-HAP fraction condensate stream. However
in correspondence to EPA regarding suggestions to the MACT
standard definitions (A-92-40, IV-D1-107), industry indicated
that a high-HAP fraction condensate stream should be designated
by a 50 percent mass split (instead of 65 percent). This
revision was necessary, according to industry, since some of the
mills with the best-performing steam stripper systems could not
meet the 65 percent mass split, even though they were sending a
large amount of HAP mass to the stripper system for treatment.
To resolve the discrepancy between the EPA and industry
percent mass split designations, additional data were evaluated
to confirm the percent mass split values presented in the
March 8, 1996 notice (A-92-40, IV-B-24). Based on the
evaluation, EPA disagrees with the percent mass split recommended
by industry (50 percent) , and has decided to keep the
65/35 percent mass split.
In their correspondence, industry also suggested that an
additional option be added to the MACT standard that would allow
for either the percent mass split to be achieved or for a minimum
HAP mass be sent to treatment. Some commenters also indicated
that they would not be able to achieve the 65/35 percent mass
split without extensive and costly modifications to their
existing equipment. Based on the analysis presented in this
memorandum, EPA believes that achieving the minimum mass
4-73
-------
requirements would also achieve their intent of controlling only
the low volume, high-HAP fraction condensate streams.
The minimum mass requirements were based on the steam
stripper performance requirements (i.e., percent and mass
removal) developed for the MACT standard. The development of the
steam stripper performance requirements is presented in a
separate memorandum (A-92-40, IV-B-10). For both bleached and
unbleached mills, the final rule requires 92 percent removal of
HAP. For bleached mills, the mass removal requirement is
5.5 kilograms or more of total HAP (measured as methanol) per
megagram of oven-dried pulp (kg HAP/Mg ODP) ; for unbleached
mills, the mass removal requirement is 3.6 kg HAP/Mg ODP
(measured as methanol) . The minimum mass requirements for each
type of mill (bleached and unbleached) were obtained by dividing
the required mass removal by the required percent removal. For
example, the mass removal required for bleached mills (5.1 kg
HAP/Mg ODP) divided by the percent removal (92 percent) yields a
minimum mass removal of 5.5 kg HAP/Mg ODP. The minimum mass
removal for unbleached mills (3.6 kg HAP/Mg ODP) was obtained
using the same procedure.
MACT floor level of control should be applicable to chlorine
dioxide preparation equipment.
Comment: One commenter (20,110) asserted that the proposed
rule did not address C1O2 preparation equipment. Two commenters
(20,091, 20,110) asserted that the rule should require control of
C102 preparation equipment and emission points. One commenter
(20,091) contended that C102 is more toxic than chlorine and the
generation of ClC»2 is likely to increase in the future as the use
of ClC>2 substitution becomes more prevalent in the industry.
4-74
-------
Response: The final rule does not require control of C102
generation equipment. No information has been submitted to the
Agency to suggest that C1C>2 generation is a significant source of
HAP emissions (C102 is not a listed HAP) or that controls exist
at the floor level of control. Based on an engineering review of
process flow diagrams of C1O2 generation processes supplied by
commenters, EPA has concluded that these processes are
essentially closed processes without significant atmospheric
vents. Facilities storing over 1,000 pounds of C102 would be
subject to the 112 (r) requirement of an approved accident
prevention and response plan.
Using applicability levels.
Comment: One commenter (20,059) argued that the use of
cutoffs or applicability levels based on cost-effectiveness
considerations would be illegal in light of Congress' rejection
of cost-benefit and cost-effectiveness as a basis for setting
MACT standards. The commenter (20,059) suggested that EPA
eliminate the cutoff applicability criteria that exempt emission
points from control.
One commenter (20,114) contended that the cutoff level of
500 ppmv for capture and incineration of vent gases contained in
the proposal should be increased. The commenter (20,114) argued
that the 500 ppmv cutoff was apparently developed using data from
the synthetic organic chemical industry. The commenter (20,114)
argued that application of data from a different industry must be
justified by EPA.
Response: In the proposed rule, the applicability criteria
were not chosen on the basis of cost-effectiveness but rather to
delineate between vents and wastewater streams that are
4-75
-------
controlled at the MACT floor level of control. Section 112(d)(2)
of the Act requires the Administrator to establish standards
based on the maximum degree of reduction in emissions of HAP' s
"taking into consideration the cost of achieving such emission
reduction. . ." This mandate was followed by EPA and the
standard was set at the floor. Applicability cutoffs were used
to distinguish between vents that were and were not controlled at
the floor level of control. Regarding the comments on the
500 ppm cutoff, the commenter appears to be confused with regard
to the applicability cutoffs specified at proposal. The 500 ppmw
(not volume) cutoff in the proposed standard applied to process
wastewater streams. Regarding the commenter's concern about
applying data from different industries, EPA interprets this
comment to address the 20 ppmv outlet concentration specified for
incinerators used to comply with the pulping process standards.
The EPA has concluded that the outlet concentration is achievable
for well designed and operated incinerators (A-92-40, IV-B-19).
Since proposal, industry submitted additional data (A-92-40,
IV-D1-29, IV-Dl-29a, IV-D1-31, IV-D1-33, IV-D1-34, IV-D1-35,
IV-D1-38, IV-D1-39, and IV-D1-41) that was used by EPA to revise
the format of the proposed rule. The format for the final rule
names specific streams to be controlled. The EPA also determined
that applicability values were appropriate for knotter and screen
systems. A more detailed discussion of this issue is presented
in the March 8, 1996 supplemental notice. Commenters are referred
to section 4.3.1.2 for a discussion of the applicability
determinations for knotter and screen systems.
MACT floor level of control applicability for sulfite mills.
Comment: One commenter (20,027) reasoned that the MACT
4-76
-------
floor level of control at sulfite mills should exclude the
control of nonhalogenated HAP's from hot caustic extraction
stages, and the control of digester or evaporator condensates
because sulfite mills do not control these processes.
Response: The EPA agrees with the commenter. As discussed
in the March 8, 1996 supplemental notice, the available data
supports the establishment of separate emission standards for
bleaching at paper-grade and dissolving-grade pulping processes
but not on the type of pulping process (e.g., kraft, soda,
sulfite, or semi-chemical pulping). The data also indicated that
bleach plant scrubbers are ineffective at removing
non-chlorinated HAP's. In the final rule, sulfite mills are
required to comply with the respective bleaching standards for
paper-grade or dissolving-grade processes, which set requirements
for chloroform and other chlorinated HAP's.
Regarding the control of digester or evaporator condensates,
EPA concurs that no existing sulfite mills control these streams.
Therefore, control of these streams is not included in the MACT
floor level of control. Considering cost and impacts, the EPA
considers the option to steam strip these streams beyond the
floor level of control, to be unreasonable.
Comment: One commenter (IV-D2-14) asked that the identity
of the specific vents to be included in the mill systems be more
explicitly stated. The commenter (IV-D2-14) provided a list of
sulfite mill vents proposed to be included in digester,
evaporator, and redstock washer systems.
Response: The EPA appreciates the commenter's support for
requiring specific named streams to be controlled. The EPA
evaluated the types of equipment controlled at existing sulfite
4-77
-------
mills in developing the MACT floor level of control and MACT
requirements (A-92-40, IV-B-8). The final rule requires existing
sulfite sources to control digester systems, evaporator systems,
and pulp washing systems. New sulfite sources are required to
control the same equipment as existing sources plus weak liquor
storage tanks, strong liquor storage tanks, and acid condensate
storage tanks. The emissions from these named systems must be
collected in a closed-vent system and routed to a control device.
Both the total methanol emissions from these named systems and
condensate streams from equipment used to reduce methanol
emissions at sodium- and calcium-based sulfite processes are to
meet an emission limit of 0.89 Ib/ODTP or are to be reduced by
92 percent. Similarly, both the total methanol emissions from
magnesium or ammonium-based sulfite processes are to meet an
emission limit of 2.2 Ib/ODTP or are to be reduced by 87 percent.
Wastewater reuse.
Comment: One commenter (20,057A2) stated that EPA
improperly determined the floor level of control for brownstock
washers to be collection and incineration of vent gases. The
commenter (20,057A2) asserted that if condensates that are
recycled to the washer are required to be treated, then the
emissions from the washer will be reduced when the cleaner
condensate is used. According to the commenter (20,057A2),
requiring treatment of condensates and collection and
incineration of brownstock washer vents is tantamount to going
above the floor level of control. The commenter (20,057A2)
indicated that EPA should perform a cost analysis for going
beyond the floor.
4-7S
-------
One commenter (20,057A2) recommended that wastewater
emission reductions be measured on a "mill-wide" basis because of
the complex processes of recycle and reuse found throughout the
mill.
Response: While EPA agrees with the commenter that reducing
the HAP concentration of shower water will reduce atmospheric
emissions from brownstock washers, EPA does not agree that steam
stripping is a beyond the floor level of control option. Based
on data submitted after proposal (A-92-40, IV-J-32), EPA
concluded that the streams that are typically recycled to
brownstock washers are not the same as the named streams that are
required to be treated in the steam stripper by the final rule.
Additionally, the final rule includes a control option allowing
mills to recycle the named condensate streams, without subsequent
treatment, to a controlled piece of process equipment. Since the
final rule requires pulp washers to be controlled, condensate
streams recycled to this piece of equipment are not required to
be treated.
Comment: One commenter (20,027) commented critically that
the prohibition of wastewater stream dilution would require
wholesale repiping of established process water flow patterns in
the industry in order to avoid the impermissible "dilution."
Response: The part 63 general provisions prohibit sources
from circumventing the control requirements of the part 63
standards. The general provisions specifically prohibit
circumventing standards by dilution. Therefore, EPA does not
consider it necessary to include similar requirements in the
rule, and the final rule does not include this language.
4-79
-------
4.3.3 Beyond the MACT Floor Level of Control.
4.3.3.1 MACT Set Beyond the MACT Floor Level of Control.
Comment: Two commenters (20,027, 20,146) stated that while
the Act allows EPA, in certain cases, to set MACT beyond floor
levels of control, EPA did not establish a foundation for such a
decision for the proposed standards. (Case law cited: Portland
Cement Assoc. v. Ruckelhaus, National Lime Assoc. v. EPA, and
Sierra Club v. Costle; A-92-40, II-C-10.)
Response: Based on information available at the time of
proposal and the statutory interpretation of the MACT floor level
of control, EPA did not propose requirements beyond the MACT
floor level of control. The EPA agrees with the commenter that
if controls beyond the level of the MACT floor level of control
were proposed, then they must be supported by sufficient
information on the balance of costs, energy, and environmental
impacts.
In the final rule, the only MACT requirements for existing
sources that are beyond the MACT floor level of control are for
soda pulping processes. Data available to EPA from the 1992
voluntary MACT survey and information received after proposal
indicate that soda mills do not currently control any of the
equipment that is subject to the MACT requirements for kraft
mills (A-92-40, IV-B-8) . However, EPA has determined that the
emissions from soda mills are similar to kraft mills and that the
control costs are similar to stand-alone semi-chemical mills.
Therefore, EPA considers going beyond the MACT floor level to
control LVHC vent emissions at soda mills to be an appropriate
level of control for MACT for these mills, taking into
consideration the costs of achieving the controls as well as the
4-80
-------
other factors, such as energy and environmental impacts (A-92-40,
IV-B-12).
4.3.4 Downtime and Back-up Control Technologies
Comment: Several commenters (20,027, 20,043, 20,054A2,
20,057, 20,066A3, 20,146) indicated that downtime of equipment is
part of the natural variability of operation, and should have
been considered by EPA when determining the MACT floor. One
commenter (20,027) further stated that continuous compliance
without downtime, as required by the proposed standards, would be
beyond the MACT floor level of control since no mill currently
operates with this type of continuous compliance. One commenter
(20,102) contended that based on past experiences of controlling
NCG and brownstock washer gases in power boilers and lime kilns,
existing control devices may not even be capable of providing
continuous compliance.
One commenter (20,027) indicated that if no allowance for
excess emissions is provided in the final rule, EPA must conduct
a cost-effectiveness analysis for the use of backup control
devices, since these control devices would be needed and the
costs and secondary impacts of backup control devices for
combustion sources and steam strippers were not addressed in the
proposed standards. Two commenters (20,027, 20,115A2) concluded
that because the downtime occurrence (and therefore the emission
reduction) would be small and the costs for backup control
devices would be large, the use of these backup control systems
cannot be justified as cost effective as a beyond-the-MACT-floor
level of control.
One commenter (IV-D1-15) stated that very few mills have
HVLC controls and only a small percentage of those mills have
4-81
-------
backup controls; therefore, backup controls are not part of the
MACT floor level of control for HVLC systems. One commenter
(20,149) indicated that existing combustion devices are down
enough that it is reasonable to expect a backup device. One
commenter (20,150) requested that new backup controls be required
by all mills using the C1O2 bleaching process. One commenter
(20,110) stated that the proposed rule should be amended to
require backup incineration devices. The commenter (20,110)
indicated that backup emissions controls are already in place in
a portion of the industry.
Several commenters (20,027, 20,057A2) disputed whether the
general provisions to part 63 would cover maintenance and
troubleshooting downtime, since the industry and regulatory
officials do not generally consider these events malfunctions.
Another commenter (20,054A2) stated that EPA had assumed that the
startup, shutdown, and malfunction allowance would cover those
events that resulted in venting of LVHC and HVLC gases from
closed-vent systems. The commenter (20,054A2) stated that not
all of the maintenance downtime associated with lime kilns and
power boilers necessarily cause a shutdown in pulp mill operation
since facilities can continue to operate the pulp mill at various
rates depending on liquor inventories and chemical make up
systems. One commenter (IV-D2-2) noted that any time that both
the mill and the NCG system are down should not be counted toward
downtime. Two commenters (20,066A3, 20,146) requested that EPA
specifically identify in the final rule which types of startup,
shutdown, and malfunction events will not require compliance with
the air emission standards. Another commenter (20,059) contended
that the general provisions were too lenient in allowing the
4-82
-------
emissions associated upsets, startup, shutdown, and maintenance
and urged EPA to close this loophole in the regulation.
Two commenters (20,027, 20,146) indicated that excess
venting is an essential safety practice, and that it would occur
even with transfer to a backup control device. One commenter
(IV-D2-11) suggested that by-pass allowances are needed for
unavoidable and safety venting events that are sometimes
difficult to define in advance. Another commenter (20,151)
indicated that uncontrolled pulping emission from bypass
collection systems should be controlled.
Several commenters (20,027, 20,043, 20,054A2, 20,057,
20,057A2, 20,066A4, 20,070A1, 20,118, IV-D2-2) presented
suggestions as well as estimated and measured downtime for
certain processes and equipment at their facilities. Several
commenters (20,027, 20,054A2, IV-D1-15, IV-D2-15) requested a
venting allowance for the pulping component standards ranging
from 2 to 4 percent outside of startup, shutdown, and malfunction
provisions in the general provisions; this would be similar to
the allowance contained in the pulp and paper NSPS. One
commenter (20,027) indicated that backup control devices would
not be needed to comply with the rule if a 4 percent allowance is
included.
One commenter (20,054A2) asserted that LVHC gas flows cannot
be automatically diverted to backup devices due to explosion
hazards. The commenter (20,054A2) stated that: (1) the burners
used in the backup devices must go through startup checks that
may result in venting for 15 to 30 minutes per episode; (2) the
frequency for diverting to a backup device varies from 1 to
10 events in a quarter (e.g., 15 to 300 minutes of venting); and
4-83
-------
(3) the operation of backup devices are checked by most mills at
least once per month. Another commenter (IV-D2-11) recommended
incorporating by-pass allowances of 2 percent for LVHC control
systems.
One commenter (20,054A2) stated that their current primary
source of combustion for LVHC gases was the lime kiln. The
commenter (20,054A2) indicated that: (1) lime kilns typically
require approximately 1.5 to 2 percent of the available annual
operating hours for rebricking; and (2) operating variabilities
(such as flame outs, ring formation, and problems with the
product removal system and wet end processes) can result in
additional downtime of 0.5 to 1 percent of the available annual
operating hours. The commenter (20,054A2) further asserted that
a minimum of one maintenance shut down per year (1 to 1.5 percent
of the available annual operating hours) is required by the mill
and that operating variables (such as variable steam load, fuel
feed system problems) can result in venting of LVHC gases from
0.5 to 1 percent of the available annual operating hours.
One commenter (IV-D1-15) stated the HVLC streams are vented
to the atmosphere during boiler or recovery furnace downtime,
which is normally down about 10 percent of the time the mill is
in operation. One commenter (IV-D2-2) suggested that a downtime
of 10 percent for HVLC systems is warranted, as these systems are
typically single line/single combustion point systems as opposed
to LVHC control systems. Another commenter (IV-D2-11)
recommended incorporating a by-pass allowance of 5 percent for
HVLC control systems.
4-84
-------
Two commenters (20,027, 20,043) recommended an allowance of
downtime for excess emissions from bleach plant scrubbers of
approximately 2 percent.
One commenter (20,011) stated that the need for backup
control devices to account for steam stripper downtime or
biological treatment system upsets has not been addressed by EPA.
One commenter (20,071) stated that all mills, even those with
backup control devices, will vent steam stripper overheads for
some percentage of the time. Another commenter (IV-D2-3)
suggested an allowance of 5 percent of the operating year for
steam stripper downtime. One commenter (IV-D2-2) suggested a
downtime of 5 percent for more reliable stand-alone stripping
systems but a downtime of 10 percent for integrated stripping
systems. One commenter (IV-D2-3) noted that downtime should only
be considered those periods when condensates are unable to be
treated and must be sewered. The commenter (IV-D2-3) also
requested that the rule explicitly state that sewering during
periods of steam stripper downtime is acceptable. Another
commenter (IV-D1-15) also indicated that the rule should take
into consideration steam stripper downtime and that mills
currently route those streams to the sewer during periods when
the stripper is not functioning and the stripper feed tanks are
full.
One commenter (20,054A2) stated that the industry
recommendation for a measure of continuous compliance is venting
time for a closed-vent system with an allowance for short term
venting which occurs due to inherent process variability. The
commenter (20,054A2) provided information on how several States
regulate continuous compliance, and indicated that the kraft pulp
4-85
-------
mill NSPS (subpart BB, 40 CFR part 60) allows two TRS exceedances
per quarter excluding startup, shutdown, and malfunctions. One
commenter (IV-D2-15) stated that excusable excursions are also
need for parameter monitoring and should be determined on an
annual basis as a percentage of the time that the process is in
operation.
Response: Since proposal, EPA has re-evaluated the need to
incorporate downtime or excess emissions allowances for LVHC,
HVLC, and steam stripper systems into the final rule. Based on
the information collected in the 1992 voluntary MACT survey, EPA
concluded that some allowance for excess emissions is part of the
MACT floor level of control. For the final rule, EPA established
excess emissions allowances to approximate the level of downtime
and backup control at the best performing mills and the
associated period of time which no control device is available
(A-92-40, IV-E-83). The excess emissions allowances in the final
rule include periods when the control device is inoperable and
when the operating parameter values established during the
initial performance test are not maintained at the appropriate
level.
Based on an analysis of the public comments and the
available data regarding excess emissions and the level of backup
control in the industry, EPA has determined that an appropriate
excess emissions allowance for LVHC systems would be 1 percent of
the operating hours on a semi-annual basis for the control
devices used to reduce HAP emissions. The best-performing mills
achieve a 1 percent downtime in their LVHC system control
devices. For control devices used to reduce emissions from HVLC
systems, EPA has concluded that an appropriate excess emissions
4-86
-------
allowance would be 4 percent. The best-performing mills achieve
a 4 percent downtime in the control devices used to reduce
emissions from their HVLC system to account for flow balancing
problems and unpredictable pressure changes inherent in HVLC
systems. For control devices used to control emissions from both
LVHC and HVLC systems, the Agency has determined that a 4 percent
excess emissions allowance is appropriate. This decision was
made because the control device would be used for the HVLC
system, which has the higher emissions allowance. For LVHC and
HVLC system control devices, the excess emissions allowances do
not include scheduled maintenance activities that are discussed
in the part 63 general provisions. The allowances address normal
operating variations in the LVHC and HVLC system control devices
for which the equipment is designed. The variations would not be
considered startup, shutdown, or malfunction under the part 63
general provisions (Air Docket A-92-40, IV-D1-103, IV-D1-110,
IV-D1-115, IV-E-83, and IV-E-85).
Although industry commenters suggested a downtime allowance
of excess emissions from bleach plant scrubbers of approximately
2 percent, no data were provided to support their suggestion.
The commenters did not address bleach plant scrubbers in their
recommendation for control device downtimes in subsequent data
submittals. Therefore, the final rule does not include downtime
allowances for bleach plant scrubbers.
The Agency determined the appropriate excess emissions
allowance for stand-alone and integrated steam stripper systems
to be 10 percent. The allowance accounts for stripper tray
damage or plugging, efficiency losses in the stripper due to
contamination of condensate with fiber or black liquor, steam
4-87
-------
supply downtime and condition control device downtime. This
downtime allowance includes all periods when the stripper systems
are inoperable including scheduled maintenance and malfunctions,
startup, and shutdowns. The stripper emissions allowances
include the part 63 general provisions allowances because
information was not available to differentiate these emissions
from normal stripper operating emissions.
Regarding the commenters' discussion of whether the general
provisions to part 63 would cover maintenance and troubleshooting
downtime, EPA has taken public comment and is currently revising
the requirements of the general provisions. Among the changes to
the language, EPA intends to incorporate safety-related venting
requirements into the general provisions. However, scheduled
maintenance activities are not considered by EPA to qualify for
excess emissions allowances. The EPA contends that the startup,
shutdown, and malfunction plan provisions specified in the
general provisions to part 63 should address the periods of
excess emissions that are caused by unforeseen or unexpected
events.
4.4 FORMAT OF THE STANDARDS
4.4.1 General Comments
4.4.1.1 A Percent Reduction Requirement is Unenforceable.
Comment: One commenter (20,059) objected to the use of
a percent reduction requirement as an emissions standard since a
reduction requirement cannot be enforced and verification cannot
be determined without measurement both before and after control.
The commenter (20,059) argued that non-complying companies could
manipulate the estimate of pre-controlled emissions to avoid
detection of violations. The commenter (20,059) indicated that
-------
EPA should establish numerical, pound-per-hour emission rates on
a continuous basis for the standards.
Response: The EPA disagrees with the commenter. Percent
reduction formats were specified in the rule only in cases where
it was not feasible to prescribe a numerical emission rate. The
most common emission rate format generally is one expressed as
mass-per-unit-of-production, since pound-per-hour rates vary with
production capacity and utilization rate. For this rule,
however, it was not always feasible to develop a
mass-per-unit-of-production format because of lack of data or
because of the degree of variability of uncontrolled emissions.
Since proposal, additional test data have been submitted to
EPA (A-92-40, IV-D1-29, IV-Dl-29a, IV-D1-31, IV-D1-33, IV-D1-34,
IV-D1-35, IV-D1-38, IV-D1-39, and IV-D1-41) to better
characterize HAP emissions from the pulp and paper processes.
Although the test data were a significant improvement over the
data available at proposal, the data were not adequate for
developing numerical emission standards for all pulping and
bleaching processes at all pulping subcategory types.
The EPA disagrees that a percent reduction format is
unenforceable. Percent reduction requirements have been included
in numerous NSPS and NESHAP and have been demonstrated to work.
State and Federal enforcement officials are accustomed to
enforcing these types of standards. Additionally, the general
provisions to part 63 specifically prohibit circumvention of the
standard by the use of diluents to achieve compliance.
Surrogate for HAP.
Comment: Several commenters (20,027, 20,056, 20,057A2,
20,071, IV-D1-15) supported the use of methanol as a surrogate
4-89
-------
measure for total HAP. One commenter (IV-D1-15) also agreed with
using methanol as a surrogate for total HAP for pulping and
wastewater sources since methanol typically constitutes
90 percent or more of the total HAP' s in these sources. For
bleach plants, the commenter (IV-D1-15) also agreed with EPA's
intent to use chlorine as a surrogate for compliance
determinations since only chlorinated HAP's are controlled by
MACT technologies. Several commenters (20,102, 20,110, 20,111,
20,129) suggested that EPA establish pollutant-specific emissions
limitations for bleaching equipment such as chloroform, chlorine,
and ClC>2 .
Several commenters (20,022, 20,049A3, 20,059, 20,090,
20,122, 20,132, 20,133) requested pollutant-specific limits on
air pollutants. One commenter (20,110) indicated that EPA should
ensure that HAP's other than methanol and chlorine, such as
phenol and chloroform, are controlled to an efficiency of at
least 90 percent. One commenter (20,102) indicated that EPA has
the authority to implement pollution prevention opportunities
such as source reduction through the MACT development process.
The commenter (20,102) suggested EPA might improve that ability
through some pollutant-specific limitations in addition to total
HAP emissions.
One commenter (20,059) argued that EPA authorized a form of
interpollutant trading by failing to establish emission limits
for individual pollutants. The commenter (20,059) indicated that
proposed process changes and control technologies differ in the
amount of specific HAP' s they reduce, yet EPA lumped all of the
pollutants together regardless of toxicity. Two commenters
(20,102, 20,103) stated that EPA should give special attention to
4-90
-------
chloroform due to its toxicity and because it is a carcinogen.
The commenters (20,102, 20,103) requested that EPA provide more
technical guidance on its evaluation and control. One commenter
(20,129) stated that emissions of carcinogenic compounds from
bleaching vents should be controlled to 10"^ to 10~6 inhalation
risk levels with BACT air cleaning technology.
Response: The final rule is a technology-based standard
with the MACT level of control based on the performance of
technologies that achieve the greatest level of emissions
reduction. The pulping process emits non-chlorinated HAP's
(predominantly methanol) while the bleaching process emits
chlorinated HAP's (such as chloroform, chlorine) and non-
chlorinated HAP's.
Each of these types of HAP (non-chlorinated and chlorinated)
has different applicable control technologies (i.e., combustion
and caustic scrubbing, respectively). For this reason, EPA re-
evaluated the floor level of control for each of these types of
HAP as discussed in the March 8, 1996 supplemental notice. As a
result, pulping standards were based on combustion with methanol
as the surrogate compound. For the bleaching process, standards
were developed for chloroform, and other chlorinated HAP's (with
chlorine as the surrogate compound).
The EPA maintains that methanol is an appropriate surrogate
for non-chlorinated HAP's since methanol is the majority of the
non-chlorinated HAP's found in pulping process vents and
wastewater based on the available data (A-92-40, IV-A-8).
Chlorine was designated as a surrogate for chlorinated HAP's
(other than chloroform) because the MACT floor level of control
technology, caustic scrubber, was installed primarily for
4-91
-------
chlorine control. Therefore, control of chlorine should indicate
proper operation of the caustic scrubber. Chloroform is
controlled through process changes such as C1C>2 substitution and
elimination of hypochlorite. The EPA contends that the control
technologies selected for the pulping and bleaching processes and
the surrogate compounds selected for measurement ensure the
adequate control of total HAP compounds.
With regard to the commenters' discussion of pollutant-
specific limits, EPA asserts that the level of the standards
would not be significantly different, if at all, had the standard
been based on specific HAP's. The rationale for this assertion
is that EPA evaluated all of the reasonably applicable control
technologies and determined that the technologies chosen in the
final rule would achieve the maximum emission control of total
HAP'S. Some other technologies (e.g., incineration of bleach
plant vents) may achieve greater control of a specific HAP, but
would achieve lesser control of other HAP's. The EPA determined
that it was not cost feasible to require these technologies in
addition to the floor level of control technologies.
Additionally, EPA does not have sufficient data to establish
pollutant specific limits for all HAP's from all emission
sources. Regarding the relative toxicity between HAP's, EPA is
not authorized under section 112(d) of the Act to establish MACT
using any type of toxicity weighing. In any case, the MACT
standards will control all HAP's.
The EPA does not believe that pollutant-specific emission
limits are needed to encourage pollution prevention. The final
rule contains provisions for a compliance alternative that
focuses on achieving the required emissions reduction from
4-92
-------
process vent emissions by reducing the HAP content of process
waters recycled or reused in various mill processes. The EPA
contends that this alternative adequately encourages mills to
pursue pollution prevention options since a specific control
technology is not identified in the compliance option.
Additionally, the effluent guidelines portion of the final rule
provides incentives for adopting pollution prevention
technologies.
Comment: One commenter (20,059) stated that EPA failed to
indicate the time period over which the percent reductions
specified in the standard are to be achieved. The commenter
(20,059) also stated that the averaging times should be short to
limit cumulative exposure and to protect the public from short-
term exposure to highly toxic pollutants.
Response: The language in the final rule has been clarified
to identify the averaging times for the specific parameters to be
monitored. The final rule specifies that mills must conduct
performance tests to determine the necessary operating parameters
such that the specified emission reduction will always be
achieved. Consequently, a violation of the parameter(s) becomes
a violation of the standards.
Innovative pollution control systems as equivalent to MACT.
Comment: One commenter (20,102) stated that pollution
prevention opportunities should be encouraged in the MACT
standards. The commenter (20,102) indicated that a method to
generate pollution credits by using a non-polluting technology
might be a good incentive. The commenter (20,102) suggested that
EPA allow mills to use a non-polluting technology to receive
credit for a percentage of the pollution that would be emitted by
4-93
-------
a facility with the same capacity using conventional technology
with emissions at the level of the MACT standards.
Response: The EPA welcomes innovative pollution control
systems and does not prohibit sources from using a different
method to achieve pollution prevention or reduction. In an
effort to encourage pollution prevention and maximize the multi-
media pollution prevention, EPA provided a 5-year compliance
extension to kraft mills for controlling HVLC vents and oxygen
delignification systems. Rationale for providing the extension
was presented in the March 8, 1996 supplemental notice. A source
may petition for equivalency based on the amount of pollution
reduction it achieves. The EPA must, however, base its MACT
standards on the reductions achievable by existing technologies
and a source must be able to demonstrate those reductions for
enforcement purposes.
Process modification should not be used as environmental
control.
Comment: One commenter (20,039) stated that it was counter-
productive to require process modifications as a means of
environmental control if the existing manufacturing process can
achieve the same environmental protection without the required
modification. The commenter (20,039) also indicated that EPA
should not set standards for points within operating systems as
complex as those found in the pulp and paper industry. The
commenter (20,039) added that standards should be established at
the intersection of the emission or discharge and the
environment.
Response: The EPA disagrees with the commenter's assertion
that it was counter-productive to require process modifications
4-94
-------
as a means of environmental control if the existing manufacturing
process can achieve the same environmental protection without the
modification.
The process modification referred to by the commenter is BAT
in the effluent limitation standards and guidelines for the
paper-grade bleaching process. The BAT requires the substitution
C102
of and elimination of hypochlorite. The MACT level of
control for paper-grade bleaching systems is control of
chloroform and the other chlorinated HAP emissions through a
combination of caustic scrubbing, 100 percent _,- substitution,
LJ.U2
and eliminating the use of hypochlorite. The BAT requirements
are at least equal to the MACT requirements. Therefore, this
level of control is required for compliance. The EPA is not
aware of any other control technology that would achieve the same
level of control. The general provisions to part 63 63. 6 (g) ] ,
IS
however, provide directions for obtaining approval of alternative
control technologies.
Limitations should be set in terms of total HAP ton of
per
production.
Comment: Two commenters (20,027, 20,045) suggested that the
emission standards should be stated in terms of a total HAP
emissions rate per ton of production, which one commenter
(20,045) stated would be similar to the effluent guidelines.
Response: The final rule contains compliance alternatives
that include a total HAP per ton of production emission
limitation for the sulfite pulping subcategory due to the
complexity of these systems and the problems that may occur when
testing these sources for compliance with the standard. The
final rule also includes a total HAP per ton of production
4-95
-------
emission limit for knotter and screen systems at kraft mills.
This limit was included in the final rule because EPA did not
have data other than mass emission rates to identify knotter and
screen systems with high emissions. The EPA does not consider a
total HAP per ton of production emission limitation to be
necessary for other pulp and paper processes.
Regarding the commenter's suggestion of using the effluent
guidelines to set total HAP emission rates, the effluent limits
are set for one or two points. There are many more air emission
points that need to be controlled. The EPA does not have
sufficient data to set pound per ton standards in most cases.
Support and comment on the named streams approach.
Comment: Several commenters (IV-D2-14, IV-D2-10, IV-D2-3,
IV-D2-8, IV-D2-7) agreed with the concept of selecting named
vents and streams for control. One commenter (IV-D2-15) stated
that EPA should list vents and streams for the sulfite pulping
subcategory whose MACT floor level of control should be no
control for non-halogenated HAP. The commenter (IV-D2-15)
suggested the following: bleach stage washers, tower vents and
seal tank vents, continuous digester steaming vessel, batch
digester fill/evacuation vent, knotter vents, screen vents,
decker (including thickeners and rewashers) vents, unwashed stock
tanks vents, intermediate filtrate tank vents, evaporator
condensate tank vents, spent sulfite liquor tanks, acid
condensate storage tanks, evaporator condensates, digester blow
gas condensates, digester relief gas condensates, and wastewater
collection, storage, and treatment vents (except to determine the
amount of methanol volatilized).
4-96
-------
Response: The EPA appreciates the commenter's support for
the approach of naming streams to be controlled. The EPA does
not intend to identify streams and vents not requiring control.
This would make the rule more confusing and is unnecessary. The
final rule specifies exactly which streams are to be controlled.
Alternative compliance determinations.
Comment: One commenter (IV-D2-15) supported EPA's intention
to allow sulfite mills to use any combination of controls to
achieve either the specified percent reduction or emission limit,
where applicable. Another commenter (IV-D2-16) agreed with EPA's
decision to incorporate a mass emission rate in addition to
a percent reduction standard into the final rule for sulfite
mills.
Response: The EPA appreciates the commenters' support. As
discussed in the March 8, 1996 supplemental notice, EPA did not
intend to specify the technology to be used to satisfy the
standards for sulfite mills. Rather, a mass emissions limit was
established for selected vents and wastewater emissions. This
format was intended to provide sulfite mills with flexibility in
complying with the sulfite rule. The final rule also includes
a percent mass reduction compliance option for additional
flexibility.
Sulfite pulping - statistical arguments concerning emission
limits and data variability.
Comment: One commenter (IV-D2-15) stated that EPA did not
establish the mass emission limits for sulfite mills properly.
The commenter stated that the mass limits are based only on "a
handful of ballpark values" and not a rigorous assessment of
emission rates. The commenter also stated that the variability
4-97
-------
of the processes and scrubbers must be considered because the
sulfite mills use variations in process parameters to make a wide
range of products.
The commenter provided several sets of data to support
concerns about process variability. The commenter also provided
several statistical analysis of the data and arrived at the
conclusion that the mass emission limit for selected vents from
magnesium and ammonium-based mills should be greater than 2.0 Ib
methanol/ODTP.
Two commenters (IV-D2-6, IV-D2-14) supported the argument
that to take the variability inherent in the industry into
account, the emission limit should not be set at the average
emission level but at an upper confidence limit based on the
relative standard deviation of the data sets.
Response: The EPA concurs with the commenters that process
variability should be incorporated into the sulfite rule mass
emission limit compliance option. Establishing the appropriate
mass emission limit was critical for the sulfite subcategory
since a reference control technology for these mills was not
identified in the standard.
As discussed in the March 8, 1996 supplemental notice, EPA
established mass emissions limits from selected vents and
wastewater for sulfite mills. At the time the supplemental
notice was published, the numerical mass limits were based on a
limited amount of data. Since that time, EPA has received test
data from several facilities (A-92-40, IV-D1-96 and IV-D1-100)
documenting the variability of process emissions over time.
These data were used to estimate the variability of the original
4-9E
-------
data set used to develop the mass emission limits presented in
the March 1996 supplemental notice.
The variability analysis was based on the 99.9 percent
confidence interval of the data supplied by the pulp and paper
industry. This amount of variability (confidence interval),
therefore, was applied to the average emission limits from the
best controlled mills to develop the final emission limit. After
the close of the March 8, 1996, Federal Register supplemental
notice comment period, additional information was provided to the
Agency that indicated that the sodium-based sulfite pulping
process is in use at some mills (A-92-40, IV-E-86, and IV-E-94).
No emissions information was available for this process.
However, the Agency determined, that due to the similarities in
processes between calcium- and sodium-based sulfite pulping
processes, the same limit developed for calcium-based mills would
be applicable to sodium-based mills. For sodium- and
calcium-based sulfite pulping processes, the final emission limit
is 0.44 kilograms of methanol per megagram ODP produced. For
ammonium- and magnesium-based sulfite pulping processes, the
final emission limit is 1.1 kilograms of methanol per megagram
ODP produced. Since the emission limits include the variability
allowance, they are never-to-be exceeded values.
Calculation of mass percent reduction.
Comment: One commenter (IV-D2-14), noted that EPA had not
determined how to measure and calculate the mass percent
reduction for sulfite mills. The commenter (IV-D2-14) proposed
the following approach:
4-99
-------
100 * (uncontrolled emissions - controlled emissions)
percent =
reduced uncontrolled emissions
where:
uncontrolled
emissions = the sum of methanol emitted from uncontrolled
selected vents and control equipment vents for the
selected vents, methanol in gases directed to
combustion sources, and methanol in wastewater
streams from selected vent control equipment; and
controlled
emissions = the sum of methanol emitted from control equipment
vents for the selected vents, and methanol emitted
from wastewater treatment that can be attributed
to wastewater streams from selected vent control
equipment.
Response: In the March 8, 1996 supplemental notice EPA
proposed to provide a mass emission limit and percent reduction
option in the final rule. The EPA evaluated commenter's
suggested equation and agreed the equation is appropriate for
determining mass percent reduction at sulfite mills. The final
rule incorporates this equation.
Compliance should be determined on test and annual averages.
Comment: One commenter (IV-D2-16) agreed that compliance
for sulfite mills should be based on test averages and that
compliance should not be based on the results of any one test.
The commenter (IV-D2-16) also suggested that compliance with any
pounds per ton emission limit should be expressed as an annual
average to account for process and testing variability.
Response: The determination of compliance, conducted with
the initial performance test, is based on the average of three
4-100
-------
1-hour tests. During the initial performance test, the mill must
select the appropriate parameters for monitoring compliance.
The EPA does not concur with the commenter that an annual
average would be sufficient for demonstrating continuous
compliance especially since the required reporting periods are
typically semi-annual or quarterly.
4.4.2 Pulping System
4.4.2.1 Operating Scenario as Compliance Alternative.
Comment: One commenter (IV-D2-5) noted that their
brownstock washers do not use condensates for washing, and as a
result, the emissions are relatively low. The commenter included
test data from the washers and requested that EPA allow this
operating scenario as a control option in place of incineration.
Response: The EPA maintains there would be significant HAP
emissions from these washers even if they did not use recycled
condensates (see the final emission factor document, A-95-40,
IV-A-8). In addition, the final rule contains a compliance
option for a HAP outlet concentration for thermal oxidizers. If
emissions from the brownstock washers are less than emissions
specified in the rule, then they would not need to meet
the percent reduction requirement.
4.4.3 Bleaching System
4.4.3.1 Alternative Compliance Determinations.
Comment: Several commenters on the proposal (20,027,
20,070A1, 20,118) requested that EPA specify alternative
compliance demonstrations for the bleaching component similar to
the options given for pulping and wastewater components. Several
commenters (20,027, 20,057A2, 20,059, 20,070A1, 20,071) suggested
the following alternatives in addition to the proposed standard
4-101
-------
(percent removal): outlet concentration-based limits, mass per
unit production-based limits, design scrubber specifications,
process changes and/or stipulation that bleaching without
elemental chlorine constitutes compliance.
One commenter (20,071) indicated that an outlet
concentration would be more desirable because the determination
of scrubber efficiency could be difficult for devices run in
series and/or parallel. Another commenter (20,057A2) suggested
that EPA revise the bleach plant scrubber removal efficiency
specification to a numerical limit of 5 ppmv chlorine. Two
commenters (20,057A2, 20,071) indicated that the removal
efficiency of 99 percent would be impossible to measure or
achieve as the amount of chlorine is reduced due to 2
substitution.
One commenter (20,070A1) stated that EPA should establish an
alternative compliance demonstration for mills that do not use
elemental chlorine. The commenter (20,070A1) argued that the
proposed process changes would greatly reduce the concentration
of elemental chlorine at the inlet to bleach plant scrubbers
which would, in turn, make demonstrating the high removal
efficiency for chlorine difficult.
One commenter on the March 8, 1996 supplemental notice
(IV-D2-15) supported the bleach plant applicability requirements
and agreed with EPA that an outlet concentration below 10 ppmv of
HAP from the scrubber exhaust is equivalent to the 99 percent
reduction standard. Three commenters (IV-D2-10, IV-D2-3,
IV-D2-8) supported the concept of a concentration limit on the
bleaching component for control of chlorinated HAP, noting that a
concentration limit in place of a percent reduction limit would
4-102
-------
not require two sampling events (inlet and outlet) and would
eliminate several potential problems in compliance determination
if the inlet is of very low concentration. Additionally, one
commenter (IV-D2-3) asked EPA to make clear that only elemental
chlorine need be measured to demonstrate compliance with the
"total chlorinated HAP" control requirement. Two commenters
(IV-D2-14, IV-D2-15) recommended establishing an alternative mass
standard for cases where low flow systems are used and would have
difficulty meeting the concentration limit. One commenter
(IV-D2-15) recommended an equivalent mass flow rate would be
equal to or less than 0.01 kg/Mg air-dried pulp (ADP)
(0.025 Ibs/air-dried ton of pulp (ADTP) ) and provided data to
support this limit. The commenters (IV-D2-14, IV-D2-25)
expressed concern that bleaching systems with new low-flow vent
systems would not be able to meet either the percent reduction or
the outlet concentration standards. Therefore, these standards
would discourage the use of new low-flow bleaching vent
technologies. Based on this concern, one commenter (IV-D2-15)
advocated a chlorinated HAP mass emission limit for bleaching
systems of 0.023 Ib of chlorinated HAP (excluding chloroform) per
ODTP produced. The commenter (IV-D2-15) claimed that a mass
emission limit would not penalize new low-flow bleaching vent
systems.
One commenter (IV-D2-14) noted that new bleaching sequences
should not be measured against the performance standards of
existing bleaching sequences. The commenter (IV-D2-14) also
noted that the chlorine scrubbing technology proposed should not
be required.
4-103
-------
Response: At proposal, the bleach plant requirements were
control of chlorinated HAP emissions (or chlorine as a surrogate)
by 99 percent. This determination was based on industry data on
scrubber performance (A-92-40, II-I-24) and irrespective of the
effect of process changes also required by the proposed rule.
After reviewing the comments on the proposal EPA analyzed the
effect of C1O2 substitution and elimination of hypochlorite use
on emissions. Based on this analysis, EPA determined that, in
some cases, emissions after process changes are enacted would be
lowered such that 99 percent reduction could not be achieved. As
stated in the March 8, 1996 supplemental notice, EPA determined
that a standard of 10 ppmv of total chlorinated HAP (other than
chloroform) is equivalent to the outlet of scrubbers achieving
99 percent removal (A-92-40, II-I-24). Chlorine may be used as a
surrogate.
Regarding the commenter's request for a mass emission limit
standard, EPA reviewed the information provided by the commenter
and emission information from sampling tests conducted on
bleaching systems. Based on available data, the Agency has
concluded that low-flow bleaching vent systems can achieve the
99 percent reduction and the 10 ppmv outlet concentration
requirements for total chlorinated HAP (other than chloroform).
Based on a review of the information provided by the commenter
and the available data on bleaching system emissions, the Agency
has concluded that the commenters recommended mass emission limit
of 0.023 Ib of chlorinated HAP (excluding chloroform) per ODTP
produced is too high. The Agency evaluated the available data
used to develop the percent reduction and outlet concentration
requirements for bleaching systems (A-92-40, II-I-24). From this
4-104
-------
evaluation, the Agency determined that a scrubber outlet mass
emission rate of 0.001 kg of total chlorinated HAP (other than
chloroform) per Mg ODP produced (0.002 Ib/ODTP) would provide
reductions equivalent to 99 percent reduction standard (A-92-40,
IV-B-29) . The mass emission limit of 0.001 kg of chlorinated HAP
(other than chloroform) per Mg ODP produced represents a mass
emission limit achievable by all units that also achieved
99 percent reduction of chlorine. Furthermore, the available
data show that some of the scrubbers achieving the 99 percent
chlorine reduction standard, and the 10 ppmv outlet concentration
limit, were also operating on low-flow bleaching vent systems.
For the final rule, the Agency has provided a mass emission
limit option for bleaching systems of 0.001 kg of chlorinated HAP
(excluding chloroform) per Mg ODP produced (0.002 Ib/ODTP). The
Agency maintains that this option allows more flexibility for
sources affected by this rule, does not penalize bleaching
systems operating with low-flow technology, and will provide
reductions in chlorinated HAP emissions (other than chloroform)
equivalent to the 99 percent reduction standard.
4.4.4 Process Wastewater System
4.4.4.1 Design steam stripper
Comment: Two commenters (20,000, 20,056) recommended that a
parametric design steam stripper not be specified in the
standards. One commenter (20,000) indicated that if mills cannot
install a steam stripper in the manner they determine to be most
cost effective there will be a heavy penalty in energy costs and
C02 emissions. The other commenter (20,056) recommended that
each mill be allowed to develop engineering calculations or
4-105
-------
perform testing to indicate stripper operating ranges which
correspond to 90 percent reduction.
Response: The EPA agrees with the commenters. In an effort
to provide sources with flexibility to comply with the wastewater
requirements the design steam stripper specified in the proposed
regulation is not required in the final rule. Sources are free
to design their own steam strippers as long as they are able to
meet the required control level (92 percent reduction, mass
removal, or outlet concentration).
Support for compliance alternatives.
Comment: One commenter (IV-D2-15) supported the inclusion
of other compliance options for the treatment of pulping process
condensates, namely, recycling to enclosed equipment or hard-
piping to a mill's biological treatment plant. The commenter
also supported EPA's decision to allow mills to choose any
wastewater treatment device for compliance purposes as long as
either the percent reduction, mass removal, or outlet
concentration is met.
One commenter (IV-D2-16) suggested requiring all pulping and
bleaching effluents be sent to a well-operated secondary
treatment plant, realizing that a well controlled operation of
this type will perform as well as possible to biodegrade methanol
and related compounds.
Response: The EPA appreciates the commenters' support of
compliance alternatives. Hardpiping the affected streams to
biological treatment is a compliance option for kraft pulping
wastewater streams in the final rule. The MACT floor level of
control for non-kraft wastewater and bleaching systems wastewater
is no control.
4-106
-------
Miscellaneous clarifications to the wastewater rule.
Comment: One commenter (20,027) requested that EPA include
the following requirements in the process wastewater standards
for sulfite mills: (1) an enclosed transport system is not
needed for a wastewater stream that has been treated to below
500 ppmw HAP by one of the proposed compliance options; and
(2) after the mandated 90 percent HAP reduction has occurred, a
treated process wastewater stream is acceptable for reuse
anywhere within the mill or for disposal in open sewers.
Response: At proposal, EPA did not establish a sulfite
subcategory. Sulfite mills were required to meet the kraft mill
requirements, including wastewater. Since proposal, EPA has
developed a sulfite subcategory. The EPA determined that control
of sulfite wastewater was not in the MACT floor level of control,
and it was cost infeasible to go beyond the floor level of
control. Therefore, the final rule does not include wastewater
requirements for sulfite mills. However, the HAP emissions from
any effluent associated with any device used to reduce vent HAP
emissions must be included in the selected vent emission limit.
Air emissions from biological treatment are unmeasurable.
Comment: One commenter (20,057A2) suggested that the
90 percent removal efficiency for biological treatment systems be
modified to pertain only to a 90 percent efficiency in water, and
should exclude de minimis air emissions. The commenter
(20,057A2) indicated that procedures do not exist to measure
de minimis air emissions from biological treatment.
Response: In the final rule, biological systems at kraft
mills are required to remove 92 percent of HAP in the wastewater
system. This efficiency was revised from 90 percent based on
4-107
-------
EPA's re-analysis of steam stripping performance (A-92-40,
IV-B-10). Kraft mills are required to demonstrate that the
removal is actually destruction of the HAP's rather than HAP
reduction due to volatilization to the air, adsorption, or HAP
loadings lost in discharge. The EPA considers this a necessary
requirement in order to ensure that kraft mills choosing the
biological treatment option are achieving the requested HAP
reduction.
Methanol from wastewater should not be included in overall
emission requirements.
Comment: One commenter (IV-D2-16) strongly disagreed with
EPA's proposal to include methanol emissions from the wastewater
treatment plant in its overall emission requirements. The
commenter (IV-D2-16) asserted that there is no practical or cost-
effective way to change the methanol removal efficiency of the
wastewater treatment system and noted that the only "measurement"
of the system output is determined by a model. The commenter
(IV-D2-16) requested that if a model is required, then the model
parameters must be spelled out explicitly to that they will not
change in the future.
Response: The wastewater control requirements are only for
kraft mills. Biological treatment of kraft wastewater is not
required in the final rule, but it is one of the control options
a mill can choose. If a mill chooses the biological treatment
option, it will need to account for the HAP emissions from the
biological treatment system. The EPA did not specify parameters
for estimating emissions using emission models because the
parameters need to be developed on a site-specific basis.
4-108
-------
Regarding the methanol removal efficiency of biological
treatment systems, data analyzed by EPA (A-92-40, IV-D1-75)
indicate that a well-operated biological treatment system can
easily achieve a methanol destruction greater than 92 percent.
Therefore, the methanol removal efficiency of biological
treatment systems does not need to be changed.
Site-specific basis for compliance determination.
Comment: One commenter (IV-D2-8), noting the site-specific
variability in model inputs to determine volatilization rates,
asserted that EPA must allow facilities to develop site-specific
rates to determine compliance. The commenter (IV-D2-8) , noting
the difficulty of such an implementation, recommended deleting
the emissions limit requirement for wastewater systems. Another
commenter (IV-D2-16) recommended that EPA only regulate process
vents and not regulate wastewater treatment plants.
Response: The EPA disagrees with the commenters. A MACT
floor level of control analysis conducted on the kraft pulping
process condensate streams shows that the average emission
limitation is control of specific condensate streams with a steam
stripper (A-92-40, IV-B-8). Therefore, the kraft pulping process
condensate streams from the digester system, turpentine recovery
system, and the weak liquor feed stages in the evaporator system
are required to be controlled. Additionally, EPA has decided to
name streams to be controlled in the final rule (see March 8,
1996 supplemental notice for a detailed discussion); therefore,
no concentration or flow cut-offs have been included in the rule.
The EPA considers this change to simplify compliance
determination.
4-109
-------
The EPA also believes that it has provided sufficient
alternatives in the wastewater requirements (hardpiping, steam
stripping achieving 92 percent reduction, or mass limit or
concentration limit) to provide sources with flexibility to meet
the standard.
4-110
-------
5.0 PULPING AREA
5.1 CONTROL OPTIONS
5.1.1 General
Comment: One commenter (20,081) objected to EPA's listing
of preferential control technologies in the December 17, 1993
proposed rule. The commenter (20,081) contended that
biofiltration should be included as a viable alternative and
listed as one of the preferred technologies. The commenter
(20,081A1, 20,081A2) provided several reasons why biofiltration
should be used instead of incineration and condensation, and
provided comparisons of cost for the technologies. The commenter
(20,081A1, 20,081A2) also provided control efficiencies for a
pilot-scale test of biofiltration.
Two commenters (20,029, 20,041) requested that other types
of oxidizers be evaluated, such as regenerative thermal oxidizers
and catalytic oxidizers for controlling emissions from the
pulping area. The commenters (20,029, 20,041) claimed that these
devices could achieve the same reduction in emissions as thermal
incinerators and may be less costly.
Response: Based on test data, and information summarized
from the 1992 voluntary MACT survey (A-92-40, IV-B-16), a level
of control was selected as MACT for each pulping subcategory.
These control levels were expressed as percent reductions or
5-1
-------
emission limits. Any control technology that can be demonstrated
to meet the requirements specified in the rule can be used.
The rule provides several options for demonstrating
compliance. For example, owners or operators of kraft, semi-
chemical and soda mills may comply with the pulping provisions by
controlling emissions from named streams by 98 percent, or by
routing named vents to lime kilns, boilers, or to a thermal
oxidizer meeting specified requirements. These control
technologies are control options in the rule because they have
been demonstrated to achieve at least 98 percent control when
operated under the specified conditions. Compliance testing to
demonstrate 98 percent reduction is not required when using these
technologies. Therefore, specifying these technologies reduces
the compliance burden. However, any alternative technology that
can be demonstrated to achieve 98 percent control may be used to
comply with the rule.
In cases where the rule specifies a design, equipment, or
operating standard, 40 CFR 63.6(g) includes procedures for
obtaining approval by the Administrator of the use of alternative
control technologies that can be demonstrated to perform as well
as, or better than, a technology specified in a rule.
Comment: One commenter (20,072) recommended that MACT for
soda mill pulping components be limited to 95 percent methanol
capture of digester blow gases, screen room closure, using mill
water supply for the brownstock washers, and no internal
recycling of digester and highly contaminated evaporator
condensates which will be directed to a mill's wastewater
treatment system. The commenter (20,072) provided estimates for
their proposed control of soda mill pulping components, which
5-2
-------
they estimate would reduce emissions by 85 percent at an
estimated capital cost of 4.1 million dollars. The commenter
(20,072) stated that their cost estimates are substantially lower
than EPA's estimate. The commenter (20,072) provided cost-
effectiveness data.
Response: Data that EPA has collected on soda mills
indicate that none are currently controlling pulping or
wastewater vents. Information supplied since proposal of the
rule has allowed EPA to characterize emissions and possible
emission controls for soda mills. Based on this new information,
EPA has determined that controlling the LVHC system is reasonable
considering the cost and other impacts; controlling additional
vent streams beyond the LVHC system or controlling the wastewater
collection and treatment system is not reasonable considering
costs and impacts. Information submitted by the industry
supports EPA's conclusions (A-92-40, IV-D1-77, IV-D1-90).
Therefore, the final rule requires that owners or operators of
soda mills control only the LVHC vents at existing mills.
Commenters are referred to supporting memorandum (A-92-40,
IV-B-13) and chapter 20 of this document for a detailed
discussion of the control options and impacts analyses.
Control cost estimates provided by the commenters outlined
results for selected control scenarios based on using mill water
supply for the pulp washers. This option would increase mill
wastewater discharge and be counter to EPA's plan to achieve a
"closed mill" from an effluent standpoint. The commenters'
recommended controls did not address all equipment included in
the LVHC system. Other equipment in the LVHC system, such as
evaporators, may emit significant HAP's. The EPA lacked
5-3
-------
sufficient information to verify the commenter's claim that their
control suggestions would achieve 85 percent reductions in HAP
emissions. The EPA determined that control of the LVHC system
(of which the digester flow gases are only a part) will result in
a significant amount of HAP emission reduction at a reasonable
cost (see discussion in March 8, 1996 supplemental notice).
Comment: Two commenters (20,053A1, 20,018) were opposed to
the floor including control of HVLC vents in the pulping system,
specifically citing brownstock washer vents, oxygen
delignification vents, and vents in the bleaching process prior
to the addition of chlorinated bleaching agents. (These
bleaching stages were included in the pulping component
definition at proposal.) Both commenters supported their
argument by pointing out potential drawbacks of controlling these
HVLC vents, citing secondary impacts such as S02 and NOX
emissions, plus additional fuel requirements for incinerator
operation. One commenter (20,053A1) argued that the proposed
emissions standards for the bleaching and wastewater components,
combined with collection and incineration of LVHC gas streams
within the pulping system, would result in substantial reductions
in HAP emissions. One commenter (20,018) contended that the
current state-of-the-art kraft mill does not collect and
incinerate the oxygen stage vents.
Response: The NESHAP are required by the Act to reflect (at
a minimum) the average emission limitation achieved by the best-
controlled 12 percent of sources. As stated in the December 17,
1993 proposal, brownstock washers, oxygen delignification units,
and bleach systems are controlled at the floor for kraft mills.
A reevaluation of this floor determination was conducted using
5-4
-------
data received since proposal. These data indicated that the
best-controlled kraft mills control HVLC vents, including vents
on pulp washing systems, decker systems, oxygen delignification
systems, and knotter and screen systems. The best-controlled
mills control chlorinated HAP's from the bleaching system vents
using caustic scrubbers and process modifications. Data indicate
that the best-controlled bleaching systems do not control non-
chlorinated HAP's; therefore, control of non-chlorinated HAP's at
bleach stages is not required in the rule. Commenters are
referred to supporting memoranda (A-92-40, IV-B-8, IV-B-16) and
chapter 20 of this document for a detailed discussion of these
analyses.
The Act requires cost and environmental impacts to be
considered only when going beyond the floor level of control.
Therefore, S02 and NOX generation, and increased fuel use were
not considered in determining the MACT floor level of control,
but were considered when evaluating controls beyond the floor.
Readers are referred to the economics assessment (A-92-40, V-A-2)
for a detailed discussion of the benefits and impacts of the
final rule.
Comment: Three commenters (20,027, 20,0531A1, 20,054A2)
expressed concern over including oxygen delignification in the
pulping system, rather than as part of the bleaching system. Two
commenters (20,027, 20,054A2) stated that because oxygen is a
bleaching agent, oxygen delignification may be more appropriately
considered as part of the bleaching system. One commenter
(20,053A1) stated that, based on the proposed bleaching system
definition, TCF mills that do not apply chlorine or chlorine-
containing compounds at any stage in the bleaching system do not
5-5
-------
have a bleaching component. Thus, all bleach plant emissions
would apparently be regulated under the pulping system. The
commenter (20,053A1) indicated that EPA should identify a point
in the TCF bleaching process where the pulping system ends and
the bleaching system begins.
Response: Oxygen delignification was included in the
pulping component because of similar control technologies and
condensate re-use practices (e.g., much of the emissions from the
oxygen delignification system are attributable to pulping-area
condensates applied to the post-oxygen delignification washers).
For the final rule, oxygen delignification is still being defined
as part of the pulping component. Information submitted by two
commenters (A-92-40, IV-D1-97, IV-D1-71) after the close of the
comment period on the proposed rule indicates that the commenters
have reversed their position on this issue. The commenters
agreed with EPA that oxygen delignification should remain in the
pulping component due to the water recycling that occurs between
the pulping component and the post-oxygen delignification
washers.
The commenters' reaction to the proposed rule may have been
due to the historical classification by the pulp and paper
industry of oxygen delignification as part of the bleach plant.
The inclusion of oxygen delignification in the pulping component
may have caused confusion. The final rule specifies vent streams
that are required to be controlled, including oxygen
delignification system vents. Therefore, whether oxygen
delignification is defined as part of the pulping system or
bleaching system does not affect the control requirements.
5-6
-------
The final rule sets standards only for chloroform and other
chlorinated HAP's in the bleaching system; therefore, TCF mills
would already be in compliance with the bleaching system
standards.
For mills utilizing TCF bleaching processes along with an
oxygen delignification stage, EPA defines the end of the pulping
section as the screened stock chest that follows the oxygen
delignification unit. Pulping processes upstream of the screened
stock chest would be subject to NESHAP requirements for pulping
processes, while processes downstream of the stock chest would be
subject to NESHAP requirements for bleaching.
5.1.2 Existing Combustion Devices
Comment: Several commenters (20,011, 20,027, 20,018,
20,118) maintained that mills lack the existing combustion
capacity to handle the quantities of gases that would require
incineration under the December 19, 1993 proposed rule. One
commenter (20,018) maintained that current combustion units are
operating at their capacity. One commenter (20,027) indicated
that combustion of HVLC streams would increase the sensible heat
loss and therefore decrease steam production capacity in existing
power boilers. One commenter (20,118) added that many mills are
steam limited and would have to add new boiler or incinerator
capacity to burn additional HAP's. The commenter (20,118)
concluded that EPA must support their assumptions on existing
combustion capacities with data. One commenter (20,027)
supported the use of boilers, lime kilns, and recovery furnaces
as appropriate combustion units for the destruction of HAP's.
However, the commenter (20,027) did express concern over HVLC
flow variations and how they would affect boiler stability. The
5-7
-------
commenter indicated that this was of greater concern in semi-
chemical and soda pulp mills where no alternative combustion
devices are present.
Response: The capacity concerns expressed by several
commenters may have been due to the fact that the rule, as
originally proposed, required the control of all vent streams
except those with very low volume and low concentrations.
Therefore, the commenters may have concluded that most streams in
the pulping system were required to be controlled. The March 8,
1996 supplemental notice specifically defined the vent streams
that are required to be controlled. The EPA has maintained the
approach of naming specific systems to be controlled in the final
rule. This approach will reduce confusion as to which equipment
needs to be considered for control, and will alleviate many of
the commenter's concerns on capacity.
An analysis of the effects of HVLC streams being combusted
in an existing power boiler determined that a 5 percent increase
in fuel usage was needed to control the HVLC (A-92-40, II-B-31) .
The results of this analysis were based on the assumption that
the boiler would maintain the same level of heat release, thus
the same level of steam production.
In response to comments received after proposal, EPA also
conducted surveys with several mills regarding their capacity to
combust additional vent streams at existing combustion devices.
Results of the survey (A-92-40, IV-E-93) indicate that two-thirds
of the surveyed mills have the capacity in the existing
combustion devices to handle combustion of the named HVLC
streams. The remaining one-third would construct a thermal
oxidizer. The assumption that two-thirds of all mills will use
-------
existing combustion devices, and one-third will use new thermal
oxidizers to control vent streams was used in determining the
national cost and environmental impacts of the final rule (see
chapter 20 of this document).
Regarding the commenter's concern about HVLC system vent
flow variability, EPA does not expect there to be significant
variability in HVLC system flows such that the operation of the
power boiler or lime kiln would be disrupted. However, if a mill
encounters significant HVLC system flow rate variability that
would affect the performance of the boiler, these occurrences
should be addressed in the downtime and malfunction allowances
provided in the final rule. The final rule does not require
control of HVLC vents at existing semi-chemical and soda mills.
Currently, only LVHC systems at semi-chemical mills and soda
mills are being controlled.
Comment: Two commenters (20,115A2, 20,054A2) contended that
introducing LVHC gases into a recovery boiler could result in a
smelt water explosion due to moisture entrained in the gases.
One commenter (20,054A2) stated that combustion of LVHC gases in
a recovery furnace is not recommended by the Black Liquor
Recovery Boiler Advisory Committee.
Response: The EPA understands the commenters' concerns, and
agrees that a possibility exists that introduction of some vent
gases into the recovery boiler could have adverse results.
However, information supplied in industry questionnaire responses
indicates that some mills are successfully routing LVHC gases to
recovery furnaces (A-92-40, IV-B-8, IV-B-16). The proposed and
final rule do not require vent streams to be combusted in
recovery furnaces. If a facility is concerned with the safety
5-9
-------
issues associated with controlling vent streams in a recovery
furnace, the facility can choose to control the vent streams in
the other acceptable control devices specified in the rule.
Comment: Several commenters (20,011, 20,027, 20,115A2,
20,118) opposed the proposed compliance option for existing
combustion devices that required vent gases to be introduced with
the primary fuel or into the flame zone. One commenter (20,027)
asserted that this requirement did not accurately reflect the
floor level of control since NCG systems utilize a separate
burner or introduce the HVLC streams with the combustion air.
Other commenters (20,011, 20,118) contended that the feasibility
of introducing vent gases in the primary fuel or into the flame
zone of a combustion device was not adequately evaluated and may
not be appropriate in all cases. One commenter (20,118) cited
their mill as an example; it introduced TRS compounds separately
into boilers and lime kilns, not with primary fuels. One
commenter (IV-D2-15) recommended allowing the HAP laden gas
streams to be introduced with the "air supply" in order to
control emissions via a combustion device, since most mills do
not currently introduce such gases with the primary fuel or into
the flame zone.
Response: The proposed rule provided owners or operators
the option to achieve compliance with the combustion requirements
by introducing vent streams with the primary fuel or into the
flame zone. The intent of this stipulation in the proposed rule
was to prevent circumvention of the combustion requirements by
introducing the vent gases at a stage that would not allow for
complete combustion. The requirement to introduce the vent
streams with the primary fuel or into the flame zone is still
5-10
-------
necessary. Therefore, these requirements are included in the
final rule. The EPA references "Reactor Processes in the
Synthetic Organic Chemical Manufacturing Industry--Background
Information for Promulgated Standards," EPA-450/3-90-016b,
March 1993 to support this conclusion. This document provides
information that shows when vent streams are introduced into the
flame zone, over 98 percent reduction is achieved. However, when
vent streams are not introduced into the flame zone, complete
combustion is uncertain.
Comment: Several commenters (20,011, 20,059, 20,102)
disagreed with the 98 percent control requirement for pulping
vents. One commenter (20,011) stated that there were no data to
indicate existing pulp mill combustion devices are capable of
achieving 98 percent HAP removal efficiency or a 20 ppmv HAP
incinerator outlet concentration.
One commenter (20,102) opposed using the 98 percent control
requirement for incinerators for all toxic compounds because some
HAP's may be more carcinogenic or more toxic than others and
should be controlled more stringently.
One commenter (20,059) asserted that EPA provided no
evidence to support the contention that a 98 percent control
efficiency for vents is the highest that can universally be
achieved by incineration. The commenter (20,059) contended that
with proper operating procedures and maintenance, a higher level
of emissions reductions is achievable. The commenter (20,059)
also stated that EPA had not evaluated the relative effectiveness
of alternate combustion devices such as flares, lime kilns, or
chemical recovery furnaces. The commenter (20,059) contended
that EPA had not explored variability in load size and
5-11
-------
composition common in the paper industry. The commenter (20,059)
stated that this analysis could assess performance or establish
more specific control requirements.
Response: Information from industry surveys showed that the
best-controlled kraft mills route vent streams to combustion
devices (A-92-40, IV-B-8, IV-B-16). Historically, EPA has
conservatively assumed that combustion devices such as thermal
oxidizers, power boilers, and lime kilns achieve 98 percent
destruction of total organic compounds based on specified
temperatures and residence times (A-79-32, II-B-31). Data
provided by industry indicates that existing thermal oxidizers
can achieve 98 percent destruction of HAP's or reduce HAP
emissions to 20 ppmv at 10 percent oxygen (A-92-40, IV-B-18).
Some devices may achieve higher destruction efficiencies for
some compounds depending on various mill-specific factors (such
as operation, fuel use, manufacturer, etc.). These factors
cannot generally be duplicated at all mills. Additionally, a
study of VOC reduction in incinerators concluded that a
98 percent reduction of VOC is the highest control level that is
consistently achievable by an incinerator considering the range
of vent stream conditions that are likely to occur (A-79-32,
II-B-31).
5.1.3 Design Incinerators
Comment: Several commenters (20,000, 20,011, 20,027,
20,041, 20,051, 20,056, 20,057A2, 20,070A1, 20,074, 20,118,
20,144, IV-D2-11) objected to the design incinerator
specifications in the December 17, 1993 proposed rule. Most of
the commenters (20,000, 20,011, 20,027, 20,041, 20,051, 20,057A2,
20,070A1, 20,074, 20,118) claimed that the incinerator
5-12
-------
requirements were erroneously based on the SOCMI industry
incinerator requirements of 1600 °P and 0.72 seconds residence
time rather than on the existing kraft pulp mill NSPS standards
for TRS. Other commenters (20,041, 20,102, 20,129) claimed that
the design incinerator parameters are inadequate to ensure high
combustion of HAP's and should be more stringent.
Several commenters (20,027, 20,051, 20,056, 20,057A2,
20,070A1, 20,074, 20,118, 20,144, IV-D2-11, IV-D2-3) contended
that EPA should change the incinerator operating provisions to
the NSPS requirements of 1200 °F and 0.5 seconds residence time.
One commenter (20,057A2) recommended that the NSPS operating
requirements and the alternative 20 ppmv HAP emissions limit be
included in the rule to ensure that current incinerators can be
used for continued control of LVHC gases. The commenter
(20,057A2) stated that this would minimize new incinerator costs
by maintaining the viability of existing equipment. One
commenter (IV-D2-11) noted that if the 1600 °F and 0.75 second
criteria are promulgated for LVHC incinerators, then the criteria
should only apply to new units so existing incinerators will not
have to be replaced.
Two commenters (20,000, 20,070A1) provided examples of
incinerators that achieve high destruction efficiencies, while
operating at a temperature of 815 °C (1500 °F) and a retention
time of 0.15 seconds. One commenter (20,070A1) provided data
from tests conducted at a mill that showed methanol destruction
greater than 99 percent. The commenter (20,070A1) stated that
preliminary results indicate that destruction efficiencies of
99 percent are being achieved for TRS and methanol on gas streams
having significant concentrations. The commenter (20,070A1)
5-13
-------
referred to NCASI for further data on HAP control efficiencies.
One commenter (20,000) claimed that the combustion efficiency was
maintained in these incinerators by using a high-intensity
combustion chamber. The chamber ensures that pollutants are
thoroughly mixed into the flame.
Two commenters (20,102, 20,129) claimed that the design
incinerator operating requirements do not offer a significant
margin of safety for high combustion efficiency combustion of
HAP'S. The commenters (20,102, 20,129) recommended requiring an
incinerator temperature of 1800 °P and a residence time of
1 second. Another commenter (20,111) indicated that a
temperature of 1600 Op does not provide an adequate margin of
safety against the formation of dioxin and should therefore be
re-evaluated. Three commenters (20,102, 20,129, 20,144) stated
that a higher combustion efficiency should be specified, such as
a minimum of 99.9 percent or 100 ppm CO corrected to 7 percent
oxygen. One commenter (20,129) noted that these are the
requirements for thermal destruction of HAP' s in the RCRA
regulations.
One commenter (20,144) suggested that the rule allow streams
to be segregated, and require those streams containing
chlorinated HAP's to have a higher temperature and residence
time, but those that do not contain chlorinated HAP' s to meet the
NSPS required temperature and residence time. Another commenter
(20,041) requested that the proposed design incinerator
requirements be deleted from the final rule. The commenter
(20,041) contended that the design incinerator operation
provisions in the proposal were based on the average operating
parameters for thermal incinerators and concluded that some HAP' s
5-14
-------
would be destroyed while others would require higher
time/temperature conditions to achieve the desired destruction
rate. The commenter (20,041) cited methyl ethyl ketone as an
example; methyl ethyl ketone requires 1 second and 1780 °F to
achieve 99 percent destruction. The commenter (20,041) concluded
that a properly designed thermal incinerator would reduce HAP
emissions by 98 percent, but indicated that an incinerator
meeting the proposed operating requirements would not.
Response: The final rule retains the incinerator operating
parameters of 1600 °F and 0.75 seconds residence time. The EPA
has decided not to change the proposed design incinerator
operating parameters for the final rule because the parameters
are necessary to meet the MACT floor level of control. The EPA
would first like to clarify that the final rule does not limit
owners or operators of incinerators to operate at the specified
temperatures and residence times. Any control device that is
demonstrated to achieve 98 percent destruction of HAP's or any
thermal oxidizer that reduces HAP emissions to a concentration of
20 ppmv at 10 percent oxygen will comply with the rule. (The
outlet concentration limit option has changed from 20 ppmv at
3 percent oxygen, at proposal, to 20 ppmv at 10 percent oxygen.
This issue is discussed later in this section. ) The 98 percent
destruction requirement represents the control level achieved by
well-operated combustion devices. The 20 ppmv limit represents
the performance achieved by well-operated combustion devices on
low concentration vent streams.
Second, EPA has made this part of the rule as flexible as
possible while still achieving a level of control reflecting
MACT. In the December 17, 1993 proposal and in this final rule,
5-15
-------
EPA developed compliance alternatives in order to reduce the
compliance testing burden. The compliance alternatives (i.e.,
operating thermal oxidizers at a temperature of 1,600 ' and a
residence time of 0.75 seconds) were developed to ensure that the
thermal oxidizers perform at a level that would meet the
destruction efficiency requirements.
Information in industry survey (A-92-40, IV-B-8, IV-B-16)
responses indicates that the best-controlled sources are
controlling vent streams using an existing boiler as a combustion
device. Power boilers operate at much higher temperatures and
residence times than the incinerators required in the kraft mill
NSPS and can achieve at least 98 percent destruction of HAP's.
The incinerator operating parameters of 1600 °F and 0.75 seconds
residence time required in the proposed rule are based on
previous Agency studies (A-79-32, II-B-31) which show that these
conditions are necessary to achieve 98 percent destruction of
Op
HAP's. However, the NSPS operating parameters (1,200 r and
0.5 seconds residence time) do not destroy HAP's to this extent.
The EPA's analysis indicates that while the NSPS
requirements of 1200 op and 0.5 seconds residence time are
sufficient to achieve 98 percent destruction of TRS compounds,
kinetic calculations for methanol (the majority of HAP in pulping
vent gases) show that the NSPS criteria will not provide the
required 98 percent reduction of HAP's (A-92-40, IV-B-18).
Additionally, EPA evaluated incinerator performance data
submitted by industry (A-92-40, IV-J-33). The data indicated
that the NSPS operating parameters were not sufficient for
achieving 98 percent destruction of methanol. This conclusion
was reached by EPA since the operating conditions (i.e.,
5-16
-------
temperature and residence time) of the incinerators that achieved
98 percent methanol destruction were greater than the levels
specified in the kraft NSPS. Therefore, the NSPS specifications
will not meet the requirements of MACT for new and existing
sources. Information supplied by industry does show that some
existing thermal oxidizers are currently meeting the 98 percent
reduction or 20 ppmv standard (A-92-40, IV-B-18).
Historically, the EPA has conservatively assumed that
combustion devices such as incinerators, power boilers, and lime
kilns achieve 98 percent destruction of total organic compounds
based on specified temperatures and residence times (A-79-32,
II-B-31). Some devices can achieve higher destruction
efficiencies for some compounds due to various mill specific
factors (such as operation, fuel use, manufacturer, etc.). Due
to the variability of combustion devices, it would not be
appropriate to require higher destruction efficiencies for all
devices based on unique characteristics of control devices at
some mills.
The EPA maintains that no significant amount of dioxin will
be generated from the combustion of these vent gases. There is
no significant level of chlorine in the pulping vents. The final
rule does not require bleaching vents, which do contain chlorine,
to be routed to a combustion device.
Comment: Three commenters (20,027, 20,070A1, IV-D2-11)
disagreed with the requirement to correct gas concentrations to
3 percent oxygen. Two commenters (20,027, 20,070A1) did not
consider any oxygen correction factor (other than the normal
oxygen content of the gas stream) to be justified for
incineration. One commenter (20,027) stated that normal oxygen
5-17
-------
content was 10 percent. Two commenters (20,070A1, IV-D2-11)
stated that for HVLC streams, the oxygen content can be between
15 to 20 percent prior to incineration and methanol content can
be less than 100 ppm. The commenters (20,070A1, IV-D2-11)
claimed that a correction to 3 percent oxygen could reduce the
20 ppmv standard for such streams to less than 5 ppmv after
correction. One commenter (20,070A1) stated that for low
concentration substances in high volume gas streams, such as
HAP/methanol in brownstock washer hood gases, a 10 percent
correction may be appropriate. The commenter (20,070A1) asserted
that the correction to 3 percent oxygen would be appropriate for
combustion units that efficiently operate at 3 percent oxygen and
are burning other fuels, or in cases where the gas stream has a
high Btu value.
Response: The final rule does not require mills to operate
combustion devices at a specified percent oxygen content. The
oxygen correction factor is used as a means of standardizing
concentration measurements to demonstrate compliance. This
standardization ensures that sources are not complying with the
concentration limit by artificially reducing the concentration by
introducing excess air into the vent stream.
The correction factor at proposal was based on previous EPA
studies for other industries. The EPA has re-evaluated the
3 percent correction factor to ensure that it is appropriate for
the pulp and paper industry. Based on thermodynamic calculations
of excess air and flame temperature relations, EPA has decided to
change the oxygen correction factor to 10 percent in the final
rule (A-92-40, IV-B-19). Therefore, the final rule allows
5-18
-------
combustion devices to be in compliance if they reduce HAP
concentrations to 20 ppmv at 10 percent oxygen.
5.1.4 Enclosures and Gas Collection Systems
Comment: One commenter (20,027) pointed out that the
assumed closed vent system requirements for the pulping area are
not practiced at any existing mill. The commenter stressed that
brownstock washers could not be tightly sealed due to the need
for frequent quality control sampling of brownstock. The
commenter (20,027) reported that EPA overestimated the extent to
which a brownstock washer can be enclosed and the amount of gas
flow that will be conveyed to a combustion device.
Response: Information received from an industry survey
(A-92-40, II-D-27) shows that several pulp mills have
successfully enclosed brown stock washers (A-92-40, IV-B-8,
IV-B-16) . Based on this information, EPA has decided to keep the
brownstock washer enclosure requirements in the final rule. The
EPA does not intend to prevent pulp sampling activities with the
enclosure requirement. Mills which have successfully enclosed
brownstock washers have access areas to allow for pulp sampling.
At mills with negative pressure enclosures, access areas do not
present emission leak concerns; however, access areas on positive
pressure enclosures will still have to pass the leak test
requirements .
Comment: One commenter (20,115A2) claimed that there is no
way to enclose the inlet to the lime kiln and simultaneously
direct air through the product coolers, a necessary conservation
step, while allowing the free flow of product lime out of the
kiln.
5-19
-------
Response: The commenter appears to have misinterpreted the
enclosure requirement. Under this regulation, the lime kiln is
not part of the pulping component and, therefore, not required to
be enclosed. The lime kiln was listed as an alternative control
device for vent stream gases in the pulping system.
Comment: One commenter (20,110) supported extending the gas
collection techniques required for kraft mills to the non-kraft
sector of the industry, particularly to control the potential for
heavy sulfur dioxide emissions during batch digester blowdowns.
Response: The proposed rule requires that kraft, soda,
semi-chemical, and sulfite mills comply with the gas collection
requirements. Although EPA subcategorized mills by pulp type for
the final rule, the gas collection and control requirements are
still applicable to all sources that are required to be
controlled, which includes batch digesters at sulfite mills.
5.2 COSTS
Comment: Two commenters (20,000, 20,070A1) contended that
the proposed incinerator design criteria would increase costs
unreasonably. One commenter (20,000) contended that proposed
design criteria would require existing incinerators to be
increased in size by a factor of five, thereby limiting
improvements to existing methods, precluding development of other
technologies to control HAP's, and adding unnecessary costs to
incinerators. Another commenter (20,070A1) stated that the
proposed design standards would require higher capital costs and
fuel costs.
Response: The final rule does not limit owners or operators
of incinerators to operate at the specified temperatures and
residence times. The final rule allows any combustion device
5-20
-------
that can be demonstrated to achieve 98 percent destruction of
HAP' s or any thermal oxidizer that reduces HAP emissions to a
concentration of 20 ppmv at 10 percent oxygen. Information
supplied by the pulp and paper industry shows that many of the
existing thermal oxidizers can meet either the 98 percent
destruction option or the 20 ppmv outlet concentration (A-92-40,
IV-J-33). Therefore, no additional cost or design is necessary
for these thermal oxidizers.
The capacity concerns expressed by several commenters to the
proposed rule may have been due to the fact that the proposed
rule required all vent streams to be controlled, except those
with very low flow and low concentrations. Therefore, the
commenters may have concluded that all pulping vents were
required to be controlled. The March 8, 1996 supplemental notice
specifically defined the vent streams that are required to be
controlled. The specific list will reduce initial estimates of
gas volume routed for combustion.
Additionally, EPA conducted surveys (A-92-40, IV-E-93) with
several mills regarding their capacity to combust additional
vents streams at existing boilers. Results of the survey
indicate that two-thirds of the surveyed mills have the capacity
in the existing boilers to handle combustion of the named HVLC
streams, and therefore, would not need to construct thermal
oxidizers. The remaining one-third would construct thermal
oxidizers. This information was used in determining the national
cost impacts of the final rule (see chapter 20 of this document
and A-92-40, IV-B-13).
Comment: Several commenters (20,011, 20,014, 20,027,
20,071, 20,123A1) claimed that EPA had significantly
5-21
-------
underestimated the cost of controlling pulping emissions and
should reassess the cost for the promulgation package. Several
commenters (20,014, 20,027, 20,057A2, 20,123A1, 20,123A7)
asserted that the costs for gas collection and treatment systems
were significantly underestimated. Three commenters (20,014,
20,027, 20,123A7) specified that EPA had underestimated the cost
of ductwork. These commenters (20,014, 20,027, 20,123A7)
asserted that using stainless steel ductwork was standard for the
industry and EPA underestimated the cost by not assuming the use
of stainless steel duct work. Three commenters (20,027,
20,057A2, 20,118) contended that EPA incorrectly predicted that
there would be no fuel penalty for combusting HVLC vent streams
and did not include the costs of control valves for gas
collection systems which are essential for pulp and paper
operations.
Response: In response to public comments, the cost
estimates for the gas collection systems were revised. The gas
collection system used in estimating cost impacts of the final
rule included stainless steel ductwork, fans, a mist eliminator,
a condenser, flame arresters, liquid sampling taps, a condensate
storage tank, and rupture disks. Additional equipment (mist
eliminator, condenser, condensate storage tank) was added for
reducing moisture in the vent stream prior to combustion
(A-92-40, IV-B-13) .
The revised cost impacts also included the cost of
1,500 feet of stainless steel ductwork to an existing combustion
device and 500 feet to a new stand-alone combustion device. The
duct lengths were based on engineering judgement and information
collected during several site visits to pulp and paper mills.
5-22
-------
The EPA considers these duct lengths to sufficiently characterize
the lengths of ducts at typical pulp and paper mills. Energy and
auxiliary fuel costs were also accounted for in the revised
impacts.
The costs and equipment designs were based on algorithms in
the Office of Air Quality, Planning and Standards (OAQPS) Control
Cost Manual (OCCM) . These algorithms have been widely used in
developing other NESHAP, and have been used by Federal, State,
and local air pollution control agencies to estimate costs.
Comment: One commenter (20,123A7) stated that EPA's
assumption that indirect costs are included in the vendor's cost
estimate may be erroneous. The commenter (20,123A7) provided
cost factors for installed cost and asserted that indirect costs
are generally 35 to 50 percent of the total installed cost
depending on the client and the type of estimate.
Response: For consistency, EPA estimates pollution control
costs using standardized cost procedures specified in the OCCM.
The EPA realizes that estimates made through the OAQPS Control
Costs Manual may be different from actual costs at individual
facilities. However, the procedure provides a reasonable
estimate of control costs on a national basis. Indirect costs
were not assumed to be included in the vendor's cost. Indirect
costs are equipment specific and range from 30 to 57 percent of
the total direct costs.
5.3 SECONDARY IMPACTS OF PULPING CONTROLS
Comment: Several commenters (20,053A1, 20,057A2, 20,059,
20,070A1, 20,103, 20,114) opposed combustion as a means of
controlling HAP's because the secondary emissions of other
pollutants would increase. Several commenters (20,018, 20,053A1,
5-23
-------
20,057A2, 20,059, 20,103) stated that EPA's approach to control
HAP' s by combustion would increase emissions of criteria
pollutants regulated under the Act. One commenter (20,114)
specifically noted that combustion of NCG containing reduced
sulfur compounds would result in increased emissions of sulfur
dioxide. Another commenter (20,010) supported using incineration
to control emissions from pulping vents. However, the commenter
(20,010) insisted that no exceedances of the existing sulfur
dioxide emissions standard should be allowed from incinerating
pulping off-gases.
One commenter (20,114) asserted that the costs and
environmental impacts associated with the by-products resulting
from combustion-based control equipment must be further assessed.
The commenter (20,114) argued that the economics assessment
erroneously indicated that the adverse effects of secondary
impact increases cannot be quantified. The commenter (20,114)
stated that EPA must explain such statements when combustion by-
products are criteria pollutants for which there are existing
ambient air quality standards. One commenter (20,103) questioned
whether tradeoffs between HAP's and criteria pollutants had even
been considered.
Response: In the final economics assessment, (A-92-40,
V-A-2), EPA outlined the estimated emission reductions resulting
from this rule (139,000 Mg/yr HAP's, 409,000 Mg/yr VOC's,
78,500 Mg/yr TRS compounds). Estimated increases in secondary
emissions due to using combustion sources for HAP control were
also presented (94,500 Mg/yr S02, 5,230 Mg/yr NOX,
8,660 Mg/yr CO). The EPA judged that the secondary impacts, due
to the use of combustion control for HAP's, were reasonable and
5-24
-------
were outweighed by the benefits of the HAP emission reductions
achieved. The readers are referred to the economics assessment
for a detailed discussion of the benefits analysis.
Comment: One commenter (20,053A1) contended that
incineration of HVLC streams would increase energy requirements
because HVLC streams do not contain compounds with sufficient
heating value to support combustion. The commenter (20,053A1)
added that increased energy consumption would have adverse
environmental effects and increase mill operating costs. The
commenter (20,053A1) concluded that these impacts were not
incorporated into EPA's analysis.
Response: The EPA agrees that some of the named vent
streams will not have sufficient fuel value to support
combustion. The environmental and cost impacts for the final
rule were revised to include increases in fuel and energy
requirements (see chapter 20 of this document).
Comment: One commenter (20,011) contended that the
secondary impacts (such as energy use) from back-up incinerators
were not assessed in the proposal. The commenter (20,011) stated
that these impacts should be addressed in the promulgation
package.
Response: Impacts were calculated assuming that the primary
combustion device was operating 100 percent of the time. The EPA
considers that if the primary device was down and a backup device
was used, similar energy and emissions would occur. Therefore,
the impacts from backup devices have implicitly been incorporated
into the impacts analysis.
Comment: One commenter (20,122) requested that EPA clarify
whether chlorinated sludge may be incinerated. The commenter
5-25
-------
(20,122) stressed that it is necessary to prevent the
incineration of chlorinated compounds that form dioxins, furans,
and other organochlorines.
Response: This NESHAP does not require the incineration of
chlorinated sludge. Solid waste handling is addressed under RCRA
regulations, and it is not in the scope of MACT regulations.
Comment: One commenter (20,027) stated that the economics
of combusting HAP' s in a boiler with an SC>2 scrubber are less
favorable than combustion in a stand-alone combustion unit
because the flue gas volume of a boiler is larger, which makes
the cost of control greater.
Response: While the rule does not require S02 control on
the boilers or incinerators used for HAP control, some existing
boilers already have SC>2 control. To avoid adverse air quality
impacts locally, some of these facilities may be required to
install SC>2 control on the boilers or incinerators that are used
for HAP control. The EPA agrees that SC>2 control costs on an
incinerator may be less than SC>2 control costs on a boiler due to
differences in flue gas volume. However, the rule allows a
facility to control HAP' s in a boiler or in an incinerator, and
the facility can make the control option decision based on their
preference.
The interaction between the NESHAP and other existing
regulations associated with collateral emissions increases are
addressed in chapter 16 of this document.
5-26
-------
6.0 BLEACHING AREA
6.1 CONTROL OPTIONS
6.1.1 Scrubbers
Comment: Several commenters (20,027, 20,029, 20,054A2,
20,074, 20,118, 20,149) discussed the chlorine removal efficiency
of existing bleach plant scrubbers. One commenter (20,027)
argued that there were no data to support EPA's assumption that
scrubbers can control 99 percent of chlorine emissions from the
bleach plant on a continuous basis. The commenter (20,027)
objected that EPA used NCASI Technical Bulletin No. 616 as the
basis for specifying 99 percent removal because the data in that
bulletin indicated that scrubbers occasionally attained the
99 percent control level, not continuously. Another commenter
(20,074) suggested a continuous chlorine/C102 removal efficiency
of 95 percent rather than the proposed value of 99 percent, and
provided data on three existing bleach plant scrubbers to justify
the recommendation.
Two commenters (20,118, 20,054A2) stated that removing
99 percent of chlorine emissions may not be possible at mills
with 100 percent 2 substitution due to low concentrations of
chlorine that would enter the scrubber.
One commenter (20,029) supported EPA's assumption that the
bleach plant chlorine scrubber could remove 99 percent of
6-1
-------
chlorine and ClO2emissions. The commenter (20,029) acknowledged
scrubbers achieving as high as 99.9 percent control efficiency.
Response: As discussed in the March 8, 1996 Federal
Register supplemental notice, data available to EPA (A-92-40,
11-1-24) support the 99 percent chlorinated HAP reduction
requirement for existing bleach plant scrubbers that use a
caustic medium and operate with high recirculation rates. The
EPA reviewed the data submitted by commenter 20,074 and found
three of the six bleach plant scrubbers analyzed by the commenter
had estimated chlorine/ClC>2 reductions of 99 percent with the
remaining three dropping off to a low of 50 percent. The three
scrubbers that did not achieve 99 percent reduction were not
caustic scrubbers and, therefore, not representative of MACT.
Moreover, the data received from commenter 20,074 was not used to
help justify the final rule because the reported reductions were
estimated and not based on test results.
The EPA agrees that a continuous 99 percent reduction of
chlorinated HAP's across a caustic scrubber may not be achievable
on streams with low concentration levels of chlorine, such as
f*l o
when 2 substitution is used. Therefore, in the final rule,
EPA has included a chlorinated HAP (other than chloroform) outlet
concentration limit of 10 ppmv and a chlorine mass emission limit
of 0.001 kg total chlorinated HAP (other than chloroform) per Mg
ODP produced as alternate compliance options. The EPA has
concluded that these compliance alternatives will achieve
chlorinated HAP reductions (other than chloroform) equivalent to
the 99 percent reduction standard (A-92-40, IV-B-29).
Comment: In regard to the December 17, 1993 proposal,
several commenters (20,024, 20,028, 20,036A1, 20,043, 20,057A2,
6-2
-------
20,071, 20,074, 20,111, 20,114) stated that EPA incorrectly
determined the level of control achieved by bleach plant
scrubbers for HAP's other than chlorine. Several commenters
(20,043, 20,057A2, 20,071, 20,074, 20,111, 20,114, 20,146)
contended that data indicated that bleach plant scrubbers are
ineffective at removing methanol from vent gases. One commenter
(20,027) provided data to demonstrate that methanol and other
non-chlorine HAP's are not removed by existing bleach plant
scrubbers. In addition, the commenter (20,027) also presented
data on the amount of methanol released from existing bleach
plant scrubbers. Another commenter (20,028) reported a methanol
removal efficiency of 40 percent with their existing bleach plant
scrubber (using weak wash as the scrubbing medium).
One commenter (20,027) pointed out that bleach plant
scrubbers typically use alkaline media rather than water as the
scrubbing fluid. Two commenters (20,027, 20,071) stated that a
second scrubber utilizing water as the scrubbing media could be
installed to remove methanol; one commenter (20,027) provided
cost information, but stated that such an option would not be
cost effective.
In response to the proposed use of incineration following
scrubbers, three commenters (20,057A2, 20,114, 20,146) claimed
that incineration following bleach plant scrubbers would not be
cost effective.
In response to the March 8, 1996 Federal Register
supplemental notice, several commenters (IV-D2-8, IV-D2-10,
IV-D2-15, IV-D2-16, IV-D2-17) agreed with EPA's decision to drop
the control of non-chlorinated HAP's from the bleaching control
requirements. However, one commenter (IV-D2-4) expressed concern
6-3
-------
over EPA's decision to only require chlorinated HAP control for
the bleaching area:
Response: As discussed in the March 8, 1996 Federal
Register supplemental notice, the proposed requirement for
control of non-chlorinated HAP's from the bleaching system was
dropped for the final rule. Evaluation of existing bleach plant
scrubber performance data provided by industry indicated that
existing scrubbers are not effective at removing non-chlorinated
HAP'S. Other control scenarios, such as incineration followed by
caustic scrubbing or a second scrubber for the control of
methanol, were evaluated for reducing non-chlorinated HAP's, but
EPA determined that these control techniques were cost
prohibitive or had adverse environmental impacts (see proposal
preamble). For example, the cost of adding a second scrubber for
the control of methanol would not be reasonable considering the
relatively low emission reduction achieved and the significant
increased generation of process wastewater requiring treatment.
Comment: Two commenters (20,059, 20,091) claimed that the
December 17, 1993 proposal may promote cross-media transfers of
certain pollutants. For example, pollutants absorbed in
scrubbing liquid may be released back to the atmosphere from the
wastewater treatment process. One commenter (20,059) also
claimed that although scrubbers may effectively remove 99 percent
of highly soluble compounds, the overall reduction of these
compounds is estimated at only 75 percent because these
pollutants re-volatize in the wastewater treatment process.
Similarly, the commenter (20,059) stated that scrubbers remove
60 percent of medium solubility compounds but the actual removal
is only 35 percent because of re-volatilization.
6-4
-------
Another commenter (20,091) expressed concern over emissions
of bleach plant pollutants from spent white liquor that has been
used as scrubbing media in the bleach plant. The commenter
(20,091) questioned whether the spent scrubbing media would be
sent to the white liquor storage tank where bleach plant
pollutants could be emitted to the atmosphere.
Response: Based on an evaluation of the industry data
collected in the 1992 NCASI voluntary questionnaire, the floor
level of control for bleach plant vents was determined to be
control of total chlorinated HAP's (other than chloroform) using
caustic scrubbing and process modifications (elimination of
hypochlorite and 100 percent substitution of chlorine). Bleach
plant scrubbers typically use white liquor as a scrubbing media
to remove chlorinated compounds. Data indicate that caustic
scrubbing can control the emissions of chlorine by 99 percent or
below 10 ppmv or to an emission limit of 0.001 kg per Mg ODP
(A-92-40, IV-B-29) . Re-volatilization of chlorinated HAP's is
not considered a concern because chlorinated compounds react with
the scrubber media to form a precipitate and would not be emitted
from the spent caustic.
Typical industry practice for spent scrubbing media is to
send the stream to the sewer followed by biological treatment.
Other control devices were evaluated for reducing non-chlorinated
HAP's but EPA determined that these control techniques were cost
prohibitive. The EPA does not have data on chlorinated HAP
emissions from white liquor storage tanks, but maintains that
they are insignificant due to the reactions between the
chlorinated compounds and the white liquor that form a
precipitate.
6-5
-------
6.1.2 Incineration
Comment: With regard to the December 17, 1993 proposal,
several commenters (20,027, 20,059, 20,071, 20,114) expressed
concerns over the potential requirement for incineration of
emissions from bleach plant vents. Three commenters (20,027,
20,071, 20,114) maintained that the incineration of emissions
from bleach plant scrubber overheads would not be cost effective
due to energy penalties and large increases in collateral
emissions. The commenters (20,027, 20,071, 20,114) also stated
that this option would not be legally defensible because the
cost-effectiveness of such a requirement could not be justified.
One commenter (20,027) stated that existing combustion
devices could not be used to combust emissions from scrubber
overhead vents because there would be periodic exposure to
chlorinated streams if the scrubber experienced downtime, or
continual exposure if the combustion device were installed prior
to scrubbing. Therefore, the commenter (20,027) concluded that a
stand-alone incinerator would have to be installed after the
scrubber. The commenter (20,027) indicated additional fuel would
be required for the incinerator, leading to collateral emission
increases and energy penalties.
One commenter (20,059) supported a control option that would
require combustion of scrubber off-gases to remove insoluble
organic compounds such as chloroform. The commenter (20,059)
contended that some pollutants in the bleaching vent streams
would not be effectively reduced with a scrubber alone. The
commenter (20,059) also asserted that EPA did not consider
intermediate options, such as the combustion of selected vent
streams with high organic content followed by scrubbing.
6-6
-------
Response: As presented in section X.E.2 of the preamble to
the December 17, 1993 proposal, EPA considered two options for
incinerating bleach plant vent gases: (1) combustion of scrubber
off-gases, and (2) combustion of emissions from bleach plant
vents followed by scrubbing. However, EPA rejected these options
because they were not reasonable considering the cost and
environmental impacts. Cost data and comments received after
proposal supported the conclusion that incineration of bleach
plant vent gases is not a viable option (20,027) .
The final rule requires caustic scrubbing for total
chlorinated HAP's, other than chloroform, for both paper-grade
and dissolving-grade mills. For paper-grade mills, the final
rule requires process modifications (e.g., 100 percent C1C>2
substitution and elimination of hypochlorite) for chloroform
emissions. The chloroform requirements for dissolving-grade
mills are still under study and are being deferred. As indicated
in the March 8, 1996 supplemental notice, compliance with the
requirements of the effluent limitation guidelines and standards
will constitute compliance with the air standards for chloroform
emissions. The EPA recognizes that these requirements do not
provide efficient control of non-chlorinated HAP's. However,
additional technologies that could reduce non-chlorinated HAP's,
such as incineration, were determined to not be reasonable based
on their cost and impacts (see proposal preamble).
6.1.3 Process Modifications
Comment: One commenter (20,027) claimed that EPA did not
systematically review industry data to predict the emission
reductions of HAP's other than chloroform from process changes
(e.g., peroxide reinforcement of the second bleaching stage,
6-7
-------
installation of oxygen delignification, extended digester
delignification, improved brownstock washing, etc.). The
commenter (20,027) concluded that, based on the ineffectiveness
of scrubbers in controlling some species of HAP emissions, it
would not be appropriate to include gas scrubbing in the
development of the MACT floor level of control. However, the
commenter (20,027) suggested that it may be appropriate to set
emission limits based on reductions obtained through process
changes.
Several commenters (20,054A2, 20,059, 20,071, 20,102,
20,144) suggested that the elimination of hypochlorite and
100 percent substitution of C1C>2 would reduce emissions of
chloroform and other HAP's. Therefore, the commenters (20,054A2,
20,071) indicated that scrubbers would not be needed for HAP
control.
Response: The EPA evaluated the industry test reports,
including the NCASI report, and prepared a document that
summarizes the calculated emission factors (A-92-40, IV-A-6).
Evaluation of these data indicate that high levels of „, _
C1U2
substitution and elimination of hypochlorite reduce chloroform
emissions, but may not significantly reduce other non-chlorinated
HAP'S.
Information obtained from the 1992 NCASI voluntary
questionnaire (i.e., the MACT survey) indicated that the floor
level of control for bleach plants is caustic scrubbing and
process modifications (100 percent substitution and elimination
of hypochlorite) to control chloroform emissions. The effluent
limitation guidelines and standards will require (as a minimum)
100 percent 2 substitution and no hypochlorite use. The EPA
-------
considers these requirements to be at least as stringent as the
process modifications in the floor level of control. Therefore,
as explained in detail in the March 8, 1996 Federal Register
supplemental notice, the final rule for bleach systems is a
combination of the effluent limitation guidelines and standards
requirements (or MACT process modification requirements) and the
bleach system vent scrubber requirements outlined in the proposed
rule.
Comment: One commenter (20,059) recommended that EPA
examine whether 100 percent C102 substitution for dissolving
kraft mills, as opposed to the 70 percent substitution
recommended at proposal, could achieve greater reductions in
chloroform and other HAP emissions. The commenter (20,059)
asserted that if it is more effective, then EPA should require
100 percent substitution combined with oxygen delignification as
the basis for controls in the dissolving kraft industry.
Response: The final rule contains standards for total
chlorinated HAP's, other than chloroform, based on scrubbing and
process modifications; either elimination of hypochlorite and
100 percent C102 substitution or the effluent limitation
guidelines and standards for paper-grade and dissolving-grade
mills. However, as stated in OW s July 15, 1996 Federal Register
notice, EPA is deferring issuing effluent limitation guidelines
and standards for dissolving-grade mills until the comments and
preliminary new data affecting dissolving-grade subcategories can
be fully evaluated. Therefore, the compliance date for
dissolving-grade mills to comply with the bleaching system
standards has been delayed until the 3 years after the effluent
limitation guidelines and standards are promulgated.
6-9
-------
Oxygen delignification is not being required as MACT because
it would not achieve significant additional air emission
reductions in the bleaching system beyond those achieved by the
final rule, and would increase emissions of HAP from the pulping
area.
Comment: Several commenters (20,036A1, 20,059, 20,110,
20,121) stated that TCF bleach plants should be exempt from the
proposed bleaching area control requirements. One commenter
(20,036A1) agreed with the requirement for enclosures in the
bleaching area if a mill uses a combination of chlorine and C1C>2
to bleach; but they argued if a mill uses a TCF bleaching
process, enclosures should not be required.
One commenter (20,036A1) requested that the requirements for
scrubbers to control chlorine/C102 emissions be eliminated if TCF
bleaching is implemented.
Response: In the December 17, 1993 proposed rule, TCF
bleach systems would have fallen under the pulping component
definition, and therefore, would be required to control all
vents. In the final rule, TCF systems are included in the
bleaching system. Bleaching systems are only required to control
total chlorinated HAP's. Since TCF systems do not use
chlorinated HAP's, TCF bleach systems are not required to be
controlled.
Comment: One commenter (20,138) requested that EPA ban the
use of chlorine to bleach pulp, and require the use of oxygen
delignification. The commenter (20,138) indicated that C102
substitution may not be good enough. One commenter (20,110)
objected to the requirement for ClOg substitution since it has a
greater inhalation toxicity than chlorine and Occupational Safety
6-10
-------
and Health Administration (OSHA)/National Institute Occupational
Safety and Health (NIOSH) exposure limits are one fifth those of
chlorine.
Response: Based on an evaluation of industry data collected
in the 1992 NCASI voluntary survey (A-92-40, IV-B-8), and data
received after proposal, EPA determined that the MACT floor for
bleach plant vents is scrubbing of total chlorinated HAP's, other
than chloroform, 100 percent substitution with ClC>2» and
elimination of hypochlorite. Therefore, substitution is required
in the final rule. The data also show that bleaching
substitution with C1C>2 reduces chloroform and other chlorinated
HAP emissions. However, the data indicate that there are no
significant increases in non-chlorinated HAP emissions. Readers
are referred to the revised emission factor document (A-92-40,
IV-A-8) for a more detailed discussion of this issue. While C±O2
is a highly toxic compound, mills that use C1O2 substitution also
have scrubbers in place to control C102 because of worker safety
concerns.
6.2 MISCELLANEOUS BLEACHING COMMENTS
Comment: One commenter (20,029) advocated the possible use
of catalytic oxidation units with appropriately formulated
catalysts to control bleaching component halogenated VOC, citing
their effectiveness in other industrial applications.
Response: A discussion of the rationale for determining the
level of control for the bleach plant is presented in chapter 4.0
of this background information document. Other control devices
than those specified in the rule can be used to meet the
performance standard in place of the reference control
technology. However, the alternate control device will still be
6-11
-------
required to meet the percent reduction or concentration limits
for chlorinated HAP's specified in the rule.
6-12
-------
7.0 PROCESS WASTEWATER AREA
7.1 DESIGN STEAM STRIPPER
Comment: Several commenters (20,000, 20,027, 20,059,
20,118, 20,147) contended that various aspects of the design
specifications for the steam stripper required by the proposed
rule were incorrect, and that some assumptions used in modeling
the steam stripper control of process wastewater streams were
incorrect. Two commenters (20,000, 20,147) reported that the
specified steam stripper design would not achieve the desired
results because the pressure was too high for the number of
theoretical stages assumed. Two commenters (20,027, 20,118)
criticized the use of a tray efficiency of 75 percent, stating
that industry typically assumes a tray efficiency of 50 percent.
The commenters (20,027, 20,118) also stated that the proposed
inlet concentration for streams to be stripped was too high.
Two commenters (20,027, 20,118) claimed that the analysis
for the proposed steam stripper appeared to be based on stripping
compounds (e.g., butadiene, toluene, naphthalene, and butanol)
unrelated to the pulp and paper industry. One commenter (20,027)
contended that these compounds do not possess the hydrophilic
properties of methanol. The commenter (20,027) also criticized
the use of Henry's law instead of empirical data to predict
liquid-gas interface relationships. The commenter (20,027)
7-1
-------
concluded that the result was that the stripper design was
undersized.
One commenter (20,147) strongly criticized the design steam
stripper parameters and indicated that EPA should simply
designate what streams require stripping and at what efficiency
and leave the design of the system to the experts.
Response: Based on comments received on the December 17,
1993 proposed rule and the March 8, 1996 supplemental notice, the
requirements for a design steam stripper were removed from the
final rule. Industry indicated that a design stripper would
likely not be used. At proposal, industry commented that a mass
removal target would be a more usable option. Prior to proposal,
EPA lacked the data necessary to establish a mass removal target.
However, additional data was submitted to EPA following the
proposal and was used to determine the appropriate mass removal
target (20,027A3). A discussion of this data was presented in
the March 8, 1996 supplemental notice. Based on this data, EPA
determined that the mass removal target is achievable (A-92-40,
IV-B-10). Additionally, information received from industry about
specific design assumptions were considered in determining steam
stripper costs (see section 7.6).
7.2 APPLICABILITY CUTOFFS
Comment: One commenter (20,027) advised that the proposed
rule (with the 500 ppmw cutoff) would require stripping of
streams with flow rates of 1,000 to 1,500 gallons per minute
(GPM) . The commenter (20,027) pointed out that no mill is known
to operate a steam stripper capable of stripping streams with
this high flow rate. The commenter (20,027) stated that the
largest steam stripper used in the industry strips 500 GPM. The
7-2
-------
commenter (20,027) asserted that stripping at the levels proposed
by EPA would require stand-alone steam strippers (A-92-40,
II-B-20). One commenter (20,043) indicated that their steam
stripper would be able to achieve about 90 percent removal of
methanol if the applicability level were raised to 3,000 ppmw so
that the large volume, dilute flows would not need to be
stripped. However, one commenter (20,059) alleged that EPA
relaxed the applicability cutoff in the final rule from 100 ppmw
to 500 ppmw due to objections from industry and Office of
Management and Budget (OMB). The commenter (20,059) argued that
the cutoff limit for steam stripping wastewater (500 ppmw
methanol) will virtually exempt all wastewater streams from steam
stripping, and consequently, from control of atmospheric VOC
emissions.
Two commenters (20,027, 20,118) asserted that steam
stripping of streams that contain black liquor (even in dilute
amounts) is not feasible due to foaming problems. As a solution,
the commenters (20,027, 20,118) proposed the restriction of the
definition of process wastewater streams to include only certain
defined streams that do not come in contact with black liquor.
This would allow condensates contaminated with liquor (carryover
or spilled liquor) to be discharged to the wastewater treatment
system.
Response: At the time of the proposal, EPA did not have
sufficient data to identify the specific streams that are
typically steam stripped. The applicability cutoffs were
developed to distinguish between those streams that were steam
stripped at the floor level of control and those that were not
stripped. Following proposal, additional data was submitted to
7-3
-------
EPA (A-92-40, IV-J-32) that identified the specific streams that
are steam stripped and streams that are uncontrolled at the floor
level of control. Control more stringent than the floor level is
not warranted given the small concentrations of HAP's and the
costs of controls. As discussed in the March 8, 1996
supplemental notice, EPA revised the format of the final rule for
kraft and sulfite wastewater by replacing the applicability
cutoffs with named pulping process condensate streams to be
controlled. The EPA contends that the revised approach contained
in the final rule more accurately reflects the floor level of
control than the applicability cutoffs in the proposed rule.
With regard to the commenter's concern of stripping streams
that contain black liquor, EPA does not have sufficient data to
assess the magnitude of this problem. However, EPA contends that
periods when condensates are untreatable should be addressed with
the downtime allowances for stand-alone and integrated steam
strippers specified in the final rule. Further discussion of
downtime issues is given in section 4.3.4.
7.3 HARDPIPING AND BIOLOGICAL TREATMENT
Comment: In regard to the EPA's request for comments and
data at proposal, several commenters (20,027, 20,039, 20,054A2,
20,067, 20,074, 20,111) supported biological treatment for the
control of HAP's from wastewater. One commenter (20,054A2)
stated that if other technologies, old or new, could achieve the
same level of reduction, they should be allowed as alternatives.
Two commenters (20,039, 20,067) stated that by not allowing
biological treatment as an alternative to steam stripping, money
and efforts would be wasted for no environmental benefit. One
commenter (20,111) reported that test results obtained from a
7-4
-------
kraft mill have shown non-detectable levels of methanol in both
the sludge and the effluent to the river, indicating that a
properly operated biological treatment system would be a better
environmental alternative than steam stripping.
One commenter (20,011) asserted that methanol's high
solubility, low volatility, and affinity for biological
destruction were not considered when EPA chose steam stripping as
the reference control technology over biological treatment. The
commenter (20,011) questioned why reference control technology
criteria for biological treatment systems have not been
established despite data demonstrating their efficient
destruction of HAP's. Two other commenters (20,027, 20,115A2)
also indicated that methanol is readily destroyed by biological
treatment systems. One commenter (20,043) noted that the
chemical characteristics of the HAP's found in pulp mill
wastewater streams are correctly described as polar and these
HAP's are not likely to volatilize readily.
Response: In both the December 17, 1993 proposal and the
March 8, 1996 supplemental notice, EPA stated that steam
stripping of pulping process condensate streams constitutes the
floor level of control for kraft pulping. The EPA asserts that
the MACT standards address total HAP emissions, not just
emissions of methanol. While EPA agrees with the commenters that
methanol is readily degraded in well-operated biological
treatment systems, information detailing the overall
effectiveness for destroying total HAP compounds is not
available.
Based on the hydrophilic nature of methanol, EPA believes
that a steam stripper removing 92 percent of methanol is
7-5
-------
achieving a substantially greater removal of total HAP
(i.e., 92 percent removal of methanol in a steam stripper
correlates to at least 92 percent removal of total HAP) . This is
not the case for biological treatment systems since methanol is
preferentially degraded over other HAP compounds
(i.e., 92 percent removal of methanol in biological treatment
does not necessarily correlate to 92 percent removal of total
HAP) . While EPA has limited data indicating that some well-
operated biological treatment systems could meet the standard
(A-92-40, IV-D1-75), EPA does not have sufficient data regarding
total HAP removal to base the floor level of control on
biological treatment.
In the proposal and final rule, methanol was selected as the
surrogate compound for measuring total HAP for most control
devices since it is the predominant compound in process vent and
wastewater streams. However, the final rule specifies that
compliance with the percent reduction standard must be
demonstrated on a total HAP basis if a biological treatment
system is used to comply with the pulping process condensate
standard.
Although EPA based the floor level of control on the
performance of steam stripping technology, the final rule
contains several compliance options. The options include
discharging condensates to a biological treatment system
achieving 92 percent destruction of HAP, recycling the pulping
process condensate streams to a piece of process equipment that
is controlled according to the pulping vent standard, achieving a
specified percent mass reduction, and achieving minimum mass
removal targets. Any HAP removed during handling and treatment,
7-6
-------
with the exception of biological treatment, must be controlled
according to the pulping vent standard.
Comment: One commenter (20,027) questioned whether
sufficient data were available to determine biological
treatment's destruction efficiency of HAP's and cautioned that
neither they nor EPA have sufficient information to predict what
the target treatment efficiency of "well operated treatment
systems" are in general. The commenter (20,027) stated that the
proposed effluent guidelines have the potential to reduce the BOD
loadings to the biological treatment system and, consequently, to
change the removal efficiency of the system. One commenter
(20,067) reported a HAP removal efficiency of 98 percent by
biological treatment systems.
Response: At proposal, EPA did not have sufficient data to
characterize the total HAP removal efficiency of biological
treatment systems. Following proposal, industry submitted data
(A-92-40, IV-D1-75) that detailed the removal efficiency of
methanol in biological treatment systems. Based on this data,
EPA concluded that a well operated treatment system can achieve
methanol removal of 98 percent. However, methanol is
preferentially degraded in biological treatment systems over
other HAP compounds. Therefore, the final rule requires owners
and operators using a biological treatment system to comply with
the pulping condensate standard to demonstrate initially and
annually that the system is achieving at least a 92 percent
reduction in total HAP, not just methanol.
Regarding the commenter's (20,027) concern about the effect
of reduced BOD loadings on biological treatment system removal
efficiency, EPA maintains that using a biological treatment
7-7
-------
system to comply with the pulping process condensate standards is
an appropriate option. If for any reason the biological
treatment system cannot be operated to achieve a 92 percent HAP
reduction on a continuous basis, then the biological treatment
option could not be used.
7.4 WASTEWATER COLLECTION AND TRANSPORT SYSTEMS
Comment: Several commenters (20,027, 20,056, 20,059,
20,074, 20,115A2) discussed the enclosure of wastewater
collection and biological treatment units. One commenter
(20,027) stated that the requirement for enclosing and combusting
the wastewater collection and transport system components was
expensive, impractical, completely unreflective of current
practice and of minimal environmental benefit. One commenter
(20,056) supported the statement in the preamble to the proposed
regulation that the release of HAP's from quiescent wastewater
units are less significant than those from turbulent systems.
The commenter (20,056) stated that control of quiescent tanks and
impoundments would not be justified since the dominant HAP is
methanol, which is extremely soluble in water and does not
readily volatilize. Another commenter (20,115A2) stated that it
would not be feasible to incinerate the large volume of air
associated with an enclosed biological treatment system. One
commenter (20,074) stated that enclosing the collection system
until biological treatment is not needed to provide equivalence
to steam stripping. The commenter (20,074) noted that steam
strippers and incinerators are far more susceptible than
wastewater plants to periods of excess emissions during startups,
shutdowns, and malfunctions.
-------
One commenter (20,059) stated that EPA should evaluate
emissions from quiescent basins and biological treatment systems
and require them to be covered and vented to a control device.
The commenter (20,059) indicated that volatile toxics evaporate
from uncovered segments of wastewater treatment and biological
treatment lagoons. The commenter (20,059) further asserted that
biological treatment of wastewaters is ineffective at controlling
VOC emissions to the atmosphere.
Response: The EPA asserts that it was not the intent of
the biological treatment compliance option specified in the
proposed rule to enclose or cover and incinerate the biological
treatment system emissions. The final rule was revised to make
the requirements for the biological treatment system compliance
option more clear. The EPA agrees with the commenter that
enclosing treatment units and incinerating the emissions would be
very costly for industry and would achieve minimal emissions
reductions. This determination is based on an evaluation of the
emissions from biological treatment units (A-92-40, IV-A-6) and
the fact that biological treatment systems are typically not
located near existing combustion devices (i.e., there would be
costly gas collection/conveyance systems). In addition, no
existing mills currently cover and vent their biological
treatment systems to control devices.
The EPA adopted a similar approach in the recently
promulgated rules controlling air emissions from hazardous waste
surface impoundments that treat volatile hazardous wastes. As
presented in the December 6, 1994 Federal Register notice
(59 FR 62917) and 40 CFR part 265.1086(a), uncovered biological
treatment systems may be utilized to comply with the rule's
7-9
-------
requirements provided that the treatment system achieves the
specified mass removal efficiency.
The proposed requirements for enclosing the wastewater
collection system were intended to prevent the volatilization of
HAP compounds from the wastewater streams before the streams
arrived at the treatment device (e.g., steam stripper or
biological treatment system). The proposed rule contained
requirements for tanks, containers, surface impoundments, and
individual drain systems. Based on industry comments, the
requirements for containers have been removed from the final rule
since they are not used in the pulp and paper industry.
Additionally the requirements for surface impoundments have been
removed from the final rule since EPA concurs that collecting and
incinerating emissions from these treatment units is not
reasonable.
The final rule retains the requirements for tanks that are
used to store or treat the pulping process condensates. The
specific individual drain system requirements contained in the
proposed rule have been removed in favor of referencing the
individual drain system requirements specified in 40 CFR
subpart RR §§ 63.960, 63.961, 63.962, and 63.964. The EPA
compared the collection system requirements contained in the
proposed rule with the requirements of subpart RR. Since the
requirements are consistent with the intent of the proposed
standards, EPA concluded that the requirements of subpart RR,
when combined with a treatment option, constitute MACT for the
pulp and paper industry.
7-10
-------
7.5 SECONDARY IMPACTS OF WASTEWATER CONTROLS
Comment: Several commenters (20,000, 20,011, 20,018,
20,027, 20,057A2, 20,067, 20,111, 20,115A2) indicated that the
collateral emissions and the waste heat load associated with
steam stripping were not adequately characterized by EPA. Two
commenters (20,027, 20,057A2) asserted that EPA did not consider
that the waste heat from steam strippers would cause water
pollution, and may impact NPDES permits. Several commenters
(20,027, 20,057A2, 20,111) noted that there would be an increase
in NOX, ^®2 i CO, and PM less than 10 microns mean aerodynamic
diameter (PM]_Q) due to the extra energy needs from the proposed
steam stripping option. Two commenters (20,000, 20,067) also
stated that the increased need for steam, normally generated by
burning fossil fuel, could lead to increased carbon dioxide (CO-?)
emissions. Two commenters (20,011, 20,018) stated that the
secondary impacts associated with routing stripper overheads to a
combustion device have not been adequately characterized. One
commenter (20,115A2) argued that incinerating the steam stripper
overheads and discharging clean, hot water was counterproductive
from an energy standpoint while achieving little more reduction
of methanol emissions than biological treatment.
Response: For the proposed and final rules, secondary
impacts (e.g., NOX, SO2 f CO, PM, etc.) were estimated for the
following areas associated with steam stripping: (1) overhead
gas combustion, (2) steam generation, and (3) electricity use.
These secondary impacts are included in the impacts analysis. A
discussion of the analysis for estimating secondary impacts is
presented in chapter 20 of this document. The commenters'
concerns regarding secondary impacts increases were referring to
7-11
-------
the impacts generated by the proposed steam stripper. At
proposal, the flow rate of condensate streams sent to the steam
stripper was estimated to be approximately 1 GPM per ADTP
production per day (GPM/tpd) . Based on data received following
proposal, the flow rate was revised to approximately 0.2 GPM per
ADTP per day. Conseguently, the energy demand required by the
steam stripper and the secondary impacts were proportionately
reduced.
Regarding overall energy concerns, the final rule specifies
that mills can rectify the steam stripper overheads to produce a
concentrated stream to be used as supplemental fuel in mill
combustion devices. This action will substantially reduce the
operating costs associated with steam stripping. Additionally,
the treated condensate from the steam stripper could be used by
mills in pulping process areas to reduce the overall demand for
fresh or mill water. If a mill elects to discharge the treated
stream to the biological treatment system, the contribution of
this stream to the total mill effluent flow rate would be
negligible (i.e., hundreds of gallons compared to millions of
gallons per day).
7.6 COSTS
Comment: Several commenters (20,011, 20,014, 20,018,
20,027, 20,043, 20,057A3, 20,114, 20,118) claimed that EPA
underestimated the cost of steam stripper installation because of
inadequate design and that EPA also overlooked equipment
requirements. Several commenters (20,014, 20,027, 20,057A2,
20,118) maintained that cooling towers will be needed to handle
the increased heat load sent to wastewater treatment systems.
Several commenters (20,011, 20,014, 20,018, 20,027, 20,057A2,
7-12
-------
20,114, 20,118) were concerned that existing steam capacity will
not be adequate for stripping the required streams. The
commenters indicated that the construction of package power
boilers would be required to generate the additional steam
necessary to strip the high volumes of wastewater. One commenter
(20,057A2) argued that the construction of new power boilers or
modifications to existing ones would trigger PSD/NSPS review and
permitting. The commenter (20,057A2) stated that EPA did not
consider the capital costs associated with this need. One
commenter (20,027) noted that the true cost of steam stripping
depends on the water and heat balances at a given mill. The
commenters indicated that these factors were not considered
properly by EPA in developing costs. One commenter (20,014)
reported that the additional steam needed to strip an estimated
1,700 GPM of condensates would cost approximately $3,500,000 per
year.
One commenter (20,027) warned that the conclusion was
incorrect that two-thirds of the industry strippers would be
integrated with evaporators. The commenter (20,027) asserted
that the proportion of integrated versus non-integrated steam
strippers is not the 66/34 percent split (integrated vs. non-
integrated) used by EPA, but closer to a 6/94 percent split (30
of 32 mills are non-integrated) . The commenter (20,027)
disagreed that integrated steam strippers are a viable option
(A-92-40, II-B-28) .
One commenter (20,043) indicated that because pulp mills
generate larger and more dilute wastewater streams than the
chemical industry, it would be cost prohibitive to transfer
chemical manufacturers' wastewater technologies (i.e., steam
7-13
-------
strippers) to the pulp industry. One commenter (20,114) urged
that the high cost of a stand-alone incinerator be considered
since wastewater treatment systems are typically located in
remote areas of the mill.
One commenter (20,027) indicated that the ASPEN model EPA
used to develop steam stripper design, cost, and model plant
parameters is used for simulating packed tower distillation
columns, not steam strippers. Therefore, the commenter asserted
that the model inputs and assumptions used at proposal were not
correct for the pulp and paper industry. The commenter (20,027)
favored projections of cost and performance made on actual
industry data rather than on a predictive model for chemical
industry equipment.
Response: The comments received regarding steam stripper
costs were made in reference to the proposed steam stripper
design. Although the cost estimation of steam stripping systems
is not critical since stripping is a floor-level technology, EPA
revised the design and performance parameters used to estimate
the capital and annual costs associated with steam stripping
(A-92-40, IV-B-17) based on comments and data received following
proposal.
At proposal, the flow rate of condensate streams sent to the
steam stripper was approximately 1 GPM per ADTP per day. Based
on data received following proposal (20,027A3), the flow rate was
revised to approximately 0.2 GPM per ADTP per day. Consequently,
the capital and annual costs associated with steam stripping were
proportionately reduced.
The EPA contends that the ASPEN model provides steam
stripper cost estimates that are comparable to the estimates
7-14
-------
provided by industry. While some mills may encounter higher or
lower capital and annual costs, EPA maintains that the cost
estimates derived from the ASPEN model are appropriate for
estimating the national impacts associate.d with steam stripping.
The costs associated with package boilers for additional
steam capacity were not included in the steam stripper costs.
The EPA maintains that the steam demand for the stripper system
is not expected to be a significant portion of the overall mill
steam generation capacity and that the affected mills will be
able to meet the increased steam demand with existing systems.
Additionally, the steam required for stripping may be generated
from other sources besides fresh steam from power boilers or
recovery furnaces (e.g., flash or waste heat sources).
The costs associated with cooling towers were not included
in the steam stripper costs for the final rule. The EPA reviewed
the data submitted by industry (A-92-40, IV-D1-46) detailing the
number of cooling towers existing in the pulp and paper industry.
The data indicated that 13 cooling towers were being used. While
some mills may need cooling towers to handle the waste heat load
from the pulping and bleaching processes, EPA's judgement was
that it was not appropriate to assign the costs for installing
and operating cooling towers to all mills that would use steam
stripping. This decision was based on the fact that the stripped
pulping condensates are typically sent to the mill's hot water
tank for distribution to other process areas and the contribution
of the stripped condensate is not expected to be significant when
compared to the total volume of mill wastewater sent to the
biological treatment system.
7-15
-------
7.7 OTHER
Comment: One commenter (20,115A2) stated that recycling
pulping process condensates could potentially increase HAP
emissions due to increased carryover to uncontrolled process
equipment such as deckers and screens.
Response: The final rule contains a compliance option for
kraft pulping system wastewaters that allows mills to recycle the
specified pulping process wastewater streams to controlled pieces
of equipment without treatment. Since the piece of equipment
receiving the untreated condensate is controlled according to the
capture and control requirements of the pulping vent standards,
EPA contends that HAP emissions would not be increased.
7-16
-------
.0 MONITORING
8.1 GENERAL
Comment: One commenter (20,059) stated that the monitoring
parameters in the December 17, 1993 proposal were insufficient.
The commenter (20,059) stated that in order to show that an
emission standard is enforceable, EPA must show that the
monitoring standard is sensitive enough to (1) detect
exceedances, (2) indicate the amount of time the source was out
of compliance, (3) show the amount of emissions in excess of the
standard, and (4) identify the pollutants emitted. The commenter
(20,059) stated that monthly measurements would not be sufficient
to track wastewater treatment performance. One commenter
(20,150) requested monitoring equipment capable of detecting any
discharge of organochlorine. The commenter (20,150) also wanted
to disallow hourly averaging. One commenter (20,151) requested
that the rule require monitoring and recordkeeping for the
potential venting of HAP's from all potential discharge
locations.
Response: This rule, as NSPS and NESHAP programs have
traditionally done, requires a combination of performance testing
and continuous monitoring of control device operating parameters
instead of monitoring the actual emission levels. Continuous
-------
parameter monitoring is consistent with section 504(b) of the
Act, which states that "continuous emission monitoring need not
be required if alternative methods are available. . .for
determining compliance." The "alternative method" presented in
the rule is to monitor the control device parameters.
The EPA has concluded in previous standards (e.g., the HON)
that there is sufficient evidence to prove that pollution control
equipment, if operating properly, can achieve high levels of HAP
destruction. Data received from industry indicate that
operational parameters provide an accurate indication of HAP
destruction and emission levels. Operation parameter levels that
ensure compliance are established during the initial performance
testing effort, and continuous monitoring of operating parameters
ensures continued compliance. Continuous emissions monitors for
individual HAP species would add significant costs and burden to
the industry without producing any environmental gain since the
standard is based on total HAP.
The parameter monitoring program contained in the final rule
provides clear criteria for what is considered to be a violation.
With the exception of biological treatment systems, a period of
excess emissions (considered a violation of the standards) occurs
when operating parameters that indicated compliance during the
initial performance tests are exceeded. The monitoring approach
for biological treatment systems is discussed later in this
section.
The final monitoring provisions are sufficient to detect
exceedances and to determine the duration and extent of
non-compliance. Providing a legal basis for effectively
-------
enforcing these emission limits does not require quantification
of specific pollutants and emission levels.
Comment: Three commenters (20,027, 20,036A1, 20,056)
claimed that the monitoring requirements are too burdensome on
the industry. One commenter (20,036A1) stated that once the
required process technology is installed and properly operated in
order to produce the desired pulp quality, the operator has
little or no effect on pollutant discharge. One commenter
(20,056) contended that unless EPA can demonstrate the need for
the proposed inspection schedule, one inspection every 6 months
is appropriate. The commenter (20,056) stated that EPA must
allow a facility the flexibility to determine the appropriate
inspection schedule considering site-specific shutdown schedules,
length of duct work, and history of repairs.
One commenter (20,027) outlined several recommendations for
changes to the proposed monitoring requirements which included
not requiring chlorine monitoring from bleaching systems (since
the effluent guidelines require C1O2 substitution) , exempting
process monitors from the monitoring plan specified in ^ 63.8(b),
specifying a monitoring plan consistent with manufacturer's
recommendation for calibration and maintenance, and allowing a
ten percent deviation range around the initial performance test
operating parameters that determine compliance.
Response: The EPA has made every effort to reduce the
monitoring burden and to require only those procedures that are
necessary to determine continuous compliance. The continuous
monitoring of control device parameters, as required by the
NESHAP, is necessary to provide information that will satisfy the
requirements of section 114 (a) (3) of the Act for enhanced
-------
monitoring, certification of compliance status, and determination
of continuous or intermittent compliance. The EPA considers the
level of monitoring specified in the rule appropriate and
necessary for compliance and disagrees with the commenters'
assertion that the level is unwarranted.
Most pollution control technologies specified in this rule
(i.e., thermal oxidizers, caustic scrubbers, steam strippers, and
closed vent systems) are not related to pulp quality. These
systems are operated separately from the pulping and bleaching
systems and must be operated such that the limits defined in this
rule are met, regardless of what pulping and bleaching process
adjustments need to be made to produce the desired pulp quality.
For process equipment that is used to reduce emissions,
appropriate monitoring parameters are required to be determined
during the performance test. The initial performance test should
be conducted during normal operation of the mill so that the
monitoring parameters determined are indicative of continuous
compliance. Process technologies that are outlined in the rule
as pollution prevention measures (e.g., total chlorine free
bleaching) satisfy the requirements of the rule when properly
operated and no further measures would be needed for compliance.
The EPA contends that the inspection schedule in the final
rule (i.e., monthly) is appropriate for ensuring continuous
compliance and does not place an undue burden on the industry. A
more frequent inspection schedule is not needed since EPA does
not expect the closed-vent systems or closed collection systems
to encounter significant breakdowns or defects that would be
associated with problems that developed over a short period of
time (e.g., one week). However, a less frequent inspection
J-4
-------
schedule is not appropriate since defects or potential problems
would not be identified in a timely manner.
The monitoring requirements specified in § 63.8 (b) will not
be waived for this rule. A 10 percent allowance above the
established operating parameter compliance level is not allowed
in the final rule. With the exception of biological treatment
systems, the operational parameter levels that are determined
during the performance test represent compliance and any
exceedance will be judged as non-compliance. However, in setting
the specified operating parameter level for determining
compliance, a mill will determine this level based on parameter
data monitored during the performance test, supplemented by
engineering assessments and manufacturers' recommendations. The
rationale and supporting information for the selected operating
parameter must be submitted to the Administrator for approval.
Comment: Several commenters (20,011, 20,027, 20,054A2,
IV-D2-15) requested that EPA allow some excursions and exemptions
from monitoring to add flexibility to the parameter monitoring
provisions. One commenter (20,054A2) supported excluding
violations during startups, shutdowns, malfunctions, and during
the first 48 hours in a reporting period. One commenter (20,011)
stated that emergency venting should be an excusable excursion
from otherwise applicable continuous monitoring requirements.
One commenter (20,027) argued that because of process variability
and lack of experience regarding continuous parameter monitoring
systems, EPA must provide some provision for a certain number of
excused excursions per reporting period. The commenter (20,027)
offered to cooperate with EPA on such a project.
-------
Response: With the exception of biological treatment
systems, monitoring excursions or exemptions during normal
operation are not allowed in the rule (other than the allowed
downtime allowances). The monitoring parameters identified
during the initial performance test should be determined in a
manner to account for process variability. If a facility
believes that the initial monitoring parameters do not accurately
demonstrate continuous compliance, the facility may retest,
before any violation of the standard, and revise the monitoring
parameters (i.e, revise their operating permit).
For biological treatment systems, the rule identifies
parameters to be monitored on a daily basis. Daily inlet and
outlet samples must also be collected and archived for 5 days.
The archived samples are used to demonstrate that the biological
treatment system is achieving 92 percent reduction of total HAP
if a specified monitoring parameter is outside the range
established during the initial performance test. Quarterly
performance monitoring for total HAP removal is also required in
the final rule. To reduce the burden of sampling for total HAP
during all four quarters, a mill may (during the first quarter
test) establish a methanol percent removal that corresponds to at
least 92 percent HAP removal, and only test for methanol percent
reduction during the remaining quarterly tests.
The general provisions allow for monitoring parameter
excursions during periods of startup, shutdown, and malfunctions.
The general provisions are being revised and will address the
issue of venting episodes that occur due to safety-related
concerns. It is important for the source to include all known
malfunctions in the startup, shutdown, and malfunction plan since
3-6
-------
a venting episode or monitoring parameter excursion that is not
included in the plan counts as a violation.
Comment: One commenter (IV-D2-15) recommended that the
final rule include more than one model for determining whether a
biological system is adequate, because models are updated and
improved frequently. The commenter requested that EPA's
recommended test method protocol also allow use of the NCASI
Organic Compound Elimination Pathway Model (NOCEPM).
Response: The EPA recognizes the NCASI NOCEPM model as a
credible biological degradation model; however, the WATERS model,
or updated versions, will be used to determine compliance because
EPA has used WATERS for demonstrating compliance with other rules
and the NOCEPM model has several limitations (A-92-40, IV-B-23).
However, industry has indicated that an updated version of NOCEPM
is expected after promulgation. The EPA may amend the rule with
a supplemental Federal Register notice to allow the use of the
updated version of NOCEPM pending evaluation of the model.
Comment: One commenter (IV-D2-10) expressed concern over
using emission factors to prove compliance with emission
standards, noting that actual emissions data from the sources
should be used whenever possible.
Response: Emission factors are not used for demonstrating
compliance with the rule. Compliance with the pulping and
bleaching standards is determined based on emissions test data
with the following exceptions. An initial performance test is
not required for a thermal oxidizer meeting the temperature and
residence time specified in the rule, nor for power boilers, lime
kilns, and recovery furnaces that are used for controlling
J-7
-------
pulping process emissions. An initial performance test is also
not required for bleaching systems that use TCF technologies.
The initial and quarterly performance test for biological
treatment systems used to comply with the pulping process
wastewater standards requires that the destruction efficiency of
the system be determined using a site-specific biodegradation
rate factor calculated using EPA's WATERS model (a system-
specific emissions model). Inputs to the model are obtained from
the biological treatment system's monitoring and operating
parameters.
For determining compliance with the clean condensate
alternative, emissions test data must be collected to determine
the baseline HAP emissions and emission reductions that would
have been achieved by implementing the MACT standards. The test
data would also be used to substantiate the HAP emission
reductions that are achieved using the alternative strategy.
8.2 CONTINUOUS EMISSIONS MONITORING
Comment: One commenter (20,049A2, 20,059) objected to EPA
requiring parameter monitoring rather than continuous emissions
monitoring systems (CEMS) . The commenter (20,059) contended that
EPA had not demonstrated that parameter monitoring is adequate
for purposes of enforcement or protection of public health. The
commenter (20,059) stated that EPA should require stringent
monitoring of control devices so that operators have the
incentive to properly maintain them and replace them before they
deteriorate.
One commenter (IV-D2-12) supported EPA's view that vents and
streams subject to the regulation should be specifically
identified and that parameter monitoring is a better approach
-------
than GEMS. Another commenter, (IV-D2-16) agreed that the proposal
for initial performance testing is reasonable because no
continuous monitoring system for methanol is available.
Response: The use of GEMS is not necessary to demonstrate
or assure compliance for certain pollutant and control strategy
combinations. As demonstrated by the history of NSPS and NESHAP
development, certain control devices are capable of achieving
continuous levels of emission control, when they are well-
designed, operated, and maintained. The EPA maintains that no
additional environmental benefit would be gained by requiring
GEMS in this rule.
Continuous parameter monitoring is consistent with
section 504 (b) of the Act, which states that "continuous emission
monitoring need not be required if alternative methods are
available... for determining compliance." The final rule requires
that HAP emissions be controlled to a specified percent
reduction, to a mass or concentration emission limit, or by
applying specific equipment. A compliance demonstration is
required for each emission point that demonstrates compliance by
meeting a control device equipment specification or a percent
reduction, mass, or concentration limit. Parameter monitoring
provides the information needed to know whether control systems
and other equipment are properly operated and maintained on a
continuous basis.
Comment: Two commenters (20,049A2, 20,059, 20,085)
contended that monitoring of specific pollutants should be
required. One commenter (20,059) argued that speciated CEM data
is needed in order to make sure that short-term averages are
being achieved for all pollutants of concern. The commenter
J-9
-------
(20,059) expressed concern that the emissions reductions promised
by this proposal will be unenforceable because of the lack of
monitoring requirements for measuring actual emissions of
specific HAP's. One commenter (20,085) asserted that if the
risks from air toxics are to be properly evaluated and
controlled, a monitoring component for specific pollutants should
be included in the rule. Otherwise, the commenter (20,085)
contended that it is unclear how the effectiveness of the MACT
rule will be evaluated or how determinations of residual risk
will be made in any meaningful way. One commenter (20,049A2)
claimed that parameter monitoring would result in an inadequate
amount of data to set "health protection" standards required by
the Act. The commenter (20,059) stated that if parameter
monitoring is used, EPA should not allow sources to select their
own measures of compliance, but EPA must identify the appropriate
range for each monitored parameter.
One commenter (20,049A2 and 20,059) considered the rule
inadequate because it allows monitoring of total HAP's rather
than specific air pollutants. The commenter (20,049A2) stated
that this would result in larger amounts of less toxic pollutants
reduced while not reducing more toxic pollutants. The commenter
(20,059) recommended that EPA investigate the applicability of
NCASI's test methods and other EPA test methods for periodic
monitoring of speciated emissions at pulp mills, and explore the
applicability of monitoring methods used by industrial
hygienists. The commenter (20,059) also asserted that speciation
would also be needed to protect the public's right to know and to
assess the seriousness of a violation.
-------
Response: MACT standards are technology-based standards and
are promulgated to achieve the maximum degree of reduction in HAP
emissions considering the costs of achieving such emission
reductions, any non-air quality health and environmental impacts,
and energy impacts. While the Agency agrees that it would be
advantageous to build a data base of specific HAP emissions for
future consideration of section 112(f) for residual risk, the
purpose of the monitoring requirements set forth in this rule is
to ensure compliance with the MACT standards. The pulp and paper
NESHAP reduces total HAP. Methanol is an appropriate indicator
of total HAP since it is the dominant HAP present in pulping
vents and condensates and since the control technologies
identified in the rule do not remove HAP's preferentially. For
bleaching vents, chlorine was designated as the surrogate for
chlorinated HAP's (other than chloroform) because the MACT floor
control technology, caustic scrubber, was installed primarily for
chlorine control.
For most systems, parameter monitoring adequately ensures
continuous compliance with the MACT standards. TO require
continuous or periodic emissions monitoring of specific HAP's is
unnecessary and will not provide additional pollution reductions.
Monitoring health risks, is outside the scope of this rule. For
biological treatment systems, continuous compliance is
demonstrated using parameter monitoring combined with emissions
modeling. The monitoring parameters specified for biological
treatment systems are appropriate indicators that the system is
being operated properly. If one of the monitoring parameters is
outside the range established during the initial performance
test, then compliance with the standard is demonstrated using the
3-11
-------
WATERS emission model. This monitoring approach was developed
since biological treatment systems and site-specific designs and
may be achieving the HAP removal efficiency required by the
standard even though one of the monitoring parameters is outside
of the established range.
With regard to the commenter's suggestion to use the NCASI
test methods for periodic monitoring, the samples collected using
the NCASI test methods must be analyzed before an indication of
emissions can be determined. Conseguently, the use of NCASI test
methods for periodic monitoring does not provide an instantaneous
indicator of continuous compliance unlike parameter monitoring.
During the initial performance test, each facility must
demonstrate compliance with applicable emission limits. At this
time, the appropriate monitoring parameter values (i.e., those
values recorded during the performance test when the source was
achieving the MACT Standard) will be determined and specified in
the source's permit. For the sulfite pulping and condensate
segregation monitoring standards, EPA did not have sufficient
information to specify the parameters that should be monitored to
demonstrate continuous compliance. For those instances, or if an
alternative parameter is chosen to be monitored instead of the
parameter specified in the standard, then sufficient rationale
must be submitted to the Administrator to justify the facility's
assertion that the parameter chosen indicates that the control
device or system is in compliance with the standard.
Comment: Two commenters (20,007, 20,059) disagreed with
EPA's decision not to require GEMS to measure total HAP's. One
commenter (20,059) contended that GEMS should be required
whenever technically feasible and for all pollutants that can be
5-12
-------
measured. The commenter (20,059) specifically stated that GEMS
should be required for combustion sources at paper mills. One
commenter (20,007) contended that they had demonstrated to EPA
that an automated gas chromatographic system could be used to
measure and speciate pertinent volatile HAP's. The commenter
(20,007) supplied a chromatogram illustrating the separation of
14 HAP's in less than 5 minutes in the 10 ppmv range. The
commenter (20,007) also claimed that such devices would increase
the accuracy of compliance demonstrations and contended that
EPA'S language in the rule regarding the technical impossibility
of GEMS would limit technical advancement in the pollution
monitoring field. The commenter (20,007) provided language to be
included in the final rule that would allow GEMS.
One commenter (20,059) contended that EPA mentioned that
flame ionization analyzer (FIA) technology offered promise as a
monitoring technique but rejected this option because it did not
measure speciated emissions. The commenter (20,059) stated that
the standards did not regulate individual pollutants and
therefore EPA may have disqualified this control option
prematurely.
Response: The EPA has concluded that the use of GEMS is not
technically feasible, does not provide any additional
environmental benefit, and could significantly increase the cost
and burden of demonstrating continuous compliance. The automated
gas chromatographic system described by the commenter is used to
comply with numerical limits for specific compounds identified in
the facility's air permit. The proposed rule addresses total HAP
emissions and does not establish numerical limits for individual
HAP compounds. Therefore, an automated system for measuring
5-13
-------
specific HAP compounds would not be applicable. Additionally,
establishing emission limits for individual HAP compounds for
demonstrating compliance would require extensive emissions
testing which would significantly increase the costs associated
with compliance without providing any environmental benefit over
parameter monitoring.
As stated in the December 17, 1993 proposal, EPA believes
that FIA technology would not increase the accuracy of compliance
demonstrations and would place an undue burden on the affected
industry. The EPA's position regarding FIA technology has not
changed since proposal.
8.3 PARAMETER MONITORING
Comment: Two commenters (20,027, 20,054A2, IV-D2-15)
recommended that EPA specify that § 63.8 of the general
provisions is not applicable to process monitors for these
standards, and that the monitoring and quality assurance plan for
the control devices must be consistent with manufacturers'
recommendations for calibration and maintenance.
One commenter (20,043) stated that maintenance and
calibration of monitoring devices was not adequately addressed in
the proposed rule. Therefore, the commenter (20,043) asserted
that requirements in the general provisions could not be met in
practice. For example, the commenter (20,043) stated that zero
and span checks on magnetic flow devices only reflect the
operation of the electronics and not the magnetic field itself.
The commenter (20,043) stated that one type of flow monitoring
device, a delta pressure cell, contains a critical orifice which
must be visually monitored to determine if its size is changing
5-14
-------
and stated that a zero and span check would not indicate this
change.
One commenter (IV-D2-15) asserted that unlike continuous
monitors, the instruments used to measure pH, steam flow, and
feed flow for steam strippers and scrubbers, cannot be checked by
a standard and cannot be evaluated using daily zero and span
checks. The commenter (IV-D2-15) recommended that monitoring
requirements recognize these differences and that EPA also
specify that § 63.8 of the general provisions is not applicable
to these process monitors.
Response: The EPA disagrees with the commenters and
believes that the specific sections in § 63.8 of the general
provisions that apply to this rule are applicable to the required
process monitors. However, if an owner or operator feels that
the monitoring requirements in § 63.8 are not appropriate, the
owner or operator may apply to the Administrator for an
alternative monitoring method as outlined in § 63.8 (f) of the
general provisions to part 63.
Comment: One commenter (IV-D2-4) maintained that periodic
performance testing is necessary to account for degradation of
the process and control equipment, to determine if the operating
and monitoring conditions initially set are still appropriate,
and to adjust the surrogate parameters when necessary.
Response: The calibration checks specified in § 63.8 of the
general provisions are intended to identify and account for drift
of monitoring devices. If the compliance status of a facility is
in question, section 114 of the Act authorizes the Administrator
to conduct performance tests at any other time. if a facility
believes that the parameter values selected during the initial
3-15
-------
performance test are no longer appropriate, the facility can
modify their operating permit to revise the initial parameter
values based on additional performance test data. if process
operating conditions change, or operation of the control device
changes from those existing during the initial performance tests,
then additional performance tests must be conducted such that
new, appropriate compliance parameters can be established.
Comment: One commenter (IV-D2-15) supported EPA's decision
to use parameter monitoring and not to establish continuously
enforceable sulfite limits. However, the commenter IV-D2-15)
expressed concern that seasonal temperature changes and various
pulp grade changes could require a lengthy period of time to
establish which parameters need to be monitored in order to
establish long-term compliance. The commenter suggested granting
sulfite mills a compliance extension of 2 years to allow
establishment of the monitoring parameters.
The commenter (IV-D2-15) expressed concern that other
enforcement initiatives will subject sulfite mills to penalties
and enforcement actions that are not intended by this rule. The
commenter urged EPA to establish clearly that: (1) monitoring
parameters are used only as an indication that a process change
has occurred, (2) if a source operates outside a parameter, then
no violation is presumed, and (3) if a facility operates outside
a parameter, then the facility's only obligation, after
reporting, is to reestablish compliance at the new conditions.
The commenter requested that if EPA could not establish these
allowances, it should develop an equipment work practice standard
for sulfite mills.
3-16
-------
One commenter (IV-D2-16) cautioned that existing sulfite
recovery systems are designed to control sulfur dioxide, not
methanol, and that there may be no practical parameter monitoring
scheme to correlate methanol emissions. The commenter (IV-D2-16)
suggested a joint industry study of methanol emission rates
versus potential operating parameters, with the caveat in the
rule that if the mill demonstrates during the test program that
its emissions are consistently below the proposed emission rate
or percent reduction requirement regardless of operating
conditions, then no further routine testing or parameter
monitoring will be required.
Response: The EPA recognizes that there may be some
difficulty in establishing appropriate monitoring parameters for
sulfite pulping processes. The compliance schedule for sulfite
processes specified in the rule is 3 years after the effective
date. The EPA maintains that this time frame is sufficient for
conducting the initial performance test to determine appropriate
monitoring of parameter values. However, if additional time is
needed to establish appropriate parameters, the mill may petition
the Administrator to extend the compliance schedule for one
additional year.
The initial performance test should be executed during
periods of normal operation. If a mill's processes are variable
from an emissions standpoint, then the initial performance test
should be conducted such that the parameters monitored are
appropriate to indicate continuous compliance under all operating
conditions that are likely to occur. if the facility later
believes that compliance of the standard can be achieved at a
different monitoring parameter value, the facility may conduct a
3-17
-------
performance test to demonstrate compliance and reestablish
appropriate monitoring parameters (i.e., revise their operating
permit) before any exceedance occurs.
Comment: Three commenters (20,011, 20,027, 20,054A2)
disagreed with EPA's assertion that the monitoring of certain
parameters was already being performed by industry and would not
impose any additional costs on the industry. One commenter
(20,054A2) explained that process monitors currently in use are
not necessarily used for compliance monitoring, but for process
information.
Two commenters (20,027, 20,054A2) contended that the inline
process parameter monitors required by these standards are not
similar to GEMS and continuous parameter monitoring would not be
feasible. Two commenters (20,011, 20,027) indicated that
continuous parameter monitoring would impose an additional cost
to the industry due to the accuracy and importance of required
information. One commenter (20,151) requested that EPA specify,
or provide guidance on what would be appropriate parameters to
monitor for biological treatment systems.
Response: Parameter values to be monitored by the
continuous recording systems are chosen by the mill and submitted
for approval by the Administrator after the initial performance
tests. Feasibility of using continuous monitoring of parameters
is based on: (1) the need to demonstrate continuous compliance,
(2) technical feasibility of the continuous parameter monitor and
(3) cost or burden imposed by such a requirement. The EPA
maintains that existing equipment can be used in most cases to
provide continuous parameter monitoring since most of the
monitoring parameters specified in the rule (e.g., thermal
3-18
-------
oxidizer temperature, steam stripper feed and steam application
rates) are currently being tracked to provide an indication of
proper operation. In other cases, new devices will need to be
installed.
Comment: One commenter (20,059) contended that EPA needs to
define combustion operating parameters to: (1) enable the
establishment of a greater than 98 percent control efficiency
requirement as MACT, and (2) ensure that the control device
functions at the required levels. The commenter argued that the
monitoring requirements were not sufficient to guarantee
98 percent HAP reduction.
Response: The EPA has concluded, based on previous Agency
studies, that temperature and residence time sufficiently define
the combustion operation with respect to HAP destruction
(A-79-32, II-B-31). For boilers and lime kilns, combustion
temperatures and residence times are more than sufficient to
ensure at least 98 percent reduction of HAP's. For thermal
oxidizers, EPA has outlined three compliance options; 98 percent
HAP reduction, 20 ppmv (at 10 percent oxygen) outlet HAP
concentration limit, or an operating level of 1,600 and
Op
0.75 seconds residence time.
Comment: Three commenters (20,027, 20,054A2, IV-D2-15)
disagreed with EPA's assumption that it is common practice to
monitor scrubber inlet gas flow. The commenters (20,027,
20,054A2, IV-D2-15) stated that the industry practice is to
monitor pH and/or scrubber liquid flow to ensure good performance
ClOo
for chlorine and * control, and some mills use
oxidation/reduction potential as an alternate to pH.
i-19
-------
One commenter (20,043) approved of using a pH threshold for
monitoring compliance because each scrubber has a specific pH
threshold above which acceptable efficiency is maintained. The
commenter (20,043) recommended that the rule should allow each
scrubber to establish its own pH threshold. The commenter
(20,043) also asserted that maintenance on scrubber monitoring
devices can only be performed by breaking the line. For all
bleach plant scrubber monitors, the commenter (20,043)
recommended placing the pH electrodes in a high flow region of
the system (i.e., just downstream of the recirculating pump) to
increase the reliability and decrease downtime. The commenter
(20,043) recommended that the final rule allow the use of sample
pots spliced off the main line to act as measurement points so
that calibration and preventative maintenance can be performed
with a minimum of lost material.
One commenter (20,043) contended that EPA should allow the
use of the last stack test for a measure of air flow as
permanently installed pitot tubes would be impossible to
accurately maintain. The commenter (20,043) claimed that
periodic scrubber media flow measurements only indicate
non-compliance when the flow approaches zero and a drop in the
flow rate is not expected to result in non-compliance.
Therefore, the commenter (20,043) concluded that monitoring of
the flow rate is useful as part of a preventative maintenance
program but a drop in flow is not expected to indicate non-
compliance .
Response: The EPA contends that monitoring of inlet gas
flow rate is necessary to prevent circumvention of the standard.
During the initial performance test, a range of flow rates should
-------
be determined that reflect normal operations so that periodic
fluctuations in the flow rate would not trigger a violation of
the standard. Previous stack data measurements of flow rate are
not allowed in the final rule.
The rule specifies that the pH or oxidation/reduction
potential of the scrubber effluent must be monitored. However,
the facility has the flexibility to determine site-specific
values.
Comment: One commenter (IV-D2-15) stated that mills that
wish to demonstrate compliance with the percent reduction limit
only be required to conduct a one-time performance test coupled
with monitoring of scrubber parameters.
Response: The final rule specifies that during the initial
performance test, appropriate parameter values are determined.
For compliance purposes, only the parameter values need to be
monitored and recorded.
Comment: One commenter (20,027) argued that EPA should
revise the monitoring requirements for steam strippers. Two
commenters (20,027, 20,054A2) stated that monitoring of the mass
feed rate is not practiced in the industry, but the industry does
monitor flow rates. Three commenters (20,027, 20,054A2,
IV-D2-15) recommended monitoring the steam-to-flow ratio which
has been demonstrated to have a direct relationship to stripper
removal efficiency, rather than monitoring the mass feed rate.
One commenter (20,043) indicated that the steam stripper
monitoring devices are inline and are not readily accessible to
routine maintenance and calibration.
Response: The EPA revised the steam stripper monitoring
requirements to include feed flow rate, steam flow rate, and feed
3-21
-------
temperature. The EPA has concluded that monitoring these
parameters will provide an acceptable indication of steam
stripper performance and HAP reduction efficiency since the
steam-to-feed ratio has the greatest influence in HAP removal.
Comment: One commenter (20,027) recommended monthly inlet
and outlet methanol concentration tests for the compliance
demonstration for mills using biological treatment. The
commenter (20,027) acknowledged that daily or weekly soluble 5
measurements could be used as an indicator of normal biological
treatment system operation since methanol has an extremely high
solubility. However, the commenter (20,027) recommended that
soluble BOD not be used as a means of determining compliance with
a MACT standard. One commenter (20,151) requested that EPA
specify, or provide guidance on, what would be appropriate
parameters to monitor for biological treatment systems.
In response to the March 8, 1996 supplemental notice, one
commenter (IV-D2-15), disagreed with the use of soluble UUiJ5 as a
compliance measure for determining whether a biological treatment
system is adequate. The commenter (IV-D2-15) supported measuring
the inlet and outlet methanol concentrations to determine
compliance. The commenter (IV-D2-15) claimed that, based on
results of NCASI testing in 1995, the following monitoring scheme
for biological treatment systems should be followed:
1. Monitoring of soluble BOD (in ppmw) into and out of the
system on the same frequency as BOD is required to be
monitored in the mill's NPDES permit.
2. Daily monitoring of methanol (in ppmw) into and out of
the system commencing within 24 hours of determining
that soluble BOD removal has dropped below 80 percent
and continuing until greater than 90 percent methanol
3-22
-------
removal or methanol outlet concentrations below 5 ppmw
are measured for 3 consecutive days.
3. Non-compliance would be determined by measured methanol
removal efficiencies below 90 percent with outlet
concentrations in excess of 5 ppmw.
Another commenter (IV-D2-5) suggested that inhibited soluble
8005 be used instead of soluble BOD as a surrogate parameter for
methanol removal efficiency because the suggested parameter
(soluble 0005) ignores the effects of ammonia and looks only at
hydrocarbons such as methanol. One commenter (20,076 and
20,045), however, suggested that measurement of soluble BOD5
would be a good indicator of methanol removal efficiency and as a
means of demonstrating compliance with the MACT rule.
One commenter (20,059) stated that EPA should require more
frequent monitoring of the HAP content of the incoming and
treated wastewater by sampling liquid streams and speciating
their constituents.
Response: In the December 17, 1993 proposed rule, the
monitoring parameters specified for biological treatment systems
were inlet and outlet methanol concentrations determined every
30 days and appropriate parameters as specified in the operating
permit and demonstrated to the satisfaction of the Administrator.
The proposed monitoring requirements for biological treatment
systems have been revised in the final rule to more accurately
reflect the operation of these systems, based on comments and
discussions with industry (A-92-40, IV-E-83, 84, 87) . The final
rule specifies the following monitoring parameters:
(1) composite daily sample of outlet soluble BQDc, concentration
to compare to maximum daily and monthly averages, (2) inlet
liquid flow, (3) mixed liquor volatile suspended solids,
3-23
-------
(4) liquid temperature, and (5) average horsepower of aerator
units. Also, daily inlet and outlet samples must be collected and
archived. If the soluble BOD, mixed liquor volatile suspended
solids, or the horsepower of the aeration units is outside of the
range established during the initial performance test, then the
archived samples must be used to demonstrate that the biological
treatment system is achieving 92 percent reduction of total HAP.
The EPA asserts that no additional environmental benefit
would be obtained by requiring monitoring of speciated HAP' s
entering the biological treatment system since the standard is
based on total HAP.
Comment: One commenter (20,110) requested that EPA amend
the rule to incorporate telemetering, alarm indications, and
other administrative controls on non-incinerated venting of
pulping component gas collection systems from dedicated bypass
vents, rupture disks, and other potential discharge locations.
Response: The EPA maintains that mills already have
indicators of venting and bypass anomalies to provide for worker
safety. The EPA has concluded that a requirement of telemetering
and alarms add unnecessary burden and do not provide an
environmental benefit. The rule requires mills to report the
date and duration of any venting anomalies.
8.4 LEAK DETECTION AND INSPECTION
Comment: The December 17, 1993 proposed rule required
closed-vent systems to be visually inspected every 30 days and
measured initially and annually to demonstrate no detectable
leaks. Several commenters (20,011, 20,027, 20,054A2, 20,056,
20,118, 20,146, IV-D2-15) asserted that requirements for leak
checks and visual inspections of closed-vent systems are costly
-------
and unnecessary and recommended that EPA provide an exemption
from monitoring for enclosures and closed-vent systems operating
under a vacuum. One commenter (IV-D2-15) stated that closed-vent
systems are designed to operate under negative pressure, and
visual inspection of negative pressure systems would not provide
any additional benefits beyond current industry inspection
methods. The commenter (IV-D2-15) suggested that visual
inspections be required only for positive pressure systems. The
commenter (IV-D2-15) also stated that when visual inspections are
appropriate, an annual inspection would be sufficient.
One commenter (20,027) asserted that there should be annual
and startup leak detection and initial and bi-annual visual
inspection for positive pressure vent systems.
Two commenters (20,027, 20,054A2) recommended a work-
practice standard that would only require visual inspections of
positive-pressure closed vent systems. Two commenters (20,027,
20,054A2) explained that inspections should be conducted upon
startup and upon at least two additional occasions annually. One
commenter (20,059) contended that EPA should require a
hydrocarbon analyzer be used instead of visual monitoring for
detecting leaks in ductwork.
One commenter (20,036A1) contended that the monitoring
requirements in the proposed rule, such as monitoring of negative
pressures, leak detection with a portable hydrocarbon detector
for leaks greater than 500 ppmv, and bypass line monitoring are
unnecessary and illogical for sulfite mills where any loss of
sulfur dioxide over one pound is a major upset and must be
reported to the National Response Center. The commenter
(20,036~) claimed that the lower odor threshold of sulfite
3-25
-------
mills, between 0.3 and 1 ppmv, makes the hydrocarbon analyzer
requirement of 500 ppmv unnecessary.
Response: The final rule makes distinctions between
positive and negative pressure portions of closed-vent systems.
For positive pressure portions of the closed-vent system, the
rule requires monthly visual inspections and initial and annual
leak detection measurements. For negative pressure portions of
the closed vent systems, the rule requires monthly visual
inspections and annual demonstrations that each enclosure opening
is maintained at a negative pressure.
The EPA concluded that leak measurements for negative
pressure systems would not be necessary if a mill could annually
document that its system was operating at a negative pressure.
This decision to remove the leak test requirement from negative
pressure systems was made because industry burden will be reduced
without sacrificing environmental benefits since any leaks in a
negative pressure closed-vent system would not cause a release of
pollution but would draw air into the system.
Comment: Two commenters (20,102, 20,129) recommended a leak
detection standard of 50 ppmv instead of 500 ppmv in the rule.
The commenters (20,102, 20,129) contended that this requirement
was reasonable because it is already used in New York and
California for detecting fugitives from local exhaust ventilation
systems from dry cleaning operations. The commenters (20,102,
20,129) proposed an alternative requirement of a local exhaust
velocity of 50 ft/min or sufficient inward air flow as indicated
by visible smoke tube tests to indicate proper inward air flow
and negative pressures to properly capture HAP emissions from
pulping equipment. The commenters (20,102, 20,129) claimed that
8-26
-------
these provisions are standard industrial hygiene ventilation
provisions that are easily used.
Response: The EPA has concluded that the 500 ppmv leak
detection standard provides an adequate level of leak prevention
since this detection standard is consistent with other leak
detection standards that EPA has promulgated. The 500 ppmv limit
is associated with the accuracy limit of the detection device
used in Method 21 (for more details see EPA Method 21). state
implementation plans have the authority to lower the leak
detection standard below 500 ppmv.
8.5 BYPASS VENTS
Comment: In the December 17, 1993 proposal, bypass line
valves were required either to (1) have a flow indicator
installed, calibrated, and maintained to indicate flow, or (2) to
be closed with a car-seal or lock-and-key type configuration and
to be visually inspected every 30 days. One commenter (20,027)
considered the sealing of bypass vents to be an emission control
requirement that must be evaluated as part of the floor. FOr
safety reasons, several commenters (20,011, 20,027, 20,054A2,
20,146) stressed that bypass lines should not be sealed and
enclosure openings should not be locked. The commenters (20,027,
20,054A2) recommended allowing other means of monitoring venting,
such as manual log entries for manually operated by-pass valves,
valve position, and flow indicators (where applicable).
Response: The purpose of establishing requirements for
bypass vents is to minimize the events in which vent streams are
released to the atmosphere. Monitoring requirements such as
bypass line seals are an aspect of compliance and are not based
on MACT floor determinations. Based on an evaluation of the
5-27
-------
industry's comments, EPA has revised the bypass line requirements
to include log entries recording pertinent information such as
valve position, flow rate, and flow direction. The requirements
for a lock-and-key type seal have been revised to specify easily
broken seals (i.e., car-seals) for bypass line valves due to
safety concerns.
Comment: One commenter (20,054A2) contended that monitoring
flow through bypass lines would not be of any benefit since the
flow indicator cannot distinguish between inward and outward
flow. Another commenter (20,056) stated that industry should be
given the flexibility to utilize other devices such as
temperature sensors and chemical sensors, and other methods such
as manufacturers' recommendations, sound engineering practices,
and professional judgement instead of specifying a flow measuring
device be installed on bypass and vent lines. The commenter
(20,056) stated that EPA may suggest compliance parameters that
may be monitored but should allow for States to allow other
parameters for source monitoring.
Response: The rule contains two sets of requirements for
monitoring bypass lines. The first set of requirements is the
installation of a flow indicator in the bypass line which
provides a record of the presence of gas stream flow in the
bypass line at least once every 15 minutes. The second set of
requirements is the installation and maintenance of a bypass line
valve, monthly log entries of valve inspections, and a seal on
the valve mechanism that ensures that the valve or closure
mechanism cannot be opened without breaking the seal. The EPA
believes that flow into a closed-vent system from the bypass line
is unlikely and that any flow in a bypass line would be outward.
-------
If a mill does not wish to use the monitoring procedures
specified in the rule, the mill may request that the
Administrator allow an alternate monitoring method through the
procedure outlined in § 63.8(f) of the general provisions to
part 63.
8-29
-------
-------
9.0 TEST METHODS AND PROCEDURES
9.1 FIVE-MILL SAMPLING PROGRAM
Comment: Three commenters (20,011, 20,027, 20,056)
contended that the data gathered from the five-mill sampling
study should not have been used to develop the December 17, 1993
proposal because the test methods and data were suspect. one
commenter (20,027) stated that EPA's main contractor had major
concerns about the use of certain methods employed in the
program, including draft Method 0011 for aldehydes and ketones
and Method 26A for hydrogen chloride (HC1) in sources with
chlorine and C1C>2 present. One commenter (20,056) stated that
the laboratory performing Method 0011 analysis of aldehydes and
ketones conceded that there was difficulty with contamination
problems for process liquid samples. The commenter (20,056)
asserted that the aldehyde and ketone data were suspect. The
commenter (20,056) also stated that EPA's characterizations of
chlorine and HC1 emissions are questionable since the test method
used (Method 26A) could produce positive bias when used in the
presence of C102 •
One commenter (20,027) expressed concerns about the validity
of the sampling results described in the BID (A-92-40, II-A-35),
since there were conflicting results for the same compound when
measured by different sampling procedures.
9-1
-------
Response: At the time of proposal, the test methods used
for the five-mill study were considered appropriate because they
were the best available test methods for measuring the pollutants
of interest at the time. The data from the five-mill study have
since been supplemented by industry-supplied test data. The
Agency recognizes the difficulties associated with the methods
used in the five-mill study, and was cautious when incorporating
the five-mill study results into the development of revised
emission factors.
As discussed in chapter 2, the proposal data base relied on
model process units derived from emission points. The emission
points were characterized by the five-mill study. Based on
comments and data received following proposal, the approach used
to develop the emission factors has been revised from an emission
point to a mill-system approach. In the revised approach, data
from the five-mill study were only used where complete mill
systems were tested (e.g., all emission points in a pulp washing
system) . This helped EPA to examine all the data on an
equivalent basis. The five-mill test data generally fell within
the range of the industry-supplied test data when evaluated on
the mill-system basis. The analyses of these data are detailed
in the revised emission factor document (A-92-40, IV-A-8).
Comment: One commenter (20,056) stated that the NCASI
methanol impinger method utilized by EPA was believed to be
susceptible to false high bias: entrained moisture (containing a
higher concentration of methanol) is trapped due to the method's
high sampling flow rate. The commenter (20,056) stated that the
new industry testing program does not contain this bias since the
9-2
-------
heated SUMMA canister method was conducted concurrently with the
NCASI impinger method for quality assurance.
Response: At four of the five sites tested, the NCASI
methanol impinger method was not operated at a sampling rate high
enough that would likely entrain liquid droplets. The methanol
data collected at the fifth site may have been susceptible to the
high bias. However, as stated earlier in this section, the data
collected in the EPA five-mill study was supplemented with data
collected by industry. The available data (both EPA and industry
data) were evaluated using a mill-system approach and the data
collected in the five-mill study was retained in the analysis
since EPA data generally fell within the range of the industry-
supplied data.
9.2 REQUIRED TEST METHODS
Comment: One commenter (20,011) claimed that Method 21 was
inappropriate for methanol, TRS, and other volatile compounds.
Response: The EPA maintains that Method 21 is appropriate
for methanol and VOC leak measurement because it is the accepted
measurement method. Additionally, the leak measurements
specified in the rule, based on Method 21, do not require the
measurement of TRS compounds.
Comment: Two commenters (20,027, 20,056) indicated that
C102 has been shown to interfere with the accuracy of test
Method 26A; thus, this method may not be appropriate for
measuring emissions from bleach plant sources. One commenter
(20,027) recommended modifying Method 26A by replacing the
alkaline impinger with a potassium iodide impinger so that
chlorine and C1O2 can be measured accurately in bleach plant
gases.
9-3
-------
One commenter (IV-D2-15) asserted that ClO^3 listed as a
potential interferant to Method 26A, which was required at
proposal to measure chlorine emissions. The commenter (IV-D2-15)
concluded that it would not be appropriate to use Method 26A for
measuring bleach plant emissions. The commenter (IV-D2-15)
recommended an NCASI method which uses potassium iodide as an
absorbing solution followed by dual pH titration. The commenter
(IV-D2-15) stated that this method has been submitted to EPA for
approval as a validated method for bleach plant sources.
Response: The EPA agrees with the commenters regarding the
potential interference from C102. Chlorine dioxide is a listed
interferant in Method 26A. The final rule contains specific
modifications to Method 26A to make the method appropriate for
determining chlorine concentration in the presence of ri(-\
Comment: Two commenters (20,027, 20,011) asserted that the
proposed test method for sampling methanol, proposed Method 308,
has not been evaluated using Method 301 validation criteria. The
commenter (20,027) submitted several minor changes to the
proposed Method 308 that should be made to allow additional
flexibility in the method. Another commenter (20,087)
specifically noted problems with sections 2.1.6, 2.1.9, 3.2.3,
3.2.4, 3.2.5, 4.1.2, 4.1.3, 4.2.1, 4.3.2.3, and 6.1 of proposed
Method 308. The commenter (20,087) suggested language that would
solve the problems.
Response: The proposed Method 308 has been validated using
Method 301 validation criteria. The validation was conducted by
the Atmospheric Research and Environmental Analysis Laboratory in
EPA's ORD. The results of the validation were reported in the
January 1995 issue of the Journal of the Air and Waste Management
9-4
-------
Association. Method 308 was promulgated in this rulemaking and
has been revised to incorporate the technical comments provided
by the commenter.
Comment: One commenter (20,027) declared that EPA should
recognize the applicability of the NCASI NOCEPM model and allow
its use as an alternative to the WATER7 model since its
applicability to the treatment of pulp and paper wastewaters is
well accepted. The commenter (20,027) noted that the biological
degradation kinetics in the WATER7 model are based on the two-
parameter Monod kinetics; however, Method 304 (the required
Method) provides a single rate parameter (first order) which
cannot be used directly as an input to the WATER7 model. Rather,
WATER7 must be "forced" to assume a first-order relationship in
order to use the results obtained from Method 304.
Response: [Note: The WATERS model is an update to the
WATER7 model]. The EPA recognizes the NCASI NOCEPM model as a
credible biological degradation model. However, the WATERS model
will be used to determine compliance because EPA has used WATERS
for demonstrating compliance with other rules and the NCASI
NOCEPM model has the following limitations: inability to support
Monod kinetics; inability to simulate plug-flow or sequential
reaction; and inability to model recycle flow, clarifiers,
collection system elements, screens, and trenches (A-92-40,
IV-B-23). However, industry has indicated that an updated
version of NOCEPM is expected after promulgation. The EPA may
amend the rule with a supplemental Federal Register notice to
allow the use of the updated version of NOCEPM pending evaluation
of the model.
9-5
-------
The WATERS model still incorporates the two-parameter Monod
kinetics; therefore, the WATERS model will need to be set up
properly if single rate parameters from Method 304 are used for
WATERS inputs. The EPA maintains that the WATERS model provides
acceptable results with the modified setup needed to incorporate
Method 304 results.
Comment: One commenter (20,011) asserted that the
Method 305 procedure for determining the HAP content of a waste
stream is unnecessarily complex for methanol, and differs from
the version proposed in the HON. The commenter (20,011) claimed
that the method has not been validated or published for comment.
Response: Method 305 in the proposed rule is the same as
the method specified for compliance testing under the HON. The
method was published for comment with the proposed HON and was
promulgated with the final HON. The method has been validated by
Method 301 and was extensively evaluated in the laboratory before
proposal with the HON.
In March 1997, industry informed EPA that it had not used
Method 305 to obtain the methanol steam stripper performance data
(which was used as the basis for the proposed pulping process
condensate standards). Instead, a direct aqueous injection gas
chromatography/flame ionization detection (GC/FID) method was
used (NCASI Technical Bulletin No. 684, Appendix I) .
Consequently, the industry contends that Method 305 should not be
specified in the final rule for determining compliance with the
pulping process condensate standards. However, the NCASI test
method has not been validated using EPA Method 301 procedures.
If the Agency approves the Method 301 validation procedures for
NCASI's GC/FID test method, this method may be referenced as
9-6
-------
either an alternative or a replacement for Method 305 (for
determining methanol concentration only) with a supplemental
Federal Register notice since it is unlikely that the test method
validation would be completed before promulgation of the MACT
standard.
Comment: One commenter (20,144) indicated that target
compounds must be specified for Method 305, because the
analytical methods are compound specific. The commenter (20,144)
asked if the compounds specified in the preamble in section X.A.3
would constitute such a list.
Response: The final rule specifies that Method 305 must be
used to determine the methanol or total HAP concentration in
process liquid streams. In determining the total HAP
concentration for use in the mass flow rate, mass per megagram of
pulp produced, or the mass percent reduction requirements
demonstrations, the final rule contains the criteria for
excluding compounds. Compounds with concentrations at the point
of determination that are below 1 ppmw or compounds with
concentrations at the point of determination that are below the
lower detection limit where the lower detection limit is greater
than 1 ppmw are not required to be included in the total HAP
concentration determination.
Comment: One commenter (20,011) stated that Method 25D for
determining compliance with the wastewater requirements in the
proposed rule was not employed during the testing program, and
may be inappropriate for use on pulp mill sources. The commenter
(20,011) requested that EPA clarify which test methods must be
used to verify compliance when biological treatment systems are
used.
9-7
-------
Response: The rule does not specify the use of Method 25D.
The rule requires the use of Method 305 for determining the
methanol (surrogate for total HAP) concentration in wastewater
streams.
For determining compliance with the biological treatment
system, the fraction of methanol degraded in the system is
determined by using the procedures specified in appendix C of
part 63 (except that the inlet/outlet test can not be used for
systems that are not well-mixed). The proposed rule incorrectly
indicated that Method 304 was contained in appendix A of part 63.
Method 304 is contained in appendix C of part 63. This change
has been made in the final rule.
Comment: Two commenters (20,102, 20,129) requested that the
test methods and procedures in § 63.451 include equations for
combustion efficiency used in RCRA 40 CFR 264 to ensure adequate
calculation and specification of combustion efficiency using CO
and CC>2 emission values.
Response: The final rule specifies the equation for
calculating the percent destruction on a mass basis. With regard
to the commenters' suggestion to monitor CO and C02 to indicate
combustion efficiency, EPA asserts that these monitoring
requirements would place additional burden on the affected
facilities without providing a substantive improvement in
monitoring combustion device efficiency. For thermal oxidizers,
the rule requires the facility to monitor the combustion device
temperature. Since this parameter is determined during the
initial performance test, EPA contends that this parameter is
sufficient for monitoring thermal oxidizer efficiency. The rule
also allows using power boilers, recovery furnaces, and lime
-------
kilns for controlling pulping process emissions. NO monitoring
requirements or initial performance tests are required for these
devices since the HAP destruction efficiency should exceed the
98 percent required in the rule when the devices are properly
operated.
Comment: One commenter (20,056) recommended allowing the
mass of pulp produced during a sampling event (for determining
the mass emission rate) to be determined over a longer period of
time than proposed sampling period to allow for periods when the
pulping process may be curtailed or even shut down while other
processes may continue to run.
Response: The EPA disagrees with the commenter's
suggestion. While some processes in the mill may have some
degree of independence due to in-process storage, EPA asserts
that periods of shutdown of all or part of the mill to which the
commenter alludes, do not constitute normal operation with regard
to any emissions testing program. The EPA recommends that any
emissions testing be conducted when the pulp production process
is in normal operation.
Comment: One commenter (20,070A1) stated that sampling
procedures required to demonstrate compliance with percent
destruction option for the combustion standards could expose
personnel to safety hazards when sampling inlet gas streams.
Response: The EPA recognizes the safety concerns expressed
by the commenter. In the final rule, several control options are
provided for complying with the pulping process standards. The
EPA's intent in providing compliance options was to allow mills
flexibility in demonstrating compliance. if a mill does not feel
comfortable with sampling inlet gas streams to demonstrate
9-9
-------
compliance with percent destruction requirements due to safety
concerns, the mill may choose one of the other control options
(e.g., outlet concentration).
Comment: One commenter (20,144) indicated that the rule
does not allow for any vacillation around the monitored
parameters (minimum or maximum) and, therefore, compliance with
the rule would be impossible. The commenter (20,144) suggested
that the rule specify an averaging time for each of the
parameters specified in § 63.452.
Response: The final rule does not specify averaging times
for monitoring parameters. Rather, the final rule specifies that
the owner or operator shall provide for the Administrator's
approval, the rationale for the selected operating parameter to
be monitored, the monitoring frequency, and the averaging time.
9-10
-------
10.0 RECORDKEEPING AND REPORTING
Comment: The reporting and recordkeeping requirements of
the general provisions apply to all sources subject to the NESHAP
unless a relevant standard specifically exempts or modifies those
requirements. The December 17, 1993 proposal was issued prior to
the final development of the general provisions. The proposal
specifically required the recordkeeping requirements located in
§
63.10 (a), (b), and (c) for monitoring parameters. The proposal
also required the reporting requirements found in the general
provisions in the following reports:
Initial Notification [§§ 63.9 (a) through (d) ;
63.10(f)] ;
Notification of Performance Tests r«« 63.7; 63.9(g)];
Notification of Compliance Status rg 63.9(h)];
Exceedance ReportstS 63.10(e) (3) (i) through (v) and
(viii)] ; and
Summary Reports (quarterly)[§ 63.10 (e) (3)] .
Several commenters (20,011, 20,027, 20,083, 20,102, 20,103)
opposed the reporting time of 45 days for Initial Notification as
being completely unrealistic in light of the realities of
compliance planning. Several commenters (20,027, 20,056, 20,083)
contended that the recordkeeping and reporting requirements were
excessive and may be contrary to the Paperwork Reduction Act.
10-1
-------
Two commenters (20,056, 20,102) indicated that the recordkeeping
and reporting required may take as long as 1 year to develop and
implement. One commenter (20,056) estimated that the industry-
wide cost for implementing a digital-based reporting system to be
between $500 million and $1 billion. Another commenter (20,083)
stated that EPA's estimate that the recordkeeping and reporting
burden of the proposed rule would require 923 to 1,797 man-hours
or approximately between one half and one person-year per source
to implement was considerable, but asserted that EPA's estimates
of recordkeeping and reporting are only a small fraction of the
true burden.
One commenter (20,018) agreed with industry recommendations
to reduce the recordkeeping and reporting burden, and provided
some additional recommendations including: eliminate the
requirement to retain monitoring values if the values show
routine compliance, retain only outlying monitoring values,
eliminate the requirement to retain all records, retain only
those records specifically identified by name, and eliminate the
reporting of data that is already required under other EPA,
State, or local rules.
One commenter (20,102) also stated that if EPA provides
guidance to State and local agencies as to what is acceptable for
notification of performance tests, then 75 calendar days for
notification is sufficient.
One commenter (20,027) concluded that EPA must amend the
monitoring, recordkeeping, and reporting requirements of the
proposal to conform to the less stringent requirements adopted in
the HON.
10-2
-------
One commenter (IV-D2-15) contended that proposed reporting,
recordkeeping, and monitoring requirements go well beyond what is
reasonable or necessary.
Response: The recordkeeping and reporting requirements in
the December 17, 1993 proposal were based on the requirements in
the proposed general provisions. The final general provisions,
which now have been promulgated, reduce some of the compliance
burden relative to the proposed version. In addition, the final
pulp and paper rule provides exemptions and modifications from
some of the general provisions. The pulp and paper rule
incorporates by reference specific sections of the general
provisions for clarity.
In the final rule, EPA revised the recordkeeping and
reporting requirements to reflect revisions to the general
provisions and to respond to concerns expressed by commenters.
Specific revisions include the following:
Due date for Initial Notification was changed from
45 days to 1 year.
Information required in Initial Notification report was
greatly reduced.
Changes were made regarding the need for, and frequency
of, quarterly excess emission reports.
Performance Test deadline was extended from 120 to
180 days, along with a change in the notification of
test date from 75 to 60 days.
Changes were in requirements for site specific test
plans.
Clarification of the difference between "performance
test" and "performance evaluation."
10-3
-------
"Step-by-step" procedures in startup, shutdown, and
malfunction plans were deleted.
A non-binding control strategy report was added to be
submitted with the Initial Notification and every
2 years, beginning 1 year following promulgation. This
requirement is for owners or operators of sources
selecting the extended compliance plan specified in
§ 63.440(d) (1) of the rule.
An option was added for Regional EPA offices to waive
duplicate submittal of notifications and reports.
A requirement for owners or operators to maintain a
record of their determination of their area source
status was added to show that a relevant standard does
not apply to them (assume this will not affect any of
NESHAP pulp and paper mills).
In addition to these specific recordkeeping and reporting
revisions, the final rule specifically names the process streams
that are subject to control by the rule. This approach also will
reduce the recordkeeping and reporting requirements, because
characteristics of individual process streams will not have to be
reported.
Comment: One commenter (20,056) objected to retaining
records for 5 years but recommended keeping them for 2 years.
Response: The EPA disagrees with the commenter and does not
believe that retaining records for 2 years is sufficient because
it does not provide adequate detail on the history of the mill.
The EPA believes that retaining records for 5 years (first
2 years on site, remaining 3 years off site) as specified in the
general provisions is appropriate. The EPA maintains that
5 years of records are needed to provide adequate compliance
history for each mill.
10-4
-------
Comment: One commenter (20,110) suggested that the rule
incorporate recordkeeping requirements and other administrative
controls on the venting of non-incinerated vent streams
(i.e., by-pass and emergency vents) from pulping component gas
collection systems, from dedicated bypass vents, from rupture
disks, and from other potential discharge locations.
Response: The EPA agrees with the commenter; however the
rule already requires owners or operators to report venting of
uncontrolled streams (i.e., by-pass and emergency vents) as
specified in § 63.10.
Comment: One commenter (20,092) contended that the rule
should require quarterly reporting until 2 years pass without an
exceedance of any State or federal emission limitations
applicable to the source. The commenter (20,092) stated that
once this occurs, semi-annual reporting is acceptable, provided
that any exceedance triggers a renewal of quarterly reporting.
Response: The EPA disagrees with the commenter's
recommendation of requiring quarterly reports for the first
2 years. The EPA maintains that semi-annual requirements, as
specified in the general provisions, is consistent with other
rules and provides sufficient reporting frequency. An owner or
operator is required to submit quarterly reports if any excess
emissions occur during the reporting period. The commenter's
recommendation would place undue reporting burden on the affected
industry without achieving any significant environmental benefit.
10-5
-------
-------
11.0 COST/ECONOMIC IMPACTS
11.1 COST IMPACTS
Comment: Several commenters (20,014, 20,018, 20,027,
20,028, 20,039, 20,046, 20,067, 20,070A1, 20,071) stated that EPA
severely underestimated the compliance costs of the proposed MACT
standards.
One commenter (20,039) stated that the final costs of
compliance with the proposed rules may approach $20 billion as
opposed to the $4 billion projected by EPA. One commenter
(20,070A1) stated that the capital costs to comply with the
cluster regulations could be at least $300 million, and may be
twice this amount depending on the degree to which the final rule
differs from the proposed rule.
One commenter (20,046) stated that EPA's compliance cost
estimate used for their mill was less than half the cost of the
estimate determined by the commenter and industry experts.
Another commenter (20,067) stated that an estimated
$350 million will be spent on compliance modifications for 11 out
of 12 mills and an additional $100 million may be spent depending
on interpretation of several vague definitions, terms, and
phrases in the proposed NESHAP. One commenter (20,014) indicated
that EPA did not take into account the cost of lost production
during construction or modification. One commenter (20,074)
11-1
-------
urged EPA to adopt the proposal made by the American Pulp and
Paper Industry which met the stated goals of EPA for this
rulemaking. The commenter (20,074) added that the cost of the
industry proposal was $1 million compared to EPA's proposed rule
at $2 million.
Another commenter (20,148) contended that EPA lacks
sufficient data for development of this standard. The commenter
(20,148) suggested that the true costs and benefits cannot be
determined until sufficient data are obtained.
Response: After review of the comments on the proposed rule
and additional data supplied by the commenters and pulp and paper
industry representatives, EPA has made significant changes for
the final pulp and paper NESHAP. Among the significant changes
are: subcategorization of the industry, requiring only specific
named vent and wastewater streams to be controlled, and providing
several options for control.
By subcategorizing the industry, EPA has evaluated the level
of control at existing kraft, soda, semi-chemical, and sulfite
mills individually. As a result, the control requirements for
soda and semi-chemical mills are significantly reduced from the
requirements for kraft mills, and the control requirements for
sulfite mills are specific to sulfite mills and not transferred
from kraft mills. The final rule requires only specific named
vent and wastewater streams to be controlled in each of the
subcategories. Therefore, the number of emission points that are
required to be controlled, and the cost of compliance, have been
significantly reduced from proposal. The EPA believes these
changes will reduce much of the commenter's concern about
confusion in the rule.
11-2
-------
The final rule also provides several options for controlling
vents and wastewater that will allow the owner or operator the
flexibility of choosing the best option for their mill. For
example, kraft mills complying with the wastewater requirements
can choose either to use a steam stripper or to hard-pipe
wastewater to a well-operated biotreatment unit. Sulfite mills
have the flexibility of complying with emission limits or percent
reduction requirements with any technology that can meet the
requirements. The EPA believes that these changes, as well as
other changes to the final rule, will significantly reduce the
compliance cost of the rule.
Additionally, EPA has revised the national cost impacts to
incorporate new data supplied by commenters and representatives
of the pulp and paper industry. The new data include:
information to characterize vent streams (temperature, flow rate,
and moisture content); description of equipment in vent gas
treatment systems; updates to the data base characterizing the
equipment and processes at pulp and paper mills, and cost
information for condensate segregation and other controls. The
commenters are referred to chapter 20 of this document and EPA's
memorandum discussing the costing changes (A-92-40, IV-B-13).
Changes made to the costs for the effluent guidelines are
discussed in the preamble for the promulgated air and water
rules. The EPA contends that the costs in the final impacts
analysis represent an appropriate estimate of the cost of
compliance with the final rule.
11-3
-------
11.2 ECONOMIC IMPACTS
Comment: Two commenters (20,115A2, 20,117) argued that EPA
did not properly evaluate the effect of the proposed rule on the
nation's economy.
Response: Total impacts on employment and output, both
direct and indirect, are estimated with final-demand national-
level input-output multipliers from the U.S. Department of
Commerce's Regional Input-Output Modeling System (RIMS II: Pigler
1993) which provides estimates of losses in employment,
shipments, and Gross Domestic Product. These impacts are
reported in the 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 (A-92-40, I-A-2)
(hereafter referred to as EA). These estimated effects on
employment, output, and shipments relate to both direct and
indirect economic impacts of the combined air and water pulp and
paper rule.
Comment: One commenter (20,025) provided a number of
comments concerning the market model used to estimate market
impacts of the regulation at proposal. The comments included
specific criticism of the supply and demand parameter assumptions
of the model, the methods used to determine market equilibrium,
and other alleged model deficiencies.
Response: The EPA used a market model and a financial model
to estimate market impacts for proposal of the regulations.
However for promulgation of the final rules, the EPA chose to use
only the financial model with some modifications to estimate
11-4
-------
market impacts and to predict mill closures. The EPA concluded
significant market changes have occurred since proposal that
would necessitate an update of data used in the market model.
These data updates could only be accomplished through an
additional updated survey of all mills in the pulp and paper
industry. Since such a survey would be burdensome to the
industry and would require significant time and resources, the
EPA elected to utilize the financial model with modifications for
promulgation. Thus comments relating specifically to assumptions
underlying the market model are moot for the economic analysis
conducted for the final rules.
Comment: Several commenters (20,009, 20,057, 20,103,
20,104, 20,115A2, 20,117) indicated that the economic burden of
the proposed rules will force some facilities to close. One
commenter (20,115A2) stated that the proposed rule will close
between 13 and 33 mills with little or no benefit to the
environment or human health. The commenter (20,115A2) stated
that EPA should determine the percentage of the total nationwide
production capacity that will be lost due because of mill
closings. The commenter (20,115A2) stated that if the demand for
paper products approaches or exceeds the remaining production
capacity of the mills then there would be a strong tendency for a
run-up in prices. One commenter (20,057) indicated that the
proposed regulation will close 33 mills and eliminate
21,800 jobs, based on an industry estimate. The commenter
(20,057) stated that EPA's economic analysis would have concluded
the same results if EPA had properly estimated the capital
requirements of the proposed rule. One commenter (20,067) argued
that EPA ignored the fact that compliance with the proposed
11-5
-------
regulations depends significantly on the ability of the affected
facility owner to raise the capital necessary to conduct
compliance modifications. One commenter (20,018) argued that the
technological and financial impact of the proposed MACT rules was
greatly underestimated by EPA and the proposed rules will have a
negative impact on the ability of American pulp and paper
companies to compete in the world market. One commenter (20,046)
argued that an economic model that used true capital costs and
inherently higher operating costs would clearly show that the
proposed cluster rules are not affordable.
Response: The costs, economic impacts, and health and
environmental benefits of the proposed air (MACT I) and water
rules were evaluated and fully discussed in the Regulatory Impact
Assessment of Proposed Effluent Guidelines and NESHAP for the
Pulp, Paper, and Paperboard Industry (EPA-821-R-93-020) . This
assessment was updated for the final rule in the EA. In addition
to assessing the impact of the MACT I final rule on the pulp and
paper industry, the impact of the final MACT III, proposed
MACT II, and the final water rules were evaluated individually
and jointly. (Note that MACT III impacts are not reported in the
EA because the MACT III rule is not expected to result in control
costs or emission reductions for the pulp and paper industry.)
The EA estimates the costs and economic impacts of the
regulation, evaluates the health and environmental benefits of
the regulation, and compares the costs of the regulation to the
benefits of the regulation. The EPA agrees with the commenter
that based on the EA, there is the potential for facility
closures to result from the regulations. Although no mill
closures are predicted to result from the final MACT I
11-6
-------
regulation, it is anticipated that as many as three mills may
close due to the combined final MACT I, proposed MACT II, and
final water regulations. Comments that mill closures will exceed
the EPA's estimate are based on the pulp and paper industry's
estimate of the cost of emission controls. The EPA evaluated the
industry estimate of the cost of emission controls and adjusted
the cost analysis where appropriate. Based on the revised cost
estimates resulting from commenters' input, as well as other
elements of the impact analysis, the EPA reassessed the economic
impacts for the final rule in the EA. Job losses, decreases in
pulp and paper shipments, and decreases in exports associated
with predicted mill closures are reported in the EA. In
addition, the price increases anticipated for pulp and paper
products are estimated and reported in the EA. The air and water
rules are not expected to significantly impact the ability of the
domestic pulp and paper industry to compete in the world market.
The economic analysis also considers the cost of financing
emission control equipment and equipment necessary to meet the
effluent guidelines. For most of the analyses conducted, a real
cost of capital or financing (discount rate) of 7 percent is
assumed. However, a sensitivity analysis of the cost of capital
using company-specific cost of capital estimates is performed and
these results are discussed in the EA. The impact of increased
capital and operating costs because of environmental controls on
the financial viability of mills affected by the regulations is
fully evaluated and reported in the EA. With regard to the
assertion that capital costs and associated operating costs are
understated, EPA has revised cost estimates based on comments and
data provided after proposal. Significant changes were also made
11-7
-------
to the requirements, such as requiring named systems to be
controlled, which reduce the costs. The costs estimated for the
promulgated rule incorporated all these changes and data. The
EPA believes that the final rule costs appropriately characterize
the costs for the industry.
Comment: One commenter (20,009) stated that EPA should try
to distinguish between those mills that might close due to
meeting new environmental control technologies and those that
would close anyway due to market and/or production constraints
and urged EPA not to dismiss technological options with large
environmental benefits just because the costs "seem" high. One
commenter (20,103) stated that EPA did not perform an adequate
evaluation of the cost of compliance with this standard and the
benefits to society and the environment. In particular, the
commenter (20,103) asserted that the overall cost of shutdown of
some facilities does not seem to have been addressed by EPA.
Another commenter (20,104) contended that the proposed rules will
force chlorine and caustic soda manufacturing facilities to
close, including three chlorine and caustic soda manufacturing
facilities in the Pacific Northwest.
Response: In the EA, the number of mills that are
anticipated to close assuming baseline conditions (no additional
environmental controls) are distinguished from the number of mill
closures expected to close as a result of the environmental
regulations. Economic impacts reported in the EA relate
specifically to mill closures resulting from the environmental
regulations. Estimates of the number of job losses anticipated
to occur and potential price increases resulting from the
regulations are reported in the EA. Job losses, decreases in
11-8
-------
shipments, and decreases in exports are not expected to be
significant due to the level of mill closures expected. The cost
of mill closures that may result from the regulations are
measured in terms of lost production and potential job losses
with the financial closure model. Lost production and potential
job losses are measures of important costs of a mill shutdown.
Estimates of the direct and indirect economic impacts of the
regulations on the national economy are also reported. A
comparison of the costs and benefits of the regulations is
conducted in the EA. Many of the health and environmental
benefits of the regulations are discussed qualitatively, and thus
the monetized benefits are compared to annualized costs with
recognition that the monetized benefits are likely understated.
The air rules are anticipated to have negligible impacts on
the consumption of chlorine at pulp and paper mills. However,
the effluent guidelines are anticipated to cause a decline in
consumption of chlorine by the pulp and paper industry. In an
article published in the November 1994 issue of Chemical and
Engineering News, "Chlorine Industry Running Flat Out Despite
Persistent Health Fears", the pulp and paper industry is reported
to have consumed approximately nine percent of the total domestic
production of chlorine in 1994. The level of chlorine
consumption by the pulp and paper industry is anticipated to
decline to approximately six percent of the total domestic
chlorine production by the year 2000. Despite the anticipated
decline in consumption of chlorine by the pulp and paper
industry, the overall growth in domestic production for the
industry is anticipated to occur at a rate of 0.8 to 1.5 percent
per year suggesting that growth in chlorine consumption is
11-9
-------
anticipated to occur in industries other than the pulp and paper
industry. (Chemical Week. Web page. 1996.
) . Since moderate growth is anticipated in
domestic chlorine production for the future, it does not seem
likely that environmental regulations for the pulp and paper
industry will result in chlorine manufacturing facility closures.
Using the estimated annualized cost of this NESHAP, an
evaluation of the economic impacts and distributional effects to
the pulp and paper industry is performed. The final rule when
evaluated independently of other regulatory requirements for air
and water pollution, is not expected to have a substantial impact
on the industry. Estimated price increases are less than
0.5 percent for bleached paper-grade kraft and sulfite,
dissolving-grade kraft and sulfite, and semi-chemical pulp and
paper products, while unbleached kraft pulp is estimated to have
a price increase of almost 5 percent. The costs imposed on
affected facilities do not result in any mill or firm closures,
thus, the rule assessed individually is not expected to alter
employment, shipments, or exports for the industry by appreciable
amounts.
Implementation of the final rule is expected to reduce
emissions of HAP, VOC, and TRS, but increase emissions of PM,
S02 , CO, and NOX• The benefits that accrue as a result of the
standard result from changes in human health effects associated
with inhalation of the above pollutants, as well as changes in
welfare effects such as visibility, crop yields, materials
soiling, and corrosion. The EPA is not able to place a monetary
value on all of the benefits achieved by the rule. Values are
11-10
-------
obtained for changes in VOC, SC>2 f and PM emissions only. Total
benefits for these pollutants range in value from $727 million to
$1,493 million.
Comment: One commenter (20,061) contended that the
compliance costs of the future MACT standards for combustion
sources would make the cost to benefit ratio even less appealing
to industry and society. One commenter (20,027) stated that EPA
was obliged to have considered the costs and other impacts of the
future combustion MACT standards when considering beyond-the-
floor technologies.
Response: At proposal of the MACT standard for non-
combustion sources, the Agency was preparing the combustion
source MACT standard. The combustion source MACT standard was
proposed concurrently with promulgation of the chemical pulping
MACT standard. The economics and benefits analyses incorporated
the impacts of both MACT standards, as well as the impacts of the
effluent guidelines portion of the final rule.
Comment: One commenter (20,018) stated that EPA's cost-
benefit analysis for emission controls should consider the
technological differences between kraft, sulfite, and neutral
sulfite pulp mills. One commenter (20,072) stated that the gap
of non-competitiveness between soda and kraft mills will be
further widened if the soda mills are required to comply with the
same compliance regulations as kraft mills.
Response: In the final rule, EPA has subcategorized the
pulp and paper industry by pulping type (kraft, soda, semi-
chemical, and sulfite). As a result, the control requirements
for soda and semi-chemical mills are significantly reduced from
kraft mills. The control requirements for sulfite mills are
11-11
-------
specific to sulfite mills. Therefore, the cost of complying with
the standard is different for kraft, soda, semi-chemical, and
sulfite mills. These differences are incorporated in EPA's cost
and economics analyses.
11-12
-------
12.0 BENEFITS
Comment: Several commenters (20,027, 20,053A1, 20,088,
20,101, 20,129) argued that the generation of collateral
emissions associated with combusting vent gases are more of a
concern than methanol, the predominant HAP compound. One
commenter (20,027) claimed that to generate the steam required
for stripping will lead to significant collateral increases of
NOX, SO2, PM, CO, and CC>2. The commenter (20,027) declared that
it is a bad trade for the environment to pay this price for a
control effort that can largely be described as methanol removal.
One commenter (20,053A1) stated that the proposed requirements
for the collection and control of high volume systems with low
concentrations of HAP emissions suggest EPA has not fully
considered the potential environmental trade-off between the
minimal HAP reductions and the increased emissions of other
pollutants. One commenter (20,101) said a drawback of the
proposed rules is the increase in CO, nitrogen oxide, sulfur
dioxide, and PM due to combustion controls. Another commenter
(20,129) stated that the emissions from combustion sources may
increase due to the need for increased boiler capacity for
extended cooking times, C1O2 generation, etc. However, the
commenter (20,129) stated a demonstrative environmental gain will
12-1
-------
result in the reduction of chlorinated compound emissions from
the bleach plant.
Response: MACT standards are required to be based on
control of HAP emissions. Although methanol is the largest
emitted HAP, there are a number of other HAP's emitted from pulp
and paper mills that are substantially reduced due to the control
requirements in the final rule, such as chloroform, o-cresol,
etc. The final rule achieves a significant reduction in the
emissions of total HAP, VOC, and TRS compounds. gQ^e Agency
recognizes that some criteria pollutants (such as PM, and
NOX) will be increased due to the control requirements (from
combustion of vent gases and fuel for energy) , and the Agency has
accounted for these increases in the impacts analysis. However,
these increases are much smaller in absolute value than the
decreases in HAP, VOC, and TRS emissions. A detailed discussion
of the benefits of the rule are presented in the preamble to the
promulgated rule and in the Economic Analysis report (A-92-40,
V-A-2).
Additionally, the Agency believes that in some cases, the
impacts have over-estimated the emissions because some mills may
be able to use existing controls to reduce emissions. However,
EPA does not have sufficient information on the number and
effectiveness of these controls, so no reductions were estimated.
Also, mills may use fuels that emit lower amounts of criteria
pollutants when combusted, or may use other control options, such
as the clean condensate alternative or lower emitting equipment,
that may not increase secondary emissions.
12-2
-------
Comment: Two commenters (20,025, 20,114) stated that EPA
used incorrect assumptions about ozone formation contending that
the relationship between VOC reductions and ozone reductions is
not linear. One commenter (20,025) continues by stating that
given the relationship between VOCs and ozone in rural areas, and
other uncertainties [relating to background ozone concentrations]
there is no basis for monetizing the agricultural benefits
relative to VOC reduction.
Response: The photochemical production of ozone is the
result of atmospheric physical processes and complex chemical
processes involving two classes of precursor pollutants: VOCs
and NOX- The analysis for the proposal of the pulp and paper
rule used the most readily available data at the time to quantify
and monetize VOC emission reductions. Since that time, a more
recent analysis (the Regulatory Impact Analyses for the
Particulate Matter and Ozone National Ambient Air Quality
Standards [NAAQS] and Proposed Regional Haze Rule) provides data
that can be used to monetize VOC emission reductions. The ozone
NAAQS analysis acknowledges the complex relationship between
emission reductions and ambient ozone concentrations by using a
variety of prognostic and empirical models to examine this issue.
The complex relationship is also incorporated into the benefits
analysis for this pulp and paper rule since the VOC benefit value
is derived from the ozone NAAQS data.
One of the methods used to value VOC emission reductions
estimated for the pulp and paper rule limits the valuation (both
health and welfare categories) of the emission reductions only to
areas with ambient ozone concentrations high enough to
potentially violate either the current ozone standard or the
12-3
-------
revised ozone standard. These areas (rural or urban) are not
only above the background concentration level, but also above the
current ozone standard or the newly promulgated ozone NAAQS. The
Ozone Staff Paper estimates the national average background ozone
concentration to be approximately 0.04 parts per million, which
is incorporated into the benefit analysis of reduced ozone
concentrations. Given this estimated background ozone
concentration, the method of valuing VOC emission reductions as
described above addresses the background ozone concern in both
urban and rural areas.
Comment: Several commenters (20,025, 20,027, 20,101,
20,114, 20,116) stated that the benefits assessment of the
proposed rules contains calculation errors. Two commenters
(20,025, 20,027) stated that their review of the Regulatory
Impact Assessment (RIA) indicates that EPA's analysis does not
employ sound science, is skewed by a large arithmetic error, uses
unrepresentative data, and is based on unwarranted assumptions.
One commenter (20,025) stated that incorrect assumptions about
ozone formation led to unjustified agricultural benefits. Three
commenters (20,025, 20,101, 20,114) stated that EPA made a
mathematical error in the use of the Office of Technology
Assessment's economic benefits analysis which resulted in a
per-metric ton benefit for VOC control that is $468 million too
high.
Response: The benefits assessment in the RIA at proposal
contained a printing error. In addition, the benefits assessment
has been updated to reflect more recent valuation estimates for
VOC, PM, and SOo emission reductions. Revisions to the analysis
also include an added explanation of the underlying assumptions
12-4
-------
and a revision of the benefit calculations due to a reevaluation
of the emission reductions and the related monetized health and
welfare benefits valuation.
Comment: One commenter (20,083) argued that EPA should not
refer to methylene chloride as a VOC in the final rule and
preamble since methylene chloride is specifically excluded from
EPA's definition of VOC. Therefore, the commenter (20,083)
stated that corrections should be made to the benefits or
tropospheric ozone reductions calculations due to the control of
methylene chloride. One commenter (20,083) argued that it is
inappropriate for EPA to use VOC reductions as a "benefit" for
supporting the stringent HAP emission standards.
Response: Methylene chloride is not referred to as a VOC in
the final rule and is not included as a VOC in the benefits
analysis. However, methylene chloride is classified as a HAP and
benefits were attributed to reductions in human health effects
from reductions in emissions.
Comment: Two commenters (20,025, 20,116) stated that EPA
used an obsolete potency factor for the inhalation route of
exposure in the chloroform risk assessment. One commenter
(20,025) stated that EPA should update the potency factor. Two
commenters (20,025, 20,114) stated that the cancer risk
reductions for formaldehyde and chloroform were overstated due to
the use of incorrect potency and scaling factors.
Response: The Agency is aware that several organizations
have reassessed the carcinogenic potency of formaldehyde and
chloroform. The reassessments have incorporated more
biologically based dose-response information. The cancer potency
factors used in the risk assessment at proposal were taken from
12-5
-------
the Integrated Risk Information System (IRIS). A recent search
of IRIS shows the same cancer potencies as were used in this risk
assessment. This program office has tended to base risk
assessments on the values contained in IRIS. The Agency did not
reassess the risk from the pollutants for the final rule, did not
place a monetary value on them, and did not base the decisions in
the final regulatory alternative on this information.
Comment: Several commenters (20,016, 20,005, 20,027,
20,101, 20,117) said the costs outweigh the economic benefits.
One commenter (20,117) suggested that EPA seriously consider the
economic impacts of the proposal and compare those impacts to the
environmental benefits. The commenter (20,117) contended that
the environmental benefits achieved by the regulation will be
small while the economic impact will be severe. Two commenters
(20,011, 20,088) stated that the costs are likely understated
while the benefits are probably overstated due to the improper
characterization of secondary pollutant impacts. One commenter
(20,016) stated that the pulp and paper industry does not oppose
environmental capital investments but asks that any environmental
requirements be based on demonstrated benefits commensurate with
the costs of the requirements.
One commenter (20,101) contended that EPA should reconsider
whether the benefits resulting from reduced air emissions
outweigh the costs of achieving those benefits. One commenter
(20,027) submitted that the realistic costs of the proposed
"cluster" standards exceed the realistic benefits by a factor of
thirty. (Case law cited: Portland Cement Assoc., National Lime
Assoc., Sierra Club.) The commenter (20,027) indicated that the
enormous costs are unwarranted considering the minimal benefits
12-6
-------
achieved with the standard. The commenter (20,027) warned that
EPA grossly overestimated the environmental benefits of the
proposed MACT standards. One commenter (20,025) added that EPA
had not cited any reliable studies to support health benefits of
reducing TRS emissions. Another commenter (20,066A4) argued that
the proposed rules were only marginally cost-effective as
underscored by EPA's benefit-cost comparison. The commenter
(20,066A4) added that in every case of these studies, the
annualized compliance cost for the mills exceeds the annualized
benefit of the rule by many millions of dollars.
Response: The EPA is limited in its ability to place a
monetary value on all of the benefit categories. Because several
health and welfare endpoints, as well as entire pollutant
categories are not monetized, the estimate of benefits is
underestimated. For instance, one category that achieves
significant reductions is TRS, which is responsible for the
malodorous smell associated with pulp and paper mills and can
result in toxic effects, as well, irrespective of odor. The
value of these reductions could be significant given the odor's
negative affect on individuals comfort and well-being, and the
toxic effects that adversely impact human health (e.g.,
headaches, nasal irritation, and respiratory and cardiovascular
impacts) . Overall, all of the information outside of the
monetized costs and benefits presentation must be considered
before a determination that costs outweigh benefits can be made.
Given the uncertainties described in the analyses, EPA cannot
make a statement in either direction. The analysis of the final
rule presents a range of benefits. The lower bound estimate
results in a net cost (i.e., costs exceed benefits) while the
12-7
-------
upper bound estimate produces net benefits (i.e., benefits exceed
costs).
Comment: One commenter (20,114) contended that EPA should
perform a cost-to-benefit evaluation to justify the RIA. One
commenter (20,129) indicated that EPA should perform risk
assessments when developing control applicability cut-off values.
Response: The EPA did present a cost-to-benefit comparison
and presented the results in a RIA for the proposed rule. For
the final rule, the EPA presented results in the EA report
(A-92-40, V-A-2). The analysis of HAP benefits relies on risk
assessments, however, this is completed independent of the
development of cut-off values.
Comment: One commenter (20,011) indicated that EPA did not
provide sufficient information to allow the industry to check the
accuracy of the cost-benefit analysis calculations.
Response: The benefits analysis, in the proposed rules RIA
and in the promulgated rules EA report (A-92-40, V-A-2), outlines
the assumptions that were used. All sources of information used
are available in the public docket.
Comment: Two commenters (20,005, 20,059) contended that EPA
will have to establish a second round of standards in 8 years to
address the residual cancer risk associated with the pulp and
paper industry because the current proposed MACT standards were
not sufficiently aggressive. One commenter (20,059) stated that
EPA should establish final MACT standards that would reduce
emissions of carcinogenic and acutely toxic compounds to levels
that will protect public health with an adequate margin of
safety. One commenter (20,005) stated that EPA should establish
-------
air toxic regulations based on risk so that emissions are reduced
by the most cost-effective methods available.
Response: The Act requires that MACT standards require
"... the maximum degree of reduction in emissions of the
hazardous air pollutants that the Administrator . . . determines
is achievable . . . through application of measures, processes,
methods, systems, or techniques . . . ." In other words, the
MACT standards are technology-based standards, rather than risk-
or health-based standards; MACT standards control total HAP
emissions, rather than each individual HAP. Therefore, EPA
cannot consider the toxicity of different compounds when
developing the standards. The EPA maintains that the final rule
requires stringent control of all HAP's. Additional control of a
few HAP's through other technologies is not warranted considering
the cost and other impacts of those technologies.
As the commenter noted, the residual cancer risk after the
standard has been promulgated will be analyzed 8 years after
promulgation. At such time, EPA will review the toxicity of
specific compounds.
Comment: One commenter (20,072A8) stressed that requiring
the soda mills to collect, transport, and incinerate vent gases
that are not an odor problem discourages the use of the soda
process without consideration of the welfare benefits (odor
reduction) associated with soda mill operation.
Response: The EPA agrees with the commenter that soda mills
do not have TRS emissions or the odor problems associated with
TRS. However, test information and information submitted by the
industry indicates that soda mills have HAP emissions comparable
to kraft mills. The EPA has determined that significant HAP
12-9
-------
reductions can be achieved for minimal cost of controlling
selected equipment at soda mills. Therefore, EPA maintains that
substantial benefit is obtained from controlling emission vents
at soda mills.
Comment: One commenter (20,059) argued that EPA did not
provide sufficient information to indicate the extent of
emissions that will go uncontrolled as a result of the
exemptions.
Response: The EPA disagrees with the commenter. In
memoranda placed in the docket at proposal and promulgation, in
the background information document (volume 1), and in chapter 20
of this document, EPA has provided the baseline emissions and
emissions reductions for each control option. The emissions that
will go uncontrolled can be calculated from this information.
Comment: Commenter (IV-D2-15) opposed language in the
March 8, 1996, Federal Register notice that stated that all the
HAP' s to be regulated at pulp and paper mills "can cause toxic
health effects following exposure, including nausea, headaches,
respiratory distress, and possible reproductive problems" because
it does not reference amount of dosage or exposure levels. The
commenter contended that EPA had not shown that the effects
described are associated with exposure levels resulting from pulp
and paper HAP emissions.
Response: The Federal Register statement was a qualitative
description of possible effects of the HAP's emitted by pulp and
paper mills. The EPA did not state that these effects were
quantified. The benefit analysis contained in the RIA at
proposal evaluates and quantifies changes in cancer incidences
12-10
-------
resulting from the proposed rule, but qualitatively discusses all
other health effect end-points.
Commenter: One commenter (20,025) contended that the EPA
had no basis for assigning benefits to acrolein emission
reductions because there are no emissions at baseline. The
commenter (20,025) also stated that the TRS emissions calculated
by EPA were erroneously high, which led to a gross exaggeration
of potential human benefits.
Response: The EPA disagrees with the commenter. Based on
information contained in emission test reports for pulp and paper
mills, emission factors were developed for acrolein. The
commenter is referred to the chemical pulping emission factor
document (A-92-40, IV-A-81 for emissions information on acrolein
obtained by the EPA. Baseline emissions of acrolein were
calculated to be 257 mg/yr. Regarding TRS emissions, the EPA
revised its estimates of emissions based on additional
information and test reports obtained since proposal. See
chapter 20 for TRS emission estimates and the chemical pulping
emission factor document (A-92-40, IV-A-8) for development of TRS
emission factors.
12-11
-------
-------
13.0 EMISSIONS AVERAGING
Comment: The December 17, 1993 proposal requested comment
on whether to include emissions averaging in the final rule.
Three commenters (20,011, 20,027, 20,056) supported including
emissions averaging to provide flexibility to the industry.
Other commenters (20,059, 20,102, 20,103, 20,129) opposed
averaging because they asserted that averaging would increase
emissions and not be enforceable.
One commenter (20,027) noted that in keeping with EPA's
approach of not distinguishing among the regulated pollutants
(HAP's) from this source category, EPA should establish an
emissions trading system. The commenter (20,027) also favored
EPA making a generic finding that emissions averaging in the pulp
and paper industry does not increase hazards. Other commenters
(20,011, 20,146) supported an emissions trading system and
claimed that an emissions trading policy is consistent with the
Act, EPA policy, the current administration's views, and general
congressional intent.
Several commenters (20,059, 20,102, 20,103, 20,129) strongly
objected to emissions averaging and supported point-by-point
compliance requirements. Two commenters (20,102, 20,103)
expressed concern that averaging between types of emissions could
minimize the public health benefits of the regulation,
13-1
-------
particularly substitution of ultra-hazardous pollutants for less
toxic ones. One commenter (20,059) indicated that the omission
of detailed discussion of interpollutant trading in
section 112(g) of the Act reflects an intention of Congress not
to authorize this practice for MACT standards. One commenter
(20,103) contended that emissions averaging would be too
difficult to enforce. One commenter (20,129) considered HAP
emissions to be highly variable and a function of chemical
pulping process conditions and air pollution control technology.
The commenter (20,129) warned that averaging HAP emissions that
are highly variable and not well known is problematic.
One commenter (IV-D2-4) , while maintaining opposition to the
use of emissions averaging, conceded that in this instance,
limited emissions averaging may be useful to minimize the overall
cost of compliance while still achieving the desired emissions
reduction. The commenter (IV-D2-4) supported requiring a
"static" vs. a "dynamic" emissions averaging scheme, restricting
emissions averaging to streams of similar pollutants, and
allowing permitting agencies to restrict the use of emissions
averaging.
Response: Based on comments received following proposal,
EPA concluded that incorporating emissions averaging in the rule
would add flexibility and could reduce the costs of compliance.
However, EPA has decided that a traditional emissions averaging
approach (as taken in the HON) is not appropriate for this
industry. The EPA and industry held several meetings after
proposal to discuss mill-wide emissions limits and emissions
averaging concepts, based on the use of emission factors
(A-92-40, IV-D1-49, 51, and 61) . The EPA concluded that
13-2
-------
currently available emission factors were sufficient for
estimating national emissions reductions and impacts; however,
they were inadequate for demonstrating compliance in a
traditional emissions averaging program. The variability between
mill operations would require a case-by-case evaluation of the
feasibility of emissions averaging. Since a significant amount
of emissions source testing would be necessary to support a
viable emissions averaging program, demonstrating compliance
would be too burdensome on industry and very difficult to
enforce.
Some commenters suggested an alternative to traditional
emissions averaging that would be more appropriate for the pulp
and paper industry. This condensate pretreatment alternative is
currently referred to in the final rule as the CCA. A brief
discussion of this alternative was presented in the March 8, 1996
Federal Register supplemental notice. A description of the
industry's assumptions used to assess the condensate pretreatment
alternative was also submitted to the Agency (A-92-40, IV-D1-59).
The CCA is based on information provided by the industry
after the December 17, 1993 proposal (A-92-40, IV-D1-29, 29a, 33,
and 38). The CCA focuses on reducing the HAP concentration in
process water (such as from the digestion and liquor evaporation
areas) that is introduced into process equipment throughout the
mill. By reducing the amount of HAP in the process water,
reductions in HAP emissions will also be achieved since less HAP
will be available to volatilize off the process to the
atmosphere. To demonstrate compliance, the mass emission
reduction of HAP's achieved by the alternative technology must
equal or exceed that which would have been achieved by
13-3
-------
implementing the kraft pulping vent controls. Eligibility for
this compliance alternative is determined on a case-by-case basis
during the permitting process.
For purposes of developing a compliance strategy, sources
may use either emission test data or engineering assessment to
determine the baseline HAP emission reductions that would be
achieved by complying with the kraft pulping vent standard. To
demonstrate that the alternative technology complies with the
emission reduction requirements of the standards, emission test
data must be used. Two conditions must be met for a CCA
compliance demonstration: (1) owners and operators that choose
this alternative must first comply with pulping process
condensate standards before implementing the alternative
technology, and (2) the HAP emission reductions cannot include
reductions associated with any control equipment required by
local, State, or Federal agencies or statutes or with emission
reductions attributed to equipment installed prior to
December 17, 1993 (i.e., the date of publication of the proposed
rule).
For purposes of the CCA, the rule provides an alternative
definition of the affected source. The alternative definition
allows for CCA to apply to process systems outside of the kraft
pulping system. The expanded source includes the causticizing
system and the paper making system. The mill must specify the
process equipment within the expanded source with which to
generate the required HAP emissions reductions using the CCA.
The mass emission reduction of HAP's must equal or exceed the
reduction that would have been achieved through application of
the kraft pulping vent standards. The final determination of
13-4
-------
equivalency shall be made by the permitting authority based on an
evaluation of the HAP emission reductions.
Comment: Several commenters (IV-D2-2, IV-D2-7, IV-D2-15,
IV-D2-19) to the March 8, 1996 Federal Register supplemental
notice supported the concept of the CCA compliance approach as
outlined by the industry. The commenters also suggested that the
final MACT rule allow individual mills to make a case-by-case
demonstration that installing and operating a condensate
treatment system and reusing the cleaned condensates in various
process areas will achieve equivalent or greater mill-wide total
HAP emissions reductions as compared to the MACT requirements.
One commenter (IV-D2-15) recommended that in order to make a
case-by-case determination of equivalent emissions reduction, a
mill would: (1) determine total HAP emissions reductions for any
pulping component systems that would be subject to a 98 percent
total HAP reduction requirement under the final rule;
(2) determine the emissions reductions from implementing the CCA
for all process units where the recycled cleaned condensates
would be used (or would affect the emissions of total HAP) , by
using estimates of relevant process liquid concentrations; (3)
verify the step 2 reductions are equal to or exceed those from
step 1; and (4) periodically monitor the methanol concentration
or other appropriate parameters on a case-by-case basis to ensure
reductions continue.
One commenter (IV-D2-15) stated that NCASI documented the
relationship between process stream methanol concentration and
air emissions for vacuum drum brownstock washer systems, oxygen
systems, smelt dissolving tanks, and paper making systems. The
commenter (IV-D2-15) also contended that a relationship has been
13-5
-------
developed for estimating total bleaching system methanol
emissions based on the amount of methanol entering the bleaching
system and the total vent gas flow rate. These relationships
could be used by mills to demonstrate the emissions reductions
achieved by the reduction of HAP concentration in reused
condensate streams, in lieu of testing.
Response: The EPA included provisions for the CCA in the
final rule. To be considered equivalent to point-by-point
control, the CCA must achieve at least the same total HAP
reductions as would be achieved if the MACT controls were
implemented on a point-by-point basis. The responsible
permitting authority will determine the adequacy of the plan.
The EPA rejected the use of the HAP emissions/process water
concentration relationship data developed by NCASI (A-92-40,
IV-D1-29, 33, and 38) as a means of demonstrating compliance with
the CCA. While a relationship may exist between the HAP
emissions from a piece of process equipment and the HAP
concentration in the process water reused or recycled to the
equipment, the information compiled thus far by NCASI is
insufficient for demonstrating compliance due to inherent process
variability between mills. These emission factors may be helpful
for screening or preliminary evaluations of the viability of the
CCA. To demonstrate compliance with the CCA, however, the rule
requires that a mill (1) perform emissions testing to establish
the baseline, uncontrolled emissions level for the pulping system
after the pulping process condensate requirements of 63.446 are
f §
met; (2) apply the 98 percent HAP emissions reduction required by
63.443(c) to obtain a compliance HAP emissions level; and
(3) after the alternative technology has been implemented, retest
13-6
-------
the pulping system to determine the HAP emissions level. To
demonstrate compliance with the Act, the HAP emissions levels
measured after the CCA technologies have been implemented must be
egual to, or lower than, the compliance level of HAP emissions
calculated from the baseline testing.
13-7
-------
-------
14.0 DEFINITIONS
14-1
-------
Term
plant
Proposal/
supplemental notice
definition
None
Affected
source
I
to
Industry recommendation Final definition
The process equipment Definition not needed.
used at sulfite mills to
produce cooking acid from
sulfurous acid, sulfur
dioxide, bisulfite salts,
and acids and various
base cations.
(A-92-40, IV-D1-104)
For the purpose of this Definition not needed.
subpart, a facility which
is a major source that
produces pulp from wood
or other fiber sources, a
facility which is a major
source that manufactures
paper and paperboard, or
a facility which is a
major source that has
integrated production of
pulp and manufacture of
paper and paperboard.
Rationale for final
definition
No definition for acid
plant was incorporated
into the proposal.
While the industry's
definition appears to
be technically correct,
the term acid plant is
not used in the rule.
Therefore, this
definition was not
needed.
The affected source is
presented in the
applicability section
of the rule.
(A-92-40, IV-D1-104)
-------
Term
Air-dried
pulp
Black liquor
Bleaching
Proposal/
supplemental notice
definition
A pulp sample with a
moisture content of
less than or equal to
10 percent by weight.
Pulp samples for the
pulpinq component shall
be unbleached pulp and
for the bleached
component shall be
bleached pulp.
[from December 17, 1993
proposal]
Pulpinq liquor from the
diqester to the point
of its incineration in
the recovery furnace of
a sulfate (kraft)
recovery process. It
contains dissolved
orqanic wood substances
and residual active
alkali compounds from
pulping process.
[from December 17, 1993
proposal, preamble]
Briqhteninq and
deliqnification of pulp
by the addition of
oxidizing chemicals.
Industry recommendation
A pulp sample at 10
percent by weiqht
moisture content. Pulp
samples for applicability
or compliance
determinations for both
the pulpinq and bleachinq
component shall be
unbleached pulp.
(A-92-40, IV-D1-104)
Pulpinq liquor from the
pulpinq process to the
point of its incineration
in the recovery furnace
of a sulfate (kraft)
recovery process. It
contains dissolved
orqanic wood substances
and residual active
alkali compounds from
pulping process.
[commenter 20,056]
Briqhteninq of pulp by
the addition of oxidizinq
chemicals or reducinq
chemicals.
Final definition
Definition not needed.
Rationale for final
definition
The units used in the
rule are based on oven-
dried pulp.
Spent cookinq liquor that The commenter's
has been separated form definition chanqes the
the pulp produced by the word "digester" to
kraft, soda, or semi-
chemical pulpinq process.
"pulping process".
However, the rule
definition has been
simplified to improve
clarity.
The industry definition
was used in the rule.
[from December
proposall
1993 (A-92-40, IV-D1-104)
Industry definition
addresses the use of
reducing chemicals.
Therefore, the
industry's definition
was used in the rule.
-------
Proposal/
supplemental notice
Term definition
Bleaching All process equipment
component beginning with the
first application of
chlorine or chlorine-
containing compound up
to and including the
final bleaching stage.
Treatment with ozone,
oxygen, peroxide may
occur before or after
the addition of
chlorine. If treatment
occurs before this
chlorine addition, then
these stages are
included in the pulping
component; if treatment
occurs after the
addition of chlorine,
then these bleaching
stages are included in
the bleaching
component.
Industry recommendation
All process equipment
after high density pulp
storage prior to the
first application of
oxidizing, purification,
or reducing chemicals
following the pulping
component up to and
including the final
bleaching stage.
(A-92-40, IV-D1-104)
Final definition
Definition not needed;
included in the
definition of "bleaching
system."
Rationale for final
definition
Industry definition
addresses the use of
oxidizing or reducing
chemicals and provides
an equipment reference
for the start of the
bleaching component.
This is necessary to
accommodate the TCF
process. The rule
definition is a
combination of the
proposal and industry
definitions.
[from December
proposal)
1993
-------
Term
Bleaching
stage
Proposal/
supplemental notice
definition
None
Bleaching
system
None
Boiler
Any enclosed combustion
device that extracts
useful energy in the
form of steam. Boilers
are not considered
incinerators.
[from December 17, 1993
proposall
Industry recommendation
All process equipment
associated with a
discrete step in the
bleaching process,
including chemical and
steam mixers, bleaching
towers, washers, and seal
(filtrate) tanks.
(A-92-40, IV-D1-104)
None
Any enclosed combustion
device whose primary
purpose is the extraction
of useful energy in the
form of steam. Boilers
are not considered
incinerators.
(A-92-40, IV-D1-104)
Final definition
All process equipment
associated with a
discrete step of chemical
application and removal
in the bleaching process
including chemical and
steam mixers, bleaching
towers, washers, seal
(filtrate) tanks, and
vacuum pumps, and any
other equipment serving
the same functions as
those previously listed.
All process equipment
after high-density pulp
storage prior to the
first application of
oxidizing chemicals or
reducing chemicals
following the pulping
system, up to and
including the final
bleaching stage.
Any enclosed combustion
device that extracts
useful energy in the form
of steam. A boiler is
not considered a thermal
oxidizer.
Rationale for final
definition
The proposal did not
contain a definition of
a bleaching stage. The
industry definition
appears to be accurate
and the EPA agrees with
the commenter. For the
rule, the industry
definition was slightly
modified to address
"chemical application
and removal" and to
include "but not
limited to" language.
"Bleaching system"
better describes the
definition than
"bleaching component."
The EPA does not
believe that the
industry's definition
adds any clarity to the
rule. Therefore, the
proposal definition was
used in the rule.
However, the term
"incinerator" was
replaced by "thermal
oxidizer" to add
clarity.
-------
Term
Brownstock
washer system
Proposal/
supplemental notice
definition
Includes rotary vacuum
drum washers, pressure
washers, diffusion
washers, horizontal
belt washers, all
filtrate tanks, and
intermediate stock
chests. The washing
system does not include
deckers, screens, stock
chests or pulp storage
tanks following the
last stage of
brownstock washing.
[from March 8, 1996
Federal Register
supplemental noticel
Industry recommendation
The eguipment used to
wash pulp and separate
spent cooking chemicals
following the digester
system and prior to the
bleaching component,
oxygen delignification
system or paper machine
system (at unbleached
mills), such as vacuum
drum washers, diffusion
washers, rotary pressure
washers, horizontal belt
filters, intermediate
stock chests, and their
associated vacuum pumps,
filtrate tanks and foam
breakers or tanks. The
washing system does not
include deckers, screens,
stock chests, or pulp
storage tanks, following
the last stage of
brownstock washing.
(A-92-40, IV-D1-104)
Final definition
All eguipment used to
wash pulp and separate
spent cooking chemicals
following the digester
system and prior to the
bleaching system, oxygen
delignification system,
or paper machine system
(at unbleached mills).
The pulp washing system
eguipment includes vacuum
drum washers, diffusion
washers, rotary pressure
washers, horizontal belt
filters, intermediate
stock chests, and their
associated vacuum pumps,
filtrate tanks and foam
breakers or tanks, and
any other eguipment
serving the same function
as those previously
listed. The pulp washing
system does not include
deckers, screens,
knotters, stock chests,
or pulp storage tanks,
following the last stage
of pulp washing.
Rationale for final
definition
The industry definition
appears to be
technically correct.
The definition (pulp
washing system) used in
the rule was modified
to address all pulp
washing systems (i.e.,
separate definitions
for brown and red stock
would not be needed) by
removing the term
1'brownlt and replacing
with word "stock" with
the word "pulp"
-------
Term
Causticizing
system
Proposal/
supplemental notice
definition
I
-J
Chemical
recovery
The process by which
pulping chemicals in
the spent cooking
liguor are extracted or
recovered after the
multiple effect
evaporator system.
[from December 17, 1993
proposal]
Industry recommendation
All lime mud washers and
storage tanks, white and
mud liquor clarifiers and
storage tanks, slakers,
slaker grit washers, lime
kilns, green liquor
clarifiers and storage
tanks, and dreg washers
ending with the white
liquor storage tanks
prior to the digester
system.
The process by which
pulping chemicals in the
spent cooking liquor are
extracted or recovered
after the multiple effect
evaporator system,
consisting of a recovery
furnace, black liquor
oxidation (if any), black
liquor storage tanks, and
ending with the smelt
dissolving tank, and
associated equipment.
Final definition
All equipment associated
with converting sodium
carbonate into active
sodium hydroxide. The
equipment includes smelt
dissolving tanks, lime
mud washers and storage
tanks, white and mud
liquor clarifiers and
storage tanks, slakers,
slaker grit washers,
kilns, green liquor
clarifiers and storage
tanks, and dreg washers
ending with the white
liquor storage tanks
prior to the digester
system, and any other
equipment serving the
same function as those
previously listed.
Definition not needed.
Rationale for final
definition
No definition was
included at proposal.
The industry definition
appears to be
technically correct.
For the rule, the
industry definition was
slightly modified to
present function first,
followed by typical
lime equipment.
The term "chemical
recovery" is not used
in the rule.
(A-92-40, IV-D1-104)
-------
Term
Chip steamer
Closed-vent
system
i
CO
Proposal/
supplemental notice
definition
None
A system that is not
open to the atmosphere
and is composed of
piping, ductwork,
connections, and, if
necessary, flow-
inducing devices that
transport gas or vapor
from an emission point
to a control device.
[from December 17, 1993
proposal]
Industry recommendation
A separate vessel for the
purpose of preheating
wood chips prior to the
digester, using flash
steam from the digester
or live steam.
A system that does not
discharge to the
atmosphere during normal
operation and is composed
of piping, ductwork,
connections, and if
necessary, flow inducing
devices that transport
gas or vapor from an
emission point to a
control device.
(A-92-40, IV-D1-104)
Final definition
A vessel used for the
purpose of preheating or
pretreating wood chips
prior to the digester,
using flash steam from
the digester or live
steam.
The proposal definition
was used in the rule.
Rationale for final
definition
No definition was
included at proposal.
The industry definition
appears to be
technically correct,
however the rule
definition was slightly
modified to remove the
term "separate" to
acknowledge that chip
steamer vessels may be
integrated into the
digester system.
The industry definition
includes language
referring to normal
operation to address
concerns regarding
malfunctions and
safety-related venting.
This language does not
add any clarity to the
definition and is
unnecessary. Also, EPA
does not intend for the
definition to depend on
operating mode.
Therefore, the proposal
definition was used in
the rule.
-------
Proposal/
supplemental notice
Term definition
Combustion An individual unit of
device equipment, including
but not limited to, an
incinerator, lime kiln,
recovery furnace,
process heater, or
boiler, used for the
thermal oxidation of
organic hazardous air
pollutant vapors.
[from December 17, 1993
proposal]
Industry recommendation
An individual unit of
equipment, including but
not limited to, a thermal
oxidizer, lime kiln,
recovery furnace, process
heater, or boiler, used
for the thermal oxidation
of organic hazardous air
pollutant vapors.
Final definition
An individual unit of
equipment, including but
not limited to, a thermal
oxidizer, lime kiln,
recovery furnace, process
heater, or boiler, used
for the thermal oxidation
of organic hazardous air
pollutant vapors.
Rationale for final
definition
The industry definition
includes the terms
"incinerator" and
"thermal oxidizer."
The EPA believes that
an incinerator is
addressed by the "term"
thermal oxidizer.
Therefore, only the
term "thermal oxidizer"
was included in the
definition for the
rule.
-------
I
M
O
Proposal/
supplemental notice
Term definition
Condensate The practice of
Segregation generating, producing,
or isolating a high-HAP
concentration/low flow
rate condensate stream
Erom process vent
vapors or gases in
order to maximize the
HAP mass and minimize
the condensate volume
sent to subsequent
treatment
[from March 8, 1996
Federal Register
supplemental notice]
Industry recommendation
The practice of
generating, producing, or
isolating a high-HAP
concentration/low flow
rate condensate stream
from process vent vapors
or gases in order to
maximize the HAP mass and
minimize the condensate
volume sent to subsequent
treatment.
For the cases where
condensate segregation is
practiced, the
segregation process must
be operated such that
either:
Final definition
Definition not needed.
Rationale for final
definition
The concept of
condensate segregation
is incorporated into
the language of the
rule.
(a) The combined high
methanol fraction streams
from one or more sources
contain 50 percent of the
total methanol in the
foul condensate streams
from the same sources; or
(b) All foul condensate
streams when combined
contain a minimum of 10
Ib methanol/ADTP for
bleached mills or 6.4 Ib
nethanol/ADTP for
unbleached mills.
-------
Term
Container
Decker system
Proposal/
supplemental notice
definition
Container means any
portable unit in which
wastewater or HAP
removed from wastewater
is stored, transported,
treated, or otherwise
handled. Examples of
containers are drums,
barrels, tank trucks,
barges, dumpsters, tank
cars, dump trucks, and
ships.
[from December 17, 1993
proposal]
A piece of equipment
used to thicken or
reduce the water
content of the pulp
slurry after the pulp
washing system.
[from December
proposal]
17, 1993
Industry recommendation
Delete entire definition.
(A-92-40, IV-D1-104)
The equipment, including
filtrate tanks, primarily
used to thicken the pulp
slurry or reduce its
liquid content after the
brownstock washer system
and prior to high density
storage.
(A-92-40, IV-D1-104)
Final definition
Definition not needed.
Rationale for final
definition
Based on industry
comments, containers
are not used in the
pulp and paper
industry. Therefore,
the rule does not have
container requirements.
Equipment used to thicken
the pulp slurry or reduce
its liquid content after
the pulp washing system
and prior to high-density
pulp storage. The decker
system includes decker
vents, filtrate tanks,
and associated vacuum
pumps, and any other
equipment serving the
same function as those
previously listed.
The industry definition
includes references to
process equipment
location and appears to
be technically correct.
However, the terms
"primarily" and
"brown" were deleted
and a list of typical
equipment was were
added to the rule
definition to broaden
the definition. The
wording was modified to
present function first,
followed by typical
equipment.
-------
Proposal/
supplemental notice
Term definition
Digester Each continuous
system digester or each set of
batch digesters used
for the chemical
treatment of wood,
including associated
flash tank(s), blow
tank(s), chip
steamer(s),
condenser (s), and pre-
hydrolysis unit(s) .
[from December 17, 1993
proposal]
to
Industry recommendation
Each continuous digester
or each batch digester
used for the chemical
treatment of wood or non-
wood fibers, including
associated flash tank(s),
blow tank(s), chip
steamer(s), blow heat
accumulator(s),
condenser (s), and pre-
hydrolysis unit(s)
preceding brownstock
washers.
(A-92-40, IV-D1-104)
Final definition
Each continuous digester
or each batch digester
used for the chemical
treatment of wood or non-
wood fibers. The
digester system eguipment
includes associated flash
tank(s), blow tank(s),
chip steamer (s) not using
fresh steam, blow heat
accumulator(s) , relief
gas condenser (s), pre-
hydrolysis unit(s)
preceding the pulp
washing system, and any
other eguipment serving
the same function as
those previously listed.
The digester system
includes any of the
liguid streams or
condensates associated
with batch or continuous
digester relief, blow, or
flash steam processes.
Rationale for final
definition
The industry's
recommended definition
includes pulping of
non-wood fibers and
includes specific
eguipment. The EPA
agrees with the
commenter's revisions.
However, the words
"brownstock washer"
were replaced with
"pulp washing system"
and "but is not limited
to" was included.
Additionally, the
language was modified
to present function
first, followed by
typical equipment.
-------
Term
Proposal/
supplemental notice
definition
Industry recommendation
Final definition
Rationale for final
definition
Emission
point
Any location within a
source from which air
pollutants are emitted,
including an individual
process vent,
wastewater collection
and treatment system,
or an open piece of
process equipment.
[from December 17, 1993
proposal]
Any location within a
source from which air
pollutants are emitted to
the atmosphere, including
an individual process
vent, an open wastewater
collection and treatment
system unit, or an open
piece of process
equipment.
(A-92-40, IV-D1-104)
I
M
U>
Any part of a stationary
source that emits
hazardous air pollutants
regulated under this
subpart, including
emissions from individual
process vents, stacks,
open pieces of process
equipment, equipment
leaks, wastewater and
condensate collection and
treatment system units,
and those emissions that
could reasonably be
conveyed through a stack,
chimney, or duct where
such emissions first
reach the environment.
The proposed definition
was expanded to add
clarity.
Evaporator
system
Any and all equipment
associated with
increasing the solids
content of spent
cooking liquor
including, but not
limited to, pre-
evaporators,
evaporators (direct and
indirect contact), and
concentrators.
[from 1996 notice]
Any and all equipment
designed to increase the
solids content of spent
cooking liquor including,
but not limited to, pre-
evaporators, multi-effect
evaporators, and
concentrators.
All equipment associated
with increasing the
solids content and/or
concentrating spent
cooking liquor from the
pulp washing system
including pre-
evaporators, multi-effect
evaporators concentrators
and vacuum systems, as
well as, associated
condensers, hotwells, and
condensate streams, and
any other equipment
serving the same function
as those previously
listed.
The industry definition
includes some minor
changes that seem
appropriate and add
clarity to the
definition. The EPA
agrees with the
commenters' revisions.
The rule definition
also has minor language
changes in order to
remain consistent with
other definitions
(e.g., brownstock to
pulp).
-------
Term
Existing
source
Flow
indicator
Proposal/
supplemental notice
definition
None
A device which
indicates whether gas
flow is present in a
closed vent system.
[from December
proposal]
1993
Industry recommendation Final definition
For the purposes of this The most current general
subpart, a source covered provisions definition was
by this subpart that is used in the rule.
not a new source.
(A-92-40, IV-D1-104)
Any device that indicates
gas or liquid flow in an
enclosed system.
[Commenter
The industry definition
was used in the rule.
Rationale for final
definition
The industry definition
is slightly different
from that in the
general provisions,
however, the rule is
not the appropriate
mechanism for changing
the general provisions
definitions. Any
revisions to the
general provisions
should be accomplished
in the ongoing
litigation. The most
current general
provisions definition
is used.
The industry definition
includes liquid flow.
The EPA agrees with the
commenter's revision.
-------
Term
Proposal/
supplemental notice
definition
Industry recommendation
Final definition
Rationale for final
definition
Foul
condensates
Any liquid streams
originating from the
following process areas
or equipment: batch
digester relief and
blow gas system
condensates; batch
digester blow heat
recovery system
condensates; continuous
digester system flash
steam condensates;
continuous digester
chip steaming vessel
condensates; turpentine
decanter underflow; NCG
system condensates; NCG
system low point
drains; and condensates
from weak liquor feed
stage(s) in the
evaporator system.
Where vapors or gases
from the digester,
turpentine recovery,
NCG, and/or evaporator
systems are segregated
into low-HAP and high-
HAP concentration
fractions though
multistage,
differential, or
selective condensation,
only the high-HAP
fraction stream is
considered foul
condensate. If
condensate segregation
is not performed on the
The following liquid
streams are considered
foul condensates:
turpentine decanter
underflow, noncondensible
gas handling system
condensates, continuous
digester flash steam
condensates, batch
digester blow steam
condensates, batch
digester relief steam
condensates, evaporator
vacuum system
condensates, and
condensed vapors from
evaporator weak black
liquor feed stage(s)
(first liquor evaporation
step). Where condensate
segregation is practiced
and vapors from
digesters (flash, blow,
and relief steam), and/or
black liquor evaporators
are condensed through
multistage, differential
or selective
condensation, to produce
low methanol and high
methanol fractions, only
the concentrated stream
(high methanol fraction)
is considered foul
condensate.
Definition not needed.
The definition for foul
condensates was removed
from the rule since the
rule format was revised
to name streams to be
controlled.
Additionally, the term
"foul" pertains to the
presence of TRS
compounds, which may or
may not be indicative
of the presence of HAP
compounds.
-------
Term
Green liquor
Hardwood
Proposal/
supplemental notice
definition
Liquor made by
dissolvinq the sodium
containinq smelt from
the kraft recovery
process in water, prior
to causticizing.
[from December 17, 1993
proposal, preamble]
Pulpwood from broad-
leaved dicotyledonous
deciduous trees.
Industry recommendation
The solution made by
dissolvinq smelt
(primarily sodium sulfide
and sodium carbonate)
from the kraft recovery
process in water, prior
to causticizinq.
[Commenter 20,027]
Any species of broad-
leaved anqiosperms
possessinq true vessels.
Final definition
Definition not needed in
the rule.
Definition not needed.
Rationale for final
definition
The industry definition
is more technically
correct (i.e., solution
versus liquor).
However, the definition
was not needed in rule.
The term "hardwood" is
not used in the rule.
*>.
t
CTi
High volume,
low
concentration
or HVLC
collection
system
[from December 17, 1993
proposal, preamble]
None None
The qas collection and This term was included
transport system used to in the rule to
convey qases from the distinquish between the
HVLC system to a control HVLC collection system
device. and the HVLC system
vents.
-------
Term
Proposal/
supplemental notice
definition
Industry recommendation
Final definition
Rationale for final
definition
Incinerator
I
M
-J
An enclosed combustion
device that is used for
destroying organic
compounds. Auxiliary
fuel may be used to
heat waste gas to
combustion
temperatures. Any
energy recovery section
present is not
physically formed into
one manufactured or
assembled unit with the
combustion section;
rather, the energy
recovery section is a
separate section
following the
combustion section and
the two are joined by
ducts or connections
carrying flue gas.
[from December
proposal]
1993
Industry recommended
replacing this definition
with the definition for
thermal oxidizer.
(A-92-40, IV-D1-104)
The definition for
thermal oxidizer was used
in the rule.
The definition for
thermal oxidizer
replaces the
incinerator definition
because it was a
broader definition.
-------
Term
Individual
drain system
rfi
I
Proposal/
supplemental notice
definition
The system used to
convey process
wastewater streams from
pulping or bleaching
process eguipment or
tanks or process
wastewater collection
and treatment system
unit to a receiving
process wastewater
collection and
treatment system unit.
The term includes all
process drains and
junction boxes,
together with their
associated sewer lines
and other junction
boxes, manholes, sumps,
and lift stations, down
to the receiving
process wastewater
treatment system. The
individual drain system
shall be designed to
segregate the vapors
within the system from
the other drain
systems. A segregated
storm water sewer
system, which is a
drain and collection
system designed and
operated for the sole
purpose of collecting
rainfall-runoff at a
facility, and which is
segregated from all
other individual drain
| Industry recommendation
The system used to convey
process wastewater
streams from the pulping
component to a receiving
process wastewater
collection and treatment
system unit. The term
includes all process
drains and junction
boxes, together with
their associated sewer
lines and other junction
boxes, manholes, sumps,
and lift stations, to the
receiving process
wastewater treatment
system.
(A-92-40, IV-D1-104)
Final definition
Definition not needed.
Rationale for final
definition
This definition was not
included in the final
rule because the
original definition was
too burdensome and did
not adeguately reflect
the streams included in
this rule. Systems
used to convey
wastewater streams from
pulping and bleaching
systems are referred to
as hardpiping.
-------
Term
Junction box
Knotter
system
Proposal/
supplemental notice
definition
A manhole access point
to a wastewater sewer
system line or a lift
station.
[from December 17, 1993
proposal]
A piece of equipment
where knots or pieces
of uncooked wood are
removed from the pulp
slurry after the
digester system and
prior to the pulp
washing system.
Equipment used to
remove oversized
particles from pulp
following the pulp
washer are considered
screens.
[from December 17, 1993
proposal]
Industry recommendation
Any structure designed
for the conjunction of
two or more sewer lines.
A junction box may allow
access to the sewer
lines.
[Commenter 20,027]
All equipment where
knots, oversized
material, or pieces of
uncooked wood are removed
from the pulp slurry
after the digester system
and prior to the
brownstock washer system.
Pieces of equipment used
to remove oversized
particles from pulp
following the brownstock
washer are considered
screens.
(A-92-40, IV-D1-104)
Final definition
Definition not needed in
rule.
Equipment where knots,
oversized material, or
pieces of uncooked wood
are removed from the pulp
slurry after the digester
system and prior to the
pulp washing system. The
knotter system equipment
includes the knotter,
knot drainer tanks and
ancillary tanks, and any
other equipment serving
the same function as
those previously listed.
Rationale for final
definition
Although EPA agrees
that the industry
revision addresses the
fact that not all
junction boxes will
have manholes or allow
access, this definition
was deleted from the
rule since the control
options for wastewater
have been simplified.
The industry definition
adds the term
"oversized material".
The EPA agrees with the
commenter's revision.
For the rule, the
proposal definition has
added more specific
equipment associated
with knotter systems to
provide greater clarity
and brownstock has been
replaced with pulp to
broaden the definition.
Additionally, the
language was modified
to present function
first, followed by
typical equipment.
-------
Term
Proposal/
supplemental notice
definition
Industry recommendation
Final definition
Rationale for final
definition
Kraft
recovery
furnace
An enclosed combustion
device where
concentrated spent
liquor is burned to
recover sodium and
sulfur, produce steam,
and dispose of unwanted
dissolved wood
components in the
liquor.
[from December 17, 1993
proposal]
Delete entire definition.
(A-92-40, IV-D1-104)
The term "recovery
furnace" replaced "kraft
recovery furnace."
Industry recommended
deleting this
definition; however,
"recovery furnace"
replaces "kraft
recovery furnace" to
broaden the definition.
The EPA maintains that
the definition for
"recovery furnace" is
needed because it is
specifically mentioned
in the rule.
Lime kiln
i
CO
o
An enclosed combustion
device used to calcine
lime mud, which
consists primarily of
calcium carbonate, into
calcium oxide.
[from December 17, 1993
proposal]
None
The proposal definition
was used in the rule.
The proposal definition
appropriately defines
the lime kiln.
-------
Term
Low volume,
high
concentration
or LVHC
collection
system
NJ
Proposal/
supplemental notice
definition
Includes batch digester
blow vents; batch
digester relief steam
condenser vents;
continuous digester
relief steam vents;
turpentine condenser(s)
vents; continuous
digester blow tank
vent; evaporator vacuum
system vents; liquor
concentrator vacuum
system vents; pre-
evaporator vacuum
system vents; steam
stripper feed tank
vents; and steam
stripper off gas vents.
[from March 8, 1996
Federal Register
supplemental notice]
Industry recommendation
Includes batch digester
blow tank and/or blow
heat recovery vents;
batch digester relief
condenser vents;
continuous digester blow
tank and/or blow heat
recovery vents;
continuous digester
relief condenser vents;
black liquor pre-
evaporator; evaporator;
and concentrator vacuum
systems vents; foul
condensate off gas vents;
and foul condensate
storage tank vents.
(A-92-40, IV-D1-104)
Final definition
The gas collection and
transport system used to
convey gases from the
LVHC to a control device.
Rationale for final
definition
The proposal definition
was modified to
distinguish between the
LVHC collection system
and the LVHC system
vents.
-------
Term
Malfunction
Proposal/
supplemental notice
definition
None
i
t\>
to
Mechanical
pulping
None
Industry recommendation
Any sudden and not
reasonably preventable
failure of air pollution
control equipment, a
process, or process
equipment to operate in a
normal or usual manner,
or the ventinq of
equipment for safety
reasons. Failures that
are caused by poor
maintenance or careless
operation and not
malfunctions.
None
Final definition
The most current qeneral
provisions definition was
used.
A pulpinq process that
only uses mechanical and
thermo-mechanical
processes to reduce wood
to a fibrous mass. The
mechanical pulpinq
processes include, but
are not limited to, stone
qroundwood, pressurized
qroundwood, refiner
mechanical, thermal
refiner mechanical,
thermo-mechanical, and
tandem thermo-mechanical.
Rationale for final
definition
The industry definition
adds lanquaqe for
safety ventinq to the
qeneral provisions
definition. However,
the rule is not the
appropriate mechanism
for chanqinq the
qeneral provisions
definitions. Any
revisions to the
qeneral provisions
should be accomplished
in the onqoinq
litigation. The most
current qeneral
provisions definition
was used.
Definition needed for
final rule. This
process was not
addressed in the
proposed rule.
-------
Term
Multiple-
effect
evaporator
system
to
UJ
New kraft
recovery
furnace
Proposal/
supplemental notice
definition
A series of evaporators
operated at different
pressures such that the
vapor from one
evaporator body becomes
the steam supply for
the next evaporator,
and associated
condenser(s) and
botwell(s) used to
concentrate the spent
cooking liquid that is
separated from the
pulp.
[from December
proposal]
None
1993
Industry recommendation
Delete definition.
(A-92-40, IV-D1-104)
Final definition
This definition was
replaced with the
evaporator system
definition.
A kraft recovery furnace
located at an existing
source covered by this
subpart, on which
construction or
reconstruction is
commenced after (proposal
date for MACT II) or a
kraft recovery furnace
located at a new source.
Definition not needed.
Rationale for final
definition
The industry recommends
deleting this
definition from the
rule since evaporator
system will be defined.
The EPA agrees with the
commenter.
The rule does not
define new or existing
equipment but
references the results
from the general
provisions litigation.
(A-92-40, IV-D1-104)
-------
Term
New source
Proposal/
supplemental notice
definition
None
i
to
Non-
condensible
gas system
Non-wood
pulping
None
Industry recommendation
For purposes of this
subpart, an affected
source on which
construction or
reconstruction is
commenced after December
17, 1993. A unit process
or component added to or
modified at an existing
facility is not a new
source, unless such
addition or
reconstruction is so
large as to make the
entire facility a new
source by virtue of the
definition of
"reconstruction" in this
subpart.
(A-92-40, IV-D1-104)
None
Final definition
Rationale for final
definition
The most current general The industry definition
provisions definition was is different from that
used in the rule. in the general
provisions. However,
the rule is not the
appropriate mechanism
for changing the
General Provision
definition. Any
revisions to the
general provisions
should be accomplished
in the ongoing
litigation. The most
current general
provisions definition
was used.
Definition not needed.
Includes pulping of flax
straw, cereal straw,
bagasse, hemp, cotton,
jute, kenaf, grasses,
leaf fibers, or secondary
fiber repulping.
[Commenter IV-D2-141
The production of pulp
from fiber sources other
than trees. The non-wood
sources include, but are
not limited to, bagasse,
cereal straw, cotton,
flax straw, hemp, jute,
kenaf, and leaf fibers.
This definition was
replaced with
definitions for LVHC
and HVLC collection
s ys terns .
The industry definition
appears to be
technically correct.
The EPA agrees with the
commenter. Language
was added to the
beginning of the
definition for clarity.
-------
Term
Nuisance
scrubber
Operating
parameter
value
i
to
ui
Proposal/
supplemental notice
definition
A minimum or maximum
value established for a
control device or
process parameter if
achieved by itself, or
in combination with one
or more other operating
parameter values;
determines that an
owner or operator has
complied with an
applicable emission
limitation or standard.
Industry recommendation
Final definition
A device which circulates Definition not
a liquid solution to
remove pollutants from a
gaseous vent stream. The
effluent from a nuisance
scrubber is sewered and
not recovered for cooking
acid production.
needed.
(A-92-40, IV-D1-104)
A minimum or maximum
value established for a
control device or process
parameter which, if
achieved by itself, or in
combination with one or
more other operating
parameter values; is an
indication that an owner
or operator has complied
with an applicable
emission limitation or
standard.
Definition not needed.
Rationale for final
definition
The industry definition
appears to be
technically correct.
However, this
definition was not
needed since the rule
does not specify or
identify control
technologies for
sulfite mills.
"Operating parameter
value" was incorporated
into the rule;
therefore, the
definition was not
needed.
[from December 17,
proposal]
1993
;A-92-40, IV-D1-104)
-------
Term
Oven-dried
pulp
Proposal/
supplemental notice
definition
None
Industry recommendation
None
i
10
Final definition
A pulp sample at zero
percent moisture content
by weight. Pulp samples
for applicability or
compliance determinations
for both the pulping and
bleaching systems shall
be unbleached pulp. For
purposes of complying
with mass emission limits
in this subpart, megagram
of ODP shall be measured
to represent the amount
of pulp entering and
processed by the
equipment system under
the specified emission
limit. For equipment
that does not process
pulp, megagram of ODP
shall be measured to
represent the amount of
pulp that was processed
to produce the gas and
liquid streams that the
subject equipment is
processing.
Rationale for final
definition
This term was needed to
define the units in the
rule.
-------
Term
Oxygen
deligni-
fication
Proposal/
supplemental notice
definition
Includes the blow tank,
the post oxygen
washers, filtrate
tanks, and any
interstage pulp storage
tanks.
[from March 8, 1996
Federal Register
supplemental noticel
to
Papermaking
component
None
Industry recommendation
The equipment that uses
oxygen to remove lignin
from pulp after
brownstock high density
storage and prior to the
bleaching component. The
oxygen delignification
system includes the blow
tank, the post oxygen
washers, filtrate tanks,
and any interstage pulp
storage tanks.
(A-92-40, IV-D1-104)
All of the equipment used
to convert pulp into
paper, paperboard, or
market pulp, including
the stock storage and
preparation systems (such
as pulp mixing and
dispersion, beating and
refining, and addition of
additives), the paper or
paperboard machine "wet
end" systems (including
sheet formation, pressing
and vacuum systems),
paper machine white water
systems, broke recovery
systems, and "dry end"
systems (including
drying, calendering, on-
machine coating, winding,
slitting, and cutting).
Final definition
The equipment that uses
oxygen to remove lignin
from pulp after high-
density stock storage and
prior to the bleaching
system. The oxygen
delignification system
equipment includes the
blow tank, washers,
filtrate tanks, and any
interstage pulp storage
tanks, and any other
equipment serving the
same function as those
previously listed.
Definition not needed; it
was included in the
definition of
"papermaking system. "
Rationale for final
definition
The industry definition
appears technically
correct and adds
specific equipment for
reference. The EPA
agrees with the
commenter. The
industry definition was
slightly modified to
remove "brownstock" and
add "but is not limited
to" to broaden the
definition in the final
rule.
This definition was
included in the
"papermaking system"
definition.
"Papermaking system"
better describes the
definition than
"papermaking
component."
-------
Term
Papermaking
system
Proposal/
supplemental notice
definition
None
to
03
Part 70
permit
None
Industry recommendation
None
A permit issued by a
state permitting
authority pursuant to a
program approved by EPA
under part 70 of this
chapter.
(A-92-40, IV-D1-104)
Final definition
All eguipment used to
convert pulp into paper,
paperboard, or market
pulp, including the stock
storage and preparation
systems, the paper or
paperboard machines, the
paper machine white water
system, broke recovery
systems, and the systems
involved in calendering,
drying, on-machine
coating, slitting,
winding, and cutting.
The most current general
provisions definition was
used.
Rationale for final
definition
"Papermaking component"
was replaced by
"papermaking system"
because "papermaking
system" better
describes the
definition. The EPA
maintains that
industry's proposed
definition of
"papermaking component"
was too specific and
was modified to be more
broad and also renamed
"papermaking system."
The industry definition
is different from that
in the general
provisions. However,
the rule is not the
appropriate mechanism
for changing the
general provisions
definition. Any
revisions to the
general provisions
should be accomplished
in the ongoing
litigation. The most
current general
provisions definition
was used.
-------
Term
Proposal/
supplemental notice
definition
Industry recommendation
Final definition
Rationale for final
definition
Point of
generation
i
to
The location where the
process wastewater
stream exits the
pulping or bleaching
process equipment or
tank prior to mixing
with other process
wastewater streams or
prior to handling or
treatment in a piece of
equipment that is not
an integral part of the
pulping or bleaching
process equipment. A
piece of equipment is
an integral part of the
process if it is
essential to the
operation of the
process (i.e., removal
of the equipment would
result in the process
being shut down).
Delete definition
entirely.
(A-92-40, IV-D1-104)
Definition not needed.
The EPA agrees with the
commenter that this
definition is not
needed since the format
of the rule has been
revised to naming
specific streams to be
controlled.
[from December 17,
proposal]
1993
Pre-washing
screening
system
Includes knotters,
knotter drain tanks,
screens, and reject
tanks prior to
brownstock washing.
[from March 8, 1996
Federal Register
supplemental notice]
None
Definition not needed.
This term is not used
in the rule.
-------
Term
Proposal/
supplemental notice
definition
Industry recommendation
Final definition
Rationale for final
definition
Primary fuel
The fuel that provides
the principal heat
input to the combustion
device. To be
considered primary, the
fuel must be able to
sustain operation of
the combustion device
without the addition of
other fuels.
[from December 17, 1993
proposal]
The combination of fuels
that provides the
principal heat input to
the combustion device.
To be considered primary,
the fuel must be able to
sustain operation of the
combustion device without
the addition of other
fuels.
(A-92-40, IV-D-97)
The proposal definition
was used in the rule.
The EPA disagrees with
the industry revision
because primary fuel is
meant to imply a single
fuel, and not a
combination of fuels.
The proposal definition
was used in the rule.
i
u>
o
-------
Term
Proposal/
supplemental notice
definition
Industry recommendation
Final definition
Rationale for final
definition
Process
emission
point
A gas stream that
contains hazardous air
pollutants discharged
during operation of
process eguipment
including, but not
limited to digesters,
evaporators, pulp
washing systems,
bleaching towers,
bleaching stage
washers, and associated
filtrate tanks.
[from December 17, 1993
proposal]
The location where a gas
stream that contains
hazardous air pollutants
is discharged from the
process eguipment in the
pulping component,
bleaching component,
process wastewater
component, chemical
recovery
component/system,
papermaking component, or
causticizing system as
defined in this section.
Process emission points
include gas streams that
are discharged directly
to the atmosphere,
discharge to the
atmosphere via vents or
open process equipment,
or after diversion
through a product
recovery device.
Definition not needed.
The EPA agrees with the
industry revisions,
however, this
definition was not
needed since the format
of the rule has been
revised to name
specific vents to be
controlled.
-------
Proposal/
supplemental notice
Term definition
Process A piece of equipment,
wastewater structure, or transport
collection mechanism used in
system conveying or storing a
process wastewater
stream. Examples of a
process wastewater
collection system
equipment include
individual drain
systems, wastewater
tanks, surface
impoundments, or
containers.
Industry recommendation
A grouping of equipment,
structures, or transport
mechanisms used in
conveying or storing a
process wastewater
stream. Examples of
process wastewater
collection system
equipment include foul
condensate drain systems,
wastewater tanks, or
surface impoundments.
Final definition
Definition not needed.
Rationale for final
definition
This definition was not
needed since the
pulping wastewater
control options have
been simplified.
i
U)
to
[from December 17, 1993
proposal]
Process Air emissions from all
wastewater process wastewater
component streams produced from
the pulping and
bleaching processes.
[from December
proposal]
1993
Delete definition.
(A-92-40, IV-D1-104)
Definition not needed.
This definition was not
needed since the format
of the rule has been
revised to name
specific vents to be
controlled.
-------
Term
Process
wastewater
stream
i
w
UJ
Process
wastewater
treatment
component
Proposal/
supplemental notice
definition
Any HAP-containing
liquid that results
from either direct or
indirect contact of
water with organic
compounds. Examples of
a process wastewater
stream include, but are
not limited to,
digester condensates,
evaporator condensates,
and NCG system
condensates.
[from December 17, 1993
proposal 1
None
Industry recommendation Final definition
Any HAP-containing liguid Definition not needed.
that results from contact
of water with organic
compounds.
(A-92-40, IV-D1-104)
Rationale for final
definition
This definition was not
needed since the format
of the rule has been
revised to name
specific vents to be
controlled.
A collection of equipment
or structures, a process,
or specific technique
that conveys or removes
or destroys any HAP in a
process wastewater
stream. Examples
include, but are not
limited to, a steam
stripping unit, or a
biological treatment
unit.
Definition not needed;
replaced by "process
wastewater treatment
system."
The definition for
process wastewater
treatment system
replaces "process
wastewater treatment
component."
(A-92-40, IV-D1-104)
-------
Proposal/
supplemental notice
Term definition
Process A process or specific
oastewater technique that removes
treatment or destroys the
system organics or any HAP in
a process wastewater
stream. Examples
include, but are not
limited to, a steam
stripping unit,
wastewater incinerator,
or biological treatment
unit.
Industry recommendation
None
Final definition
A collection of
equipment, a process, or
specific technique that
removes or destroys the
HAP's in a process
wastewater stream.
Examples include, but are
not limited to, a steam
stripping unit,
wastewater thermal
oxidizer, or biological
treatment unit.
Rationale for final
definition
"Process wastewater
treatment component"
was changed to "process
wastewater treatment
system" because
"process wastewater
treatment system"
better describes the
definition.
[from December
proposal]
1993
u>
-------
Term
Pulp washing
system
i
u»
in
Proposal/
supplemental notice
definition
Pulp or brownstock
washers and associated
vacuum pumps, filtrate
tanks, and foam
breakers or tanks used
to wash the pulp to
separate spent cooking
chemicals following the
digestion system and
prior to the bleaching
component.
[from December 17, 1993
proposal)
Industry recommendation
None
Final definition
All equipment used to
wash pulp and separate
spent cooking chemicals
following the digester
system and prior to the
bleaching system, oxygen
delignification system,
or paper machine system
(at unbleached mills).
The pulp washing system
equipment includes vacuum
drum washers, diffusion
washers, rotary pressure
washers, horizontal belt
filters, intermediate
stock chests, and their
associated vacuum pumps,
filtrate tanks, and foam
breakers or tanks, and
any other equipment
serving the same function
as those previously
listed. The pulp
washing system does not
include deckers, screens,
knotters, stock chests,
or pulp storage tanks,
following the last stage
of pulp washing.
Rationale for final
definition
The definition used in
the rule was modified
to address all pulp
washing systems (i.e.,
separate definitions
for brown and redstock
would not be needed) by
removing the term
"brown" and replacing
with word "stock" with
the word "pulp".
-------
Term
Proposal/
supplemental notice
definition
Industry recommendation
Final definition
Rationale for final
definition
Pulping
component
i
u>
en
All process equipment,
beginning with the
digester system, and up
to and including the
last piece of pulp
conditioning equipment
prior to the bleaching
component, including
treatment with ozone,
oxygen, or peroxide
before the first
application of chlorine
or chlorine-containing
compounds.
[from December
proposal]
1993
The wood storage and
preparation area
(including debarking and
chipping), the digester
system, knotter systems,
brownstock washer system,
pulp storage, turpentine
recovery system, multiple
effect evaporator system,
causticizing systems,
weak and strong black
liquor storage tanks,
tall oil recovery system,
oxygen delignification
system, deckers and
screens. The pulping
component ends with the
last stage of brownstock
washing, deckers and/or
screens, or the last
stage of post-oxygen
washing.
Definition not needed;
included in definition of
"pulping system."
Industry indicated that
they want to include
wood storage and
preparation. The EPA
disagrees because the
rule does not address
emissions from wood
storage and preparation
areas. This definition
was incorporated into
the "pulping system"
definition.
-------
Term
Pulping
process
condensates
Proposal/
supplemental notice
definition
Industry recommendation
None
Final definition
Any HAP-containing liguid
that results from contact
of water with organic
compounds in the pulping
process. Examples of a
process condensates
stream include digester
system condensates,
evaporator system
condensates, LVHC, and
HVLC system condensates,
and any other condensates
from eguipment serving
the same function as
those previously listed.
Liquid streams that are
intended for by-product
recovery are not
considered process
condensate streams.
Rationale for final
definition
"Process wastewater
stream" was replaced by
"pulping process
condensates" because
"pulping process
condensates" better
describes the
definition.
-------
Term
Proposal/
supplemental notice
definition
Pulping
system
None
Industry recommendation
None
03
Purchased
pulp
Virgin pulp purchased
from an off-site
facility or obtained
from an inter-company
transfer from another
site.
[from December 17, 1993
proposal, preamble]
Pulp purchased from an
off-site facility or
obtained from an inter-
company transfer from
another site.
[Commenter 20,027]
Final definition
All process equipment,
beginning with the
digester system, and up
to and including the last
piece of pulp
conditioning equipment
prior to the bleaching
system, including
treatment with ozone,
oxygen, or peroxide
before the first
application of a chemical
bleaching agent intended
to brighten pulp. The
pulping system includes
pulping process
condensates and can
include multiple pulping
lines.
Definition not needed.
Rationale for final
definition
"Pulping component" was
changed to "pulping"
system" because
"pulping system" better
describes the
definition.
The term is not used in
the rule.
-------
I
U)
vo
Term
Reconstruc-
tion
Proposal/
supplemental notice
definition
None
Industry recommendation
The replacement of
components at a source
subject to this subpart
to such an extent that:
(1) The fixed capital
cost of the new
components exceeds 50% of
the fixed capital cost
that would be required to
construct a comparable
new source; and (2) It is
technologically and
economically feasible for
the reconstructed source
to meet the relevant
standard (s) established
in this subpart. Any
reconstructed source is
subject to relevant
standards for new
sources, including
compliance dates,
irrespective of any
change in emissions or
hazardous air pollutant
from that source.
(A-92-40, IV-D1-104)
Final definition
The most current general
provisions definition was
used in the rule .
Rationale for final
definition
The industry definition
adds language for
safety venting.
However, the rule is
not the appropriate
mechanism for revising
the general provisions
definitions. Any
revisions to the
general provisions
should be accomplished
in the ongoing
litigation. The most
current general
provisions definition
was used.
-------
Proposal/
supplemental notice
Term definition
Recovery An individual unit of
device equipment, such as an
absorber or a
condenser, capable of
and used for the
purpose of recovering
chemicals for use,
reuse, or sale.
Industry recommendation
Delete definition
entirely.
(A-92-40, IV-D1-104)
Final definition
Definition not needed.
Rationale for final
definition
This term is not used
in the rule.
Recovery
furnace
i
*>
o
Redstock
washer system
[from December 17,
1993 proposal]
An enclosed combustion
device where
concentrated spent
liquor is burned to
recover sodium and
sulfur, produce steam,
and dispose of unwanted
dissolved wood
components in the
liquor.
[from December 17,
proposal]
None
1993
Delete definition.
(A-92-40, IV-D1-104)
The proposal definition
was used in the rule.
The EPA believes this
definition was
necessary since
recovery furnaces are
referenced in the
pulping control
options. "Recovery
furnace" replaces
"kraft recovery
furnace" because
"recovery furnace" is a
broader definition.
The equipment used to Definition not needed. The brownstock and
wash sulfite pulp and to "Red stock washer system" redstock washer system
separate spent sulfite was included in the "pulp definitions have been
liquor (which is returned washing system" incorporated into the
for recovery) following definition. "pulp washing system"
the digester system. definition.
-------
Term
Relief valve
M
•IS.
I
•to
system
Proposal/
supplemental notice
definition
A valve used only to
release an unplanned,
nonroutine discharge.
A relief valve
discharge can result
from an operator error,
a malfunction such as a
power failure or
eguipment failure, or
other unexpected cause
that reguires immediate
venting of gas from
process eguipment to
avoid safety hazards or
equipment damage.
[from December 17, 1993
proposal]
A piece of process
eguipment where pieces
of oversized particles
are removed from the
pulp slurry after the
pulp washing system and
prior to the
papermaking eguipment.
Eguipment used to
remove uncooked wood
prior to the pulp
washing system are
considered knotters.
[from December
proposal]
1993
Industry recommendation
None
Final definition
Delete the definition.
Rationale for final
definition
This term is not used
in the rule.
A piece of process
eguipment in which
oversized particles are
removed from the pulp
slurry after the
brownstock washer system
and decker system and
prior to the bleaching or
paper machine component
washed stock storage.
Pieces of eguipment used
to remove knots,
oversized materials, or
pieces of uncooked wood
prior to the brownstock
washer system are
considered knotters.
All eguipment in which
oversized particles are
removed from the pulp
slurry prior to the
bleaching or papermaking
system washed stock
storage.
The EPA agrees that the
industry definition
adds clarity to the
rule. The proposal
definition was slightly
modified (e.g.,
"brownstock" to "pulp")
to broaden the
definition.
-------
Term
Secondary
Fiber pulping
Ja>
to
Segregated
condensate
stream (high-
HAP fraction)
Proposal/
supplemental notice
definition
None
Any condensate stream
that contains at least
65 percent by weight of
the total HAP mass
(measured as methanol)
that is present in the
vapor stream prior to
condensation or
isolation.
Industry recommendation
None
No definition was
provided by industry.
However, they indicated
in the July 9, 1996
letter that the percent
split between high and
low fractions should be
50/50 (not 65/35) .
Final definition
A pulping process that
converts a fibrous
material, that has
previously undergone a
manufacturing process,
into pulp stock through
the addition of water and
mechanical energy. The
mill then uses that pulp
as the raw material in
another manufactured
product. These mills may
also utilize chemical,
heat, and mechanical
processes to remove ink
particles from the fiber
stock.
Definition not needed.
Rationale for final
definition
Definition needed for
final rule. This
source was not in the
proposed rule.
This term is
incorporated into the
rule language.
[from March 8, 1996
Federal Register
supplemental notice]
-------
Term
Proposal/
supplemental notice
definition
Industry recommendation
Final definition
Rationale for final
definition
Semi-chemical
pulping
A pulping process that
combines both chemical
and mechanical pulping
processes.
[from December 17, 1993
proposal]
A pulping process that
combines both chemical
and mechanical pulping
processes with typical
pulping yields of 65
percent or greater based
on the dry weight of
pulpwood.
(A-92-40, IV-D1-104)
A pulping process that
combines both chemical
and mechanical pulping
processes. The semi-
chemical pulping process
produces intermediate
yields ranging from 55 to
90 percent.
The EPA agrees with the
commenter that the
definition needs to be
clarified to
distinguish semi-
chemical pulping from
mechanical pulping
where small amounts of
chemicals are used.
For the rule, the
proposal definition was
revised to reflect this
difference.
Sewer line
A lateral, trunk line,
branch line, or other
conduit including, but
not limited to, grates,
and trenches used to
convey process
wastewater streams or
any HAP removed from
process wastewater
streams to a downstream
unit in the process
wastewater collection
and treatment system.
[from December 17, 1993
proposal]
A lateral line, trunk
line, branch line, or
other conduit including,
but not limited to,
grates, and trenches used
to convey process
wastewater streams to a
downstream unit in the
process wastewater
collection and treatment
system.
(A-92-40, IV-D1-104)
Definition not needed.
This definition was not
included in the final
rule because it is too
burdensome and did not
adequately reflect the
streams included in the
rule. Systems used to
convey wastewater
streams from pulping
and bleaching systems
will be referred to as
hardpiping.
Soda pulping
A chemical pulping
process that uses
sodium hydroxide as the
active chemical in the
cooking liquor.
[from December 17, 1993
proposal]
None
The proposal definition
was used in the rule.
The proposal term
appropriately defines
the soda pulping
process.
-------
Term
Spent liquor
Steam
stripper
system
Sulfite
pulping
Proposal/
supplemental notice
definition
Cooking liquor from a
digestion or pulp-
washer process,
containing dissolved
organic wood materials
and residual cooking
compounds.
[from December 17, 1993
proposal]
A column, and
associated condensers
or heat exchangers,
used to strip compounds
from wastewater, using
air or steam.
[from December 17, 1993
proposal]
A chemical pulping
process that uses a
mixture of sulfurous
acid and bisulfite ion
as the cooking liquor.
Industry recommendation
Process liquid generated
from the separation of
black liquor from pulp by
the pulp washing process,
containing dissolved
organic wood materials
and residual cooking
compounds.
(A-92-40, IV-D1-104)
A column, and associated
stripper feed tank,
condensers or heat
exchangers, used to
remove compounds from
foul condensate, using
air or steam.
(A-92-40, IV-D1-104)
None
Final definition
Process liquid generated
from the separation of
cooking liquor from pulp
by the pulp washing
system containing
dissolved organic wood
materials and residual
cooking compounds.
A column (including
associated stripper feed
tank, condensers, or heat
exchangers) , used to
remove compounds from
wastewater or condensates
using steam. The steam
stripper system also
contains all equipment
associated with a
methanol rectification
process including
rectifiers, condensers,
decanters, and storage
tanks, and any other
equipment serving the
same function as those
previously listed.
The proposal definition
was used in the rule.
Rationale for final
definition
The industry definition
appears technically
correct. The final
definition was a
modified version of
industry's definition
with "pulp washing
system" replacing "pulp
washing process."
The EPA believes that
industry comments
improve the definition
and have been
incorporated into the
definition.
Additionally, the
definition was revised
to address methanol
rectification.
The proposed term
appropriately defines
the sulfite pulping
process.
[from December
proposal]
1993
-------
I
rfi
U1
Proposal/
supplemental notice
Term definition
Surface A unit which is a
impoundment natural topographic
depression, manmade
excavation, or diked
area formed primarily
of earthen materials
(although it may be
lined with manmade
materials), which is
used for the purpose of
treating, storing, or
disposing of wastewater
and is not an injection
well. Examples of
surface impoundments
are equalization,
settling, and aeration
pits, ponds, and
lagoons.
[from December 17, 1993
proposal]
Temperature A piece of equipment
monitoring used to monitor
device temperature and having
an accuracy of +1
percent of the
temperature being
monitored expressed in
degrees Celsius or +0.5
degrees Celsius (°C) ,
whichever is greater.
[from December
proposal]
Industry recommendation
Delete the definition.
(A-92-40, IV-D1-104)
Final definition
Definition not needed.
Rationale for final
definition
This term is not used
in the rule.
A piece of equipment used A piece of equipment used The EPA agrees with the
to monitor temperature
and having an accuracy of
+1.0 percent of the
temperature being
monitored expressed in
degrees Celsius or +0.5
degrees Celsius (°C) ,
whichever is greater.
(A-92-40, IV-D1-104)
to monitor temperature
and having an accuracy of
+1.0 percent of the
temperature being
monitored expressed in
degrees Celsius or +0.5
degrees Celsius (°C) ,
whichever is greater.
(A-92-40, IV-D1-104)
industry definition.
1993
-------
Term
Thermal
oxidizer
Proposal/
supplemental notice
definition
Industry recommendation
Thermal oxidizer means an
enclosed combustion
device that is designed
for thermally oxidizing
gaseous organic
compounds.
(A-92-40, IV-D1-104)
rfi
I
Turpentine
recovery
system
None
The decanters and storage
tanks used for recovering
turpentine from the
digester system.
(A-92-40, IV-D1-104)
Final definition
Rationale for final
definition
Thermal oxidizer means an Industry indicated that
enclosed combustion
device that destroys
organic compounds by
thermal oxidation.
All eguipment associated
with recovering
turpentine from digester
system gases including
condensers, decanters,
and storage tanks, and
any other eguipment
serving the same function
as those previously
listed. The turpentine
recovery system includes
any liguid streams
associated with the
turpentine recovery
process such as
turpentine decanter
underflow. Liguid
streams that are intended
for byproduct recovery
are not considered
turpentine recovery
system condensate
streams.
the term thermal
oxidizer includes
incinerators. The EPA
agrees with the
commenter. Therefore
the rule definition for
"thermal oxidizer"
includes incinerators.
Also, the auxiliary
fuel and energy
recovery language in
the incinerator
definition was removed
since it is not needed.
The industry definition
is specific to
decanters and storage
tanks. The EPA
modified the definition
to add more clarity,
specifically including
eguipment used in
turpentine recovery
systems.
-------
Term
Proposal/
supplemental notice
definition
Weak black None
Liquor
storage tanks
Weak liquor None
storage tanks
Working day None
Industry recommendation
All storage tanks
containing black liquor
recovered from the
brownstock washer system
and prior to the multiple
effect evaporator system.
(Brownstock or decker
filtrate tanks are not
weak black liquor storage
tanks.)
(A-92-40, IV-D1-104)
None
Final definition
Definition not needed.
Any day on which the
federal government
offices are open for
normal business.
Saturdays, Sundays, and
official federal holidays
are not working days.
Any storage tanks except
washer filtrate tanks
containing spent
recovered from the
pulping process and prior
to the evaporator system.
Definition not needed.
Rationale for final
definition
The EPA agrees with the
industry definition.
For the rule, the
definition was slightly
modified so that it is
not specific to kraft
mills only.
Furthermore, weak black
liquor storage tanks
were included in the
definition for weak
liquor storage tanks.
This definition
replaces weak black
liquor storage tanks.
The definition was
expanded such that it
was not specific to
kraft mills.
Industry indicated that
this definition was
left out of the general
provisions, however,
the rule is not the
mechanism for changing
general provisions
definitions. Any
revisions to the
general provisions
should be accomplished
in the ongoing
litigation. This
definition is not
needed in the rule.
-------
-------
15.0 INTEGRATED RULE INTERACTION
15.1 GENERAL
Comment: Several commenters (20,019, 20,027, 20,039,
20,051, 20,057A2, 20,088, 20,089, 20,091, 20,115, 20,153)
supported the concept of an integrated rule. However, several
commenters (20,027, 20,039, 20,057A2, 20,059, 20,088, 20,115)
criticized combining air and water regulations into one rule
because they believed it failed to fully consider the cross-media
impacts of each of the regulations.
One commenter (20,091) stated that they were impressed with
the coordinated effort by EPA to develop air and effluent
guidelines but thought that pollution prevention should be
carried further. One commenter (20,088) supported EPA's effort
to combine air and water regulations, but stated that EPA should
use a life cycle analysis, or holistic approach to evaluate the
effectiveness of the combined rule. Several commenters
(20,049A3, 20,059, 20,082, 20,129, 20,132, 20,133) indicated that
EPA has authority under the Act to establish emission limits for
non-HAP's. One commenter (20,049A3) contended that EPA has the
authority to set limits for other pollutants under the Act and
should propose enforceable emissions limits for all air
pollutants of concern including criteria and non-conventional
pollutants. One commenter (20,059) cited section lll(d)(l) of
the Act as authorization for EPA to establish "existing source"
15-1
-------
performance standards to control non-criteria pollutants, such as
TRS, and to require State Implementation Plans (SIP) to
incorporate these standards.
One commenter (20,122) argued that a truly cross-media
rulemaking would consider the impacts on workers, products,
chemical accident potential, and hazardous waste generation.
Response: The EPA disagrees with the commenters' suggestion
that the cross-media impacts associated with the combined rule
have not been addressed. All of the information submitted to the
Agency following the December 17, 1993 proposed rule and the
March 8, 1996 supplemental notice has been considered in
developing the final rule. For example, the effluent limitation
guidelines and standards, established by EPA's Office of Water,
have the potential to increase the solids loading sent to the
recovery process. This scenario was considered by OW in
developing their cost and benefit analysis.
In another case, the MACT standards and the effluent
limitation guidelines and standards require 100 percent
substitution. This process modification would reduce the
chlorine and chlorinated HAP ' s being sent to the bleach plant
scrubber. Comments received following proposal indicated that
the percent removal requirements for bleach plant scrubbers would
be difficult to achieve if the mass of chlorine and chlorinated
HAP ' s sent to the scrubber were reduced. In response to these
comments, EPA included chlorine outlet concentration and outlet
mass emission limit compliance options.
Regarding pollution prevention efforts, the final rule will
contain provisions for complying with the kraft pulping standards
using a strategy that focuses on removing HAP ' s from in-process
recycled or reused condensate streams before they are allowed to
15-2
-------
be volatilized into the atmosphere. Since the final rule does
not identify the specific control technology to be used, this
compliance alternative (the CCA), provides industry with the
opportunity to implement pollution prevention projects that can
achieve the HAP reductions equivalent to the MACT standards.
Additionally, the effluent guidelines contain voluntary
performance-based incentive programs designed to compliment the
baseline BAT to encourage individual mills to evaluate and
install technologies that could achieve further pollutant
reductions.
Regarding the comprehensiveness of the regulations, NESHAP
standards are limited to addressing the compounds contained in
the HAP list in section 112 (b) of the Act and emitted from the
all significant sources at pulp and paper mills. Although there
are some areas of the mill that are not specifically covered by
the pulp and paper rule, EPA maintains that the rule addresses
the pollutants, and pollutant sources deemed most critical at
pulp and paper mills. The effects on workers and product markets
caused by the MACT standards are evaluated in the EA (A-92-40,
V-A-2). Chemical accident potentials are addressed under
section 112(r) of the Act for applicable facilities. Residual
risks of this rule will be addressed under section 112 (f) of the
Act. Hazardous waste generation is addressed and regulated under
RCRA. The EPA maintains that all cross-media impacts will be
considered, if not specifically under this rule, under other
rules that are already in effect.
Comment: One commenter (20,122) indicated that TCF and
secondarily chlorine free (SCF recycled paper products that have
not been secondarily bleached with chlorine and chlorine
compounds) technologies should be evaluated from both air and
15-3
-------
water perspectives. The commenter (20,122) stated that it was
inappropriate for EPA not to consider the applicability of
TCF/SCF technologies under the Act simply because they were
eliminated from consideration under the CWA. The commenter
(20,122) further asserted that EPA must develop a way to phase in
TCF/SCF technologies within the rule and other authorities.
Another commenter (20,102) encouraged incentives for producing
paper using TCF and other environmentally friendly technologies.
Response: The EPA has included incentives for facilities to
use TCF processes in the final rule. For this NESHAP, all kraft
mills have been given a total of 8 years to comply with the
standard. This additional 5 years was given to allow facilities
to install process equipment, such as oxygen delignification and
TCF bleaching. The OW has also included several incentive
packages in the effluent guidelines. These incentives would
provide mills with additional compliance time, up to 16 years
beyond the date of promulgation, to meet limitations more
stringent than BAT. Qualifying technologies more stringent than
BAT include oxygen delignification and TCF bleaching.
Additionally, EPA considers that the TCF technologies would
constitute compliance with the bleaching component of the MACT
requirements. Therefore, in the air portion of the combined
rule, EPA has indicated that application of TCF technologies for
bleaching would comply with the bleaching standards.
Comment: One commenter (20,145) opposed the expansion of
the combined rule to a multimedia permit concept. The commenter
(20,145) indicated that a single multimedia permit at a facility
would prevent new projects and delay major expansions. The
commenter (20,145) opposed multimedia permits because they would
limit the flexibility of industry to choose options for reducing
15-4
-------
pollutants and because changes that affect one permitted media
would open the permit for review on all permitted media.
Response: Compliance with the combined rule will not
require a multimedia permit. The air and water regulations were
developed jointly because of the multimedia nature of pollution
control in this industry. The air and water regulations are
being promulgated simultaneously to facilitate coordinated
compliance planning. However, the regulations are being
promulgated individually under the respective authorities of the
Act and the CWA. Accordingly, each regulation will be
implemented under the authority of its respective Act.
Permitting requirements, therefore, will be unchanged. This
NESHAP will be implemented according the requirements of the
part 63 general provisions and each mill's title V operating
permit. New source review permits will be required for any new
or modified sources. Water regulations will not be addressed in
air permits, nor will air regulations be addressed in the permits
required under the CWA.
15.2 EFFLUENT GUIDELINES
Comment: Two commenters (20,018, 20,027) questioned whether
EPA had effectively evaluated the integration of air and water
standards for the bleaching component. One commenter (20,057A2)
stated that EPA has not evaluated the impact of the proposed air
standards on the effluent guidelines.
One commenter (20,018) indicated that there appeared to be
inconsistencies between the proposed MACT standards for the
bleaching component and the technology requirements applied to
pulp bleaching in the effluent guidelines due to lack of
coordination between air and wastewater groups. The commenter
(20,018) stated that the proposed effluent guidelines require
15-5
-------
complete substitution of chlorine with C102 for kraft bleaching.
The commenter (20,018) contended that MACT for the bleaching
component should be complete C102 substitution since emissions of
chlorine and chloroform from a C102 bleaching stage are
significantly lower than a chlorine bleaching stage.
One commenter (20,027) stressed that the impact of the
effluent guidelines proposed process changes on elemental
chlorine emissions should have been considered when determining
the air control options for the bleach plant. The commenter
(20,027) stated that substitution of C1C>2 for elemental chlorine
plus oxygen delignification would greatly reduce the
concentration of elemental chlorine at the inlet to bleach plant
scrubbers. The commenter (20,027) pointed out that this would
make it difficult to demonstrate the high removal efficiency
required by MACT and significantly increase the probability of a
calculated exceedance when, in reality, the actual emissions of
chlorine were very small.
Response: The EPA has conducted several impact analyses on
the integration of the air and water standards and maintains that
a sufficient evaluation of the integration of these two standards
has been completed. The EPA has analyzed air emissions after
implementation of the effluent limitation guidelines and
standards options (referred to as OW Options A and B) , and TCF.
Results of these analyses are presented in chapter 20 of this
document.
The December 17, 1993 proposed rule specified that HAP
emissions from chlorine or chlorinated compound application
stages must be reduced by 99 percent. The EPA agrees with the
commenters that C1C>2 substitution, required by OW and the MACT
floor, will decrease the amount of chlorine in bleach plant
15-6
-------
scrubber inlets and that a 99 percent reduction of chlorinated
HAP' s from a C102 application stage may not be feasible. As
discussed in the March 8, 1996 supplemental notice, EPA
incorporated a scrubber outlet chlorine concentration and
considered a mass emission limit as options for the bleaching
system requirement. The outlet concentration, mass emission
limit, and the percent mass reduction options are considered by
EPA to be equivalent (A-92-40, II-I-24, IV-B-29).
Comment: One commenter (20,027) stated that the waste heat
that will accompany steam stripping will have adverse water
pollution consequences. The commenter (20,027) noted that many
mills (particularly in southern States) have heat-limited
effluent treatment systems. The commenter (20,027) declared that
requiring increased steam stripping at such mills would probably
lead either to noncompliance with NPDES limits or to the need to
construct cooling towers.
Response: The EPA reviewed the data submitted by industry
(A-92-40, IV-D1-46) detailing the number of cooling towers
existing in the pulp and paper industry. (The data indicated
that 13 cooling towers were being used.) No additional data were
submitted to EPA regarding potential conflicts with steam
stripping and the NPDES permit program or the prevalence of
cooling towers used in conjunction with steam stripping systems.
While some mills may need cooling towers to handle the waste heat
load from the pulping and bleaching processes, EPA's judgment is
that it is not appropriate to assign the costs for installing and
operating cooling towers to all mills for estimating national
impacts.
15-7
-------
15.3 COMBUSTION MACT
Comment: Several commenters (20,027, 20,018, 20,043,
20,054A2, 20,056, 20,057A2, 20,146) argued that all processes in
the mill are interrelated and that EPA failed to consider this
when it failed to propose combustion MACT standards with the MACT
standards for other sources. Two commenters (20,027, 20,057A2)
claimed that EPA proposed an integrated rule that requires
changes in the technology and engineering of process-source
emissions without considering the impact of those changes on the
design, capacity, and engineering of the liquor recovery process.
Several commenters (20,011, 20,014, 20,027, 20,043, 20,046)
stated that the lack of integration between the process sources
and combustion source rulemakings has several technical,
engineering, emissions, and economic implications that were not
considered by EPA. One commenter (20,014) stated that it is
difficult to evaluate the combined regulations because the air
emission regulations for other sources at the mill
(i.e., recovery furnaces, lime kilns, smelt dissolving tank
vents, oxidizers, and power boilers) are not known. Another
commenter (20,046, 20,046A2) contended that the costs associated
with the combustion sources must be considered as part of the
ultimate cost/benefit analysis of the combined rule.
Several commenters (20,027, 20,043, 20,056, 20,114) urged
EPA to integrate the combustion and process MACT standards and
re-propose the NESHAP. One commenter (20,059) indicated that the
deferral of proposals for combustion and certain non-combustion
sources frustrates the objective of a coordinated pollution
prevention approach. Several commenters (20,011, 20,027, 20,043,
20,059, 20,066A3) indicated that EPA must consider the impacts
the proposed non-combustion standards and combustion standards
15-8
-------
would have on each other because the two types of sources are so
interrelated. Three commenters (20,027, 20,043, 20,066A3) cited
examples of pulping emission units at sulfite mills that are
currently controlled by combustion source control equipment and
combustion emission sources (white liquor production) that are
affected by water reuse patterns. One commenter (20,057)
indicated that the combustion and non-combustion standards must
be consistent with each other.
Three commenters (20,046A2, 20,056, 20,074) indicated that
the costs of the pending combustion source MACT regulations are
likely to be very high, and in order to assess the total costs of
the regulations on mills, EPA should wait until the combustion
source requirements are clearly understood and then integrate
them into the cost-effectiveness assessments.
Response: In the preamble to the December 17, 1993
proposal, EPA indicated that the combustion source MACT standards
were expected to be proposed in 1994 and be promulgated together
with the standards for the non-combustion source emission points
and effluent guidelines. After further evaluation and analyses,
EPA proposed the combustion source standards as the non-
combustion MACT standards and the effluent guidelines were
promulgated.
The EPA contends that the Agency has considered the
interrelated nature of pulp and paper mills and impacts of the
combined rule (i.e., combustion sources, non-combustion sources,
and effluent guidelines). The non-combustion source standards
address HAP emissions associated with pulping and bleaching
processes. The only potential conflict between the combustion
and non-combustion source standards is the use of recovery
furnaces as emissions control devices. While combustion sources
15-9
-------
(e.g., chemical recovery operations) are not covered in the scope
of the non-combustion source standards, the recovery furnace has
been identified as a control device for pulping emissions used at
a limited number of existing facilities. Comments received
following proposal have stressed that industry groups strongly
recommend that recovery furnaces not be used for controlling
pulping emissions due to serious explosion risks. Although EPA
agrees with the industry's concerns regarding explosion hazards,
the final rule contains a control option for routing pulping
emissions to a recovery furnace, power boiler, or lime kiln to
provide individual mills flexibility in complying with the non-
combustion source standards.
The effluent limitation guidelines and standards contain
requirements that have the potential to affect combustion
sources. Most notably, the effluent limitation guidelines and
standards for handling black liquor spills which will likely
result in increased solids loading to chemical recovery
processes. The EPA's OW has taken these interactions into
account in their costs and impacts analyses.
With regard to sulfite mills, EPA has established a separate
subcategory for these pulping processes and re-evaluated the
floor level of control. This analysis was discussed in the
March 8, 1996 supplemental notice. Consequently, EPA determined
that sulfite pulping emissions are typically controlled using the
acid making/chemical recovery systems at these mills. The acid
making/chemical recovery systems at sulfite mills should not be
affected by the combustion source standards.
Comment: One commenter (20,053A1) suggested that emission
standards for black liquor oxidation (BLOX) systems should be
included in the proposed pulping emission standards. The
15-10
-------
commenter (20,053A1) asserted that control of HAP emissions from
BLOX systems would be substantially more cost effective, and
would result in significantly greater environmental benefits than
treatment of insignificant HAP sources such as brownstock washer
vents and oxygen delignification vents.
Response: The EPA disagrees with the commenter regarding
the insignificance of brownstock washer vents and oxygen
delignification vents. As presented in the revised emission
factor document (A-92-40, IV-A-8), HAP emissions from these
sources are not insignificant. With regard to the appropriate
placement in the rule for BLOX systems, EPA contends that these
systems should be considered part of the chemical recovery
process (i.e., combustion sources). The purpose of BLOX systems
is to convert sodium sulfide into thiosulfate. This conversion
is done to prevent the stripping of hydrogen sulfide gas in the
chemical recovery process and is therefore best considered under
the combustion source MACT standards.
15-11
-------
-------
16.0 INTERACTION WITH OTHER RULES
16.1 NEW SOURCE REVIEW/PREVENTION OF SIGNIFICANT DETERIORATION
Comment: Industry and some States have commented
extensively on the potential problems that could result from the
interaction between the December 17, 1993 proposed rule and the
NSR program. The NSR program includes the PSD and nonattainment
NSR preconstruction permit programs.
Regarding the December 17, 1993 proposed rule, many
commenters (20,027, 20,057, 20,057A2, 20,071, 20,103, 20,111,
20,118) stated that: (1) the control equipment and process
changes required to comply with the rule will increase emissions
of S02 and NOX; (2) these compounds are generated from the
combustion of vent gases required by the rule; and (3) the
increases in S02 emissions could be of such magnitude to trigger
the need for preconstruction permits under the PSD/NSR program.
Several commenters (20,027, 20,043, 20,053A1, 20,054A2,
20,057, 20,057A2, 20,146) maintained that there are issues and
impacts of PSD/NSR review that were overlooked by EPA.
Commenters indicated that NSR review would: (1) cost the pulp
and paper industry significantly more for permitting and
implementation of NSR and PSD requirements than predicted by EPA
(2) impose a large permitting review burden on State air quality
offices; and (3) present difficulties for mills to meet the
16-1
-------
proposed NESHAP compliance schedule of 3 years due to the time
required to obtain a preconstruction permit.
One commenter (20,071) indicated that, for some sources, it
would be difficult or impossible to obtain permits due to
emission caps and difficulties obtaining offsets. The commenter
(20,071) indicated that sources would be required by one set of
regulations to install emission controls and constrained from
beginning construction on them by another set of regulations.
One commenter (20,027) noted that the PSD/NSR review could
preclude existing combustion devices from controlling vent gases
(i.e., stand-alone thermal oxidizers would have to be used). One
commenter (20,043) also noted that a steam-limited facility near
a Federal Class I PSD area (61 FR 38250, July 23, 1996) area may
not be able to perform the required steam stripping because an
increase in criteria air pollutant emissions, resulting from
increased steam production, may be prohibited, or limited, by the
PSD air quality restraints.
Many commenters (20,010, 20,011, 20,027, 20,011, 20,057A2,
20,010, 20,111, 20,118) made recommendations on how EPA should
handle the issue of PSD/NSR in the final rule. Several
commenters (20,010, 20,011, 20,057A2, 20,111, 20,118) stated
their views on whether or not the pollution controls required by
MACT should be excluded from PSD/NSR review. Three commenters
(20,027, 20,011, 20,057A2) proposed that controls installed to
comply with MACT standards be granted an explicit "pollution
control" exclusion from PSD review and NSR. One commenter
(20,057A2) recommended that EPA include language in the PSD and
NSPS regulations to exempt sources that install controls as a
result of MACT standards, rather than in each MACT standard.
16-2
-------
One commenter (20,010) indicated that the PSD regulations
should not be bypassed for situations where the installation of
new incineration equipment results in increases in criteria
pollutants. One commenter (20,111) indicated that the
installation of MACT controls should trigger PSD or NSR
requirements if there is an associated increase in pulp
production or in the permitted emission levels of the existing
boiler.
Response: An industry-wide NSR exemption for pulp and paper
mills is not necessary because EPA already has an existing policy
for excluding from NSR pollution control projects (PCP) at
existing sources (July 1, 1994 memorandum from John Seitz,
"Pollution Control Projects and NSR" and proposed revisions to
40 CFR parts 51 and 52 at 61 FR 38250) . The PCP exclusion is
granted on a case-by-case basis by the permitting authority.
Under this policy, projects that are eligible include physical or
operational changes whose primary function is the reduction of
air pollutants subject to regulation under the Act (e.g., MACT
standards). To obtain the exclusion, a mill must submit a
request for exclusion to the permitting authority through either
a minor NSR permitting process, a State non-applicability
process, or other similar process. Modifications to existing
combustion devices (e.g., boilers) or the addition of a stand-
alone thermal oxidizer to comply with a MACT standard are the
types of technologies that would qualify for a PCP exclusion.
To grant a PCP exclusion, the permitting authority must
determine that the project is "environmentally beneficial" and
would cause no adverse air quality impacts. An adverse air
quality impact is defined as causing or contributing to the
violation of a national ambient air quality standard (NAAQS), a
16-3
-------
PSD increment, or an air quality related value (AQRV) in a
Federal Class I PSD area (e.g., national parks). The AQRV's are
specified by the responsible Federal Land Manager (FLM).
In the March 8, 1996 supplemental notice, EPA presented a
strategy for streamlining the process of granting a PCP
exemption. Based on an evaluation of pollutant reductions and
environmental and energy impacts, the notice proposed a policy
statement that projects implemented to comply with the MACT
portion of the December 17, 1993 proposed rule were to be
considered "environmentally beneficial" under the Agency's PCP
policy. This determination would mitigate one of the two case-
by-case determinations required by the permitting authorities. A
case-by-case determination that PCP would pose no adverse air
quality impacts would still be required in order for the
exemption to be granted.
The March 8, 1996 supplemental notice requested comment on
the determination that these MACT control projects are
environmentally beneficial and eligible for the PCP exemption.
The EPA also solicited comments on providing a specific exclusion
in the NSR rule for these types of controls installed to comply
with MACT.
Issues related to the time required to obtain a PSD/NSR
review and the impact on the compliance schedule are addressed in
the response to the following comment summary.
Comment: With regard to the March 8, 1996 supplemental
notice, several commenters (IV-D2-2, IV-D2-3, IV-D2-7, IV-D2-10,
IV-D2-11) stated that the guidance by EPA regarding the existing
PCP exclusion was inadequate, and recommended including specific
language in the pulp and paper MACT rule exempting MACT
compliance projects from PSD/NSR review. One commenter
16-4
-------
(IV-D2-15) supported EPA's determination that MACT compliance
projects will be environmentally beneficial and should qualify
for exemption under EPA's PCP exclusion guidance. However, the
commenter (IV-D2-15) contended that EPA should recognize that
when mills install the controls required for MACT, they will
likely upgrade other parts of the operation at the same time,
including increases in capacity. The commenter (IV-D2-15)
requested that the exclusion be broad enough to include all the
actions taken concurrently with the MACT installation. The
commenter (IV-D2-15) requested a firm commitment from EPA that
MACT compliance projects will be expressly excluded from coverage
in the new NSR reform regulations. Two commenters (IV-D2-11,
IV-D2-15) urged that EPA include language in the pulp and paper
MACT rule that will expressly exclude any project installed for
the purpose of complying with MACT from NSR or PSD review without
any need for a site-by-site air quality benefit analysis.
Another commenter (25,538) objected that both the PCP
exemption and proposed NSR reform rule provide inadequate relief.
Both policies still give the FLM an opportunity to conduct an
AQRV review. The commenter (25,538) also objected that these
policies apply only to "modified" sources and not to new emission
units (e.g., new boilers) that may be constructed to comply with
the rule. The commenter (25,538) indicated that, because the
granting of NSR relief is voluntary by the State, it is doubtful
that the States will confer the NSR relief that EPA has proposed.
As a result, the commenter (25,538) asserts that additional
controls on these collateral emissions will be required, and EPA
has not taken into account the cost of these additional controls.
Two commenters (20,057A2, 20,118) on the December 17, 1993
proposed rule had noted that EPA failed to consider the
16-5
-------
additional burden that PSD/NSR review would have on State
permitting agencies.
One commenter (IV-D2-4) strongly opposed the specific
exclusion of these types of projects in the NSR rules. The
commenter (IV-D2-4) noted that these projects are not necessarily
environmentally beneficial and should not be eligible for
automatic exemption from major source NSR. The commenter
(IV-D2-4) also expressed concern that EPA's statement that MACT
compliance projects are "environmentally beneficial" and would
limit the States' authority to apply the environmental safeguards
available in the July 1, 1994 policy. These safeguards are
required to ensure that progress made by permitting authorities
to reduce air pollution is not compromised. One commenter
(20,103) questioned what the appropriate response would be where
an increase in S02 results in potential violation of S02 NAAQS.
Response: The PCP exemption offered by the current policy
(July 1, 1994 memorandum from John Seitz, "Pollution Control
Projects and NSR" and proposed revisions to 40 CFR parts 51 and
52 at 61 FR 38250) and in the proposed NSR reform (61 FR 38250)
provides adequate relief from any cost or schedule impacts of NSR
that are unreasonable. In the December 17, 1993 proposed rule,
EPA has concluded that projects implemented to comply with the
rule are "environmentally beneficial," under the context of the
NSR program, based on the overall environmental impacts
associated with this rule. This conclusion, along with the
proposed NSR reform rule, should reduce some of the uncertainty
with the policy and help provide uniformity in its application.
The Agency does not believe, however, that an automatic exemption
from NSR is appropriate or necessary. Case-by-case review and
approval by the permitting authority is a necessary and
16-6
-------
appropriate step to ensure that the environmental safeguards are
met and that the approval is subject to public notice. The
environmental safeguards are protection of the NAAQS, PSD
increments, and AQRV's in Federal Class I areas; and the securing
of offsetting emission reductions if the project results in a
significant increase of a nonattainment pollutant. Designation
of MACT projects as "environmentally beneficial" does not limit
the States' authority to apply these environmental safeguards, as
provided in the July 1994 policy.
The case-by-case nature of the PCP exemption should not
impede the granting of exemptions. The objections to the
proposed exclusion that were raised by the State and local air
pollution control agencies pertained primarily to EPA's request
for comment on the proposed option of granting in the rule an
automatic exemption from major NSR. An exemption in the rule
could allow significant emission increases even in cases with
local air quality problems. Their position was that the
environmental safeguards in the current policy should be
retained. The EPA agrees that an automatic exemption from major
NSR is not appropriate. In addition, the procedure of a PCP
application and review by the permitting authority is necessary
to ensure that the PCP exemption is not applied to projects
performed concurrently with NESHAP compliance that would result
in an increase in process utilization or emissions. The types of
projects suggested by one commenter as being candidates for PCP
exemption include such activities as concurrent process upgrades
to increase production capacity. The EPA believes that these are
the types of projects that the Act presumes should be subject to
NSR preconstruction review. Review by the permitting authority
16-7
-------
through the minor NSR process, therefore, is a necessary
safeguard to monitor appropriate application of the policy.
Obtaining a minor NSR permit should not impose any
significant delays that would adversely affect the ability of
pulp and paper mills to comply with the NESHAP on time. Possible
triggers for NSR/PSD review would be SO2 and NOX emission
increases associated with the control of pulping systems. First,
most of the mills that will experience significant SC>2 and NOX
emission increases will be kraft mills, and the final rule allows
8 years for compliance with the kraft pulping HVLC system
requirements. In addition, a mill can request a compliance date
extension of up to 1 year if needed for the installation of
controls. Delays in the ability to install controls that are
caused by the permitting process could qualify for this
additional extension.
In addition, the proposed NSR reforms contain measures to
reduce the delays that sometimes are associated with permitting
near Federal Class I PSD areas. The proposed NSR reforms better
define the role of the FLM and the procedures to follow for an
AQRV analysis. The proposed NSR reforms require that the FLM
provide to the applicant, in advance, a current list of relevant
AQRV, sensitive receptors, critical pollutant loadings for each
AQRV, and the methods available to analyze potential impacts.
The rule also will define the role of the FLM and set a timetable
for FLM involvement in the permitting process. These reforms
should streamline the process by reducing much of the uncertainty
inherent in the current process.
The claim that the PCP exemption is a "voluntary" action by
the permitting authority is somewhat misleading. The current
policy and the proposed NSR rules contain specific criteria for
-------
issuing an exemption. Emissions cannot cause or contribute to a
violation of a NAAQS, and PSD increment, or an AQRV. These
criteria involve local air quality impacts that must be
considered on a case-by-case basis. Also, review by the local
authority is needed to ensure that the proposed project is a MACT
compliance activity. Projects that meet these criteria qualify
for an exemption from major NSR, and there is no reason to
believe that the permitting authority will not issue the
exemption if all the legal criteria are met.
In conclusion, EPA maintains that for the majority of pulp
and paper mills, compliance with the NESHAP will not trigger
major NSR because most mills will qualify for, and obtain, a PCP
exemption from NSR. Since NSR is not expected to occur in a
widespread or frequent manner, it is not appropriate to account
for additional costs of NSR in the national impacts of the rule.
Likewise, a significant burden increase on State permitting
agencies is not expected. The 8-year compliance period for
designated HVLC system operations at kraft mills with potential
NSR problems allows the time to explore alternative pollution
prevention programs that have less secondary impacts, like those
anticipated with the clean condensate alternative.
Comment: One commenter (20,057A2) suggested that EPA
evaluate the impact of the proposed rule on the northeast
Transport Region ozone non-attainment areas (NAA). The commenter
(20,057A2) requested that EPA provide guidance on meeting the NOX
reasonably achievable control technology (RACT) standards and for
new source compliance with LAER.
Response: The EPA has concluded that this rule should not
have an adverse impact on ozone attainment in the northeast Ozone
Transport Region (OTR) because the decreases in VOC emissions are
16-9
-------
very large compared to the potential increases in NOX emissions
(about 75 to 1) . The rule will decrease VOC emissions by
409,000 Mg/yr and may increase NOX emissions by 5,230 Mg/yr,
nationally. The EPA recognizes that some of those increases will
occur in the northeast OTR. These increases in NOX are very
small in comparison with current national and regional NOX
emissions and with current NOX emissions from pulp and paper
plants subject to the final rule. Increases in NOX emissions
from compliance with the MACT standard are estimated at about
5,230 Mg/yr (5,753 tons/yr) . National NOX emissions in 1994 were
approximately 21.4 million Mg/yr (23.6 million tons/yr).
Increases in NOX emissions resulting from compliance with
the NESHAP are primarily due to increased steam demand for steam
stripping the pulping process condensate streams. Combustion of
pulping vent streams accounts for a minority of the estimated
increases in NOX emissions.
Facilities installing boilers or increasing boiler capacity
to meet the increased steam demand may have to meet NOX, RACT
standards or LAER. For these facilities, EPA has provided
guidance on meeting NOX standards in a documented entitled
Alternative Control Techniques Document -- NOX Emissions from
Industrial/Commercial/'Institutional (ICI) Boilers,
EPA-453/R-94-022, published in March 1994. This document
outlines several options for reducing NOX from industrial
boilers.
For combustion devices used to combust pulping vent streams,
EPA does not believe that guidance on NOX RACT and LAER is
necessary. The EPA has concluded that approximately 70 percent
of all facilities will comply with the pulping standard by
routing vents to existing combustion devices and 30 percent will
16-10
-------
construct incinerators to control the vents (A-92-40, IV-E-93) .
Analysis of existing combustion sources shows that a 5 percent
increase in fuel use is required to incorporate the vent streams
and keep the combustion device at a consistent level of operation
(A-92-40, II-B-31). Such a small increase in fuel requirements
should result in minimal NOX increases at these mills. The
facilities using existing combustion devices should not trigger
any additional NOX RACT requirements beyond those already in
place for these devices.
For the 30 percent of facilities EPA estimates will install
new incinerators, NOX increases are not expected to trigger LAER,
because the anticipated emissions increases are below the
emission thresholds for nonattainment NSR. For incinerators, the
rule requires 98 percent HAP reduction. The rule requires
98 percent reduction of HAP's or an operating temperature of
1600 °F at a residence time of 0.75 seconds. Sources are not
expected to operate incinerators at temperatures significantly
higher than 1600 °F due to added fuel costs. Analysis of NOX
formation mechanisms show that below 1800 °p, negligible levels
of NOX are generated. Therefore, the standard tends to minimize
the additional formation of NOX. In the event that a facility
that does become subject to NOX RACT, available technologies
include lcw-NOx burners, selective catalytic reduction (SCR), or
selective non-catalytic reduction (SNCR) technologies.
Comment: One commenter (20,027) maintained that the
exclusion from NSR and PSD review for the installation of
pollution control projects (including process modifications such
asClO2 substitution) is legally required based on the definition
of "modification" in NSPS regulations.
16-11
-------
Response: The comment is incorrect. A pollution control
project modification is not legally required to be exempted from
NSR and PSD review based on the NSPS definition of "modification"
in 40 CFR part 60. For purposes of NSPS, the addition or use of
any system or device whose primary function is the reduction of
air pollutants is not considered to be a "modification." This
definition, however, has no application to the NSR/PSD rules. A
separate and distinct definition of "modification" is specified
for NSR/PSD implementation in 40 CFR parts 51 and 52.
16.2 RESOURCE CONSERVATION AND RECOVERY ACT/BOILERS AND
INDUSTRIAL FURNACES
Comment: Regarding the December 17, 1993 proposal, two
commenters (20,011, 20,027) agreed with the proposed requirement
of combusting steam stripper overheads in the process wastewater
area. According to three commenters (20,027, 20,057, 20,057A2)
the overhead stream should be condensed to enhance the fuel value
by concentrating the methanol. Three commenters (20,027, 20,057,
20,146) pointed out that increasing the concentration of methanol
would increase the cost-effectiveness of this control option;
however, burning a waste-derived fuel would likely trigger the
Resource Conservation and Recovery Act/Boilers and Industrial
Furnaces (RCRA/BIF) rules. Several commenters (20,011, 20,027,
20,054A2, 20,057, 20,057A2, 20,071) indicated that EPA failed to
address the potential RCRA/BIF implications of the combined rule.
Three of these commenters (20,027, 20,057A2, 20,071) urged EPA to
exempt the burning of methanol condensates (from steam-stripping
devices installed to meet the HAP reduction requirements) from
the RCRA/BIF rules under a "clean fuels exemption," as long as
they are combusted on site and only exhibit the characteristic of
ignitability.
16-12
-------
One commenter (20,078) indicated that the need for both a
title V air permit and a RCRA permit (when methanol is
concentrated and burned) was redundant and suggested issuing only
air permits. The commenter (20,078) stated that a similar
process is currently in place for water quality/RCRA issues using
a NPDES permit.
In the March 8, 1996 supplemental notice, EPA proposed to
exclude from RCRA/BIF requirements the combustion of steam
stripper system condensates. Three commenters (IV-D2-7,
IV-D2-15, IV-D2-19) supported EPA's decision that stripper vent
gases that were condensed and combusted on site to meet MACT
requirements should not be subject to the RCRA/BIF requirements.
One commenter (IV-D2-15) expressed concern that EPA limited the
scope of its decision to only those stripper overheads that have
been concentrated before being combusted. The commenter
(IV-D2-15) stated that some mills may be able to meet the MACT
requirements without rectifying their vent gases, but because
those gas streams may condense naturally, they would fall under
RCRA/BIF. The commenter (IV-D2-15) stated that although these
mills may not be utilizing their methanol streams to their
greatest energy potential, the environmental risk posed by
burning the unconcentrated materials would be no greater than
that for the rectified materials. The commenter (IV-D2-15)
recommended that EPA modify its proposal in the final rule to
allow mills that simply condense stripper vent gases and burn
these condensates to be excluded from regulations under RCRA/BIF.
One commenter (IV-D2-3) recommended that the methanol
rectification system (steam stripper, rectifier, separate tanks,
and delivery system) be exempted as a whole from RCRA/BIF. Such
an explicit exclusion will ensure that a facility may efficiently
16-13
-------
re-use methanol fuel without component pieces of the handling
system being re-regulated by BIF.
One commenter (IV-D2-14) noted that the proposed standards
for pulp mills may trigger RCRA/BIF regulations, and specifically
stressed concern over RCRA/BIF applicability at the point of
generation of the pulping process condensate waste stream. To
resolve the conflict, the commenter (IV-D2-14) suggested that the
collection and subsequent stripping or alternative treatment of
pulping process condensate should be exempt from RCRA/BIF and
should only be regulated by MACT standards. The commenter
(IV-D2-14) also requested that EPA include red and foul oil in
the MACT exemption from RCRA/BIF for stripper overhead products.
The commenter (IV-D2-14) also requested that EPA explicitly
define turpentine, red oil, or foul oil burning at kraft mills as
activities currently regulated by the Act and exempted from RCRA.
Response: As explained in the March 8, 1996 supplemental
notice, EPA has concluded that regulation of the combustion of
condensates, whether rectified or dilute, is not needed under
RCRA/BIF because the MACT controls will be adequately protective
(and certainly sufficiently protective to eliminate the need for
RCRA controls until the residual risk determination under
section 112 (f) of the Act is conducted) . The condensate does not
contain chlorinated HAP's, and any organic HAP's in the
condensate would be controlled to the level specified by the MACT
standards. In addition, EPA maintains that the burning of this
condensate does not produce any additional HAP's due to the high
temperature and residence times found in the combustion devices
that would be used to comply with the kraft pulping standards.
Therefore, burning condensate will not increase the potential
environmental risk over the burning of the steam stripper vent
16-14
-------
gases prior to condensation. Additionally, the use of the
condensate as a fuel could reduce or eliminate the need for
supplemental firing of fossil fuels in such combustion devices.
The potential cost savings produced by allowing the burning of
condensed steam stripper vent gases would be significant.
Industry estimates that annual cost savings would be
approximately $850,000 per mill, or $100 million for the entire
kraft industry. Cost savings would come primarily through the
reduction in fossil fuel purchases.
In summary, regulation under RCRA is not necessary since the
practice in question would not increase environmental risk,
reduces secondary impacts, and provides a cost savings. Further
considerations of risk should appropriately be handled as part of
the section 112(f) residual risk determination required for all
sources after implementation of MACT standards. For these
reasons, EPA will exclude from the BIF requirements of RCRA
combustion sources that burn condensates derived from steam
stripper overheads.
This decision is consistent with RCRA section 1006, which
requires EPA to "integrate all provisions of [RCRA] for purposes
of administration and enforcement and . . . avoid duplication, to
the extent practicable, with the appropriate provisions of the
Clean Air Act . . .". The EPA acknowledges that the imposition
of RCRA regulations in this instance would result in the types of
unnecessary duplication that section 1006 is intended to prevent.
The EPA maintains that steam stripping with rectification
followed by combustion of the concentrated condensates is MACT,
given the energy, economic, and environmental impacts. See
generally 60 FR 32587, 32593 (June 23, 1995), and 59 FR 29570,
29776 (June 9, 1994) where EPA similarly found that RCRA
16-15
-------
regulation of secondary lead smelter emissions was unnecessary,
at least until completion of the residual risk process.
Some mills further process the condensate to extract
turpentine and red and foul oils. The EPA notes that it
considers the residues that are generated as part of this
processing of the condensates to be within the scope of the
exclusion when such residues are burned as fuels for the same
reasons given above. (These residues also may not exhibit the
ignitability characteristic, and so would not be hazardous in any
case.) Also, the Agency notes that the turpentine and red and
foul oils, which can be put to use as raw materials or non-fuel
products, are not subject to RCRA under the existing regulations
either because they are co-products and not secondary materials
(see 40 CFR 261, l(b) (3) and 261, 2(c)), or because they are used
as ingredients or as substitutes for commercial chemical products
(40 CFR 261, 2(e) (1) (i) and (ii)).
16.3 SECTION 112 RULES (112(g), 112(j), 112(r))
Comment: Section X.L. of the preamble to the proposed
regulation discussed regulations under development that could
affect new, modified, or reconstructed sources at pulp and paper
mills. The preamble encouraged commenters concerned with the
interaction between the proposed 112(d) NESHAP, section 112(g),
and 112 (j) rules to submit those concerns as comments to the
proposed 112(g) rule.
One commenter (20,054A2) indicated that they did not have
time to fully evaluate the impact or interaction of the 112(g)
rule because it was proposed during the MACT comment period. The
commenter (20,054A2) requested that EPA provide an additional
comment period for further comments on the relationship between
16-16
-------
the broad definition of source proposed in the MACT standards and
112(g).
Another commenter (20,027) asserted that pulp and paper
sources should be exempt from 112(g) review. The commenter
(20,027) also asserted that EPA's proposed MACT standards should
not be used as the starting point for 112 (g) determinations in
view of the flaws contained in the proposal. The commenter
(20,027) advised that EPA should issue a public statement to that
effect.
Response: These comments address issues that are no longer
relevant to the pulp and paper industry because section 112(g)
will not apply to sources covered by this NESHAP. At the time
that the pulp and paper standards were proposed (December 1993) ,
the section 112 (g) rules had not been proposed. Since then, the
112 (g) rules have been proposed (63 FR 15504, April 1, 1994) and
the public comment period was reopened (61 FR 13125,
March 26, 1996) in a draft final rule. These actions have
addressed issues associated with the relationship between
section 112(g) and the MACT standards. Moreover, on
February 14, 1996, the Agency published an interpretive notice
(60 FR 8333) that deferred the applicability of section 112(g)
until after the final section 112(g) regulations are promulgated.
The section 112 (g) rule was promulgated on December 27, 1996
(61 FR 68384).
The final 112 (g) rule should have no effect on the pulp and
paper processes covered by the section 112(d) MACT standards.
The section 112 (g) program is a transitional measure to protect
the public from HAP's until EPA issues the MACT standards for a
listed source category. As stated in the final 112(g) rule, only
sources that commence construction or reconstruction after
16-17
-------
June 29, 1998, will be effected, unless a State program to
implement the 112 (g) provisions is adopted sooner. Since the
112 (d) MACT standard for pulp and paper mills will be promulgated
in 1997, the sources addressed in this rule will be exempted from
112(g) review.
Comment: Two commenters (20,102, 20,103) suggested that EPA
use its 45-day review period on all part 70 (title V) permits to
evaluate case-by-case MACT determinations for consistency with
any proposed but not yet promulgated MACT standards to provide
consistency between 112(g) and MACT standards.
Response: The 45-day review period for title V permits is
in place to determine the adequacy and completeness of the permit
application, and is not the place for reviewing consistency
between rules. However, a case-by-case MACT determination under
section 112(g) is not required for emission sources that are
regulated by or specifically exempted by a relevant MACT
standard. Modifications at the affected sources outlined in this
rule are not subject to 112 (g) review.
Comment: One commenter (IV-D2-15) stated that 112(g) should
have no applicability to any units that are either covered by
standards promulgated in this rule, or that are the subject of a
"no regulation" decision. The commenter (IV-D2-15) asserted that
MACT III sources would fall in the second category. The
commenter (IV-D2-15) believes a broader definition of source than
that proposed would allow the flexibility to implement equipment
retrofits and rebuilds, restructure production processes and
install new technology to comply with the MACT standard without
triggering section 112(g) review. The commenter (IV-D2-15)
contended that a plant-wide definition of source for implementing
112(g) is both legally required and represents sound policy.
16-18
-------
Response: For the final regulation, EPA is defining the
affected source to which existing MACT requirements apply to
include the total of all HAP emission points in the pulping and
bleaching system (including pulping condensates). The EPA agrees
with the commenters that certain emission points that are
excluded from the definition of affected source in the rule, or
are subject to a determination that MACT for these operations is
no control, should not be required to undergo Act section 112(g)
review. The sources that have been so identified in are wood
yard operations (including wood piles), tall oil recovery
systems, pulping systems at mechanical, secondary fiber, and non-
wood fiber pulping mills, and paper making systems. With regard
to wood yard operations, tall oil recovery systems, and pulping
systems at mechanical, secondary fiber, and non-wood fiber
pulping mills, EPA has determined that these sources do not
generally emit large quantities of HAP' s and is not aware of any
reasonable technologies for controlling HAP's from these sources.
For paper making systems, EPA has not identified any reasonable
control technology, other than the clean condensate alternative,
that can reduce HAP emissions attributable to HAP's present in
the pulp arriving from the pulping and bleaching systems.
Additionally, EPA has determined that the use of paper making
system additives and solvents result in negligible emissions of
HAP's. Therefore, based on the applicability requirements of
section 112(g) (40 CFR 63 part B, 63.40(b)), wood yard
operations; pulping systems at mechanical, secondary fiber, and
non-wood fiber mills; and paper making systems would not be
required to undergo section 112(g) review.
Comment: One commenter (20,027) requested that, because the
provisions of 112(j) may be misconstrued to apply to sulfite
16-19
-------
recovery furnaces independent of the December 17, 1993 proposed
rule, EPA should make a definite statement as to the status of
sulfite recovery furnaces. The commenter (20,027) expressed
concern that recovery furnaces will not be covered by the
proposed non-combustion or combustion MACT standards and
therefore will be subject to 112 (j).
Response: The EPA intends to cover sulfite recovery
furnaces under a separate NESHAP for combustion sources at pulp
and paper mills. The NESHAP covering combustion sources at pulp
and paper mills will be proposed concurrently with the
promulgation of this rule. Since the pulp and paper source
category has been listed for promulgation by November 15, 1997,
the section 112(j) provisions will not apply unless the
combustion NESHAP is not promulgated by May 15, 1999.
Comment: One commenter (20,059) indicated that because the
proposed rule would increase the reliance on the substitution of
chlorine by C1O2, EPA should use its section 112(r) authority to
establish accident prevention standards or monitoring
requirements to minimize the risks of accidental release. The
commenter (20,059) stated that such standards could ensure that
is manufactured on site in small quantities for use in closed-
loop systems.
Response: Accident prevention regulations under 112 (r) were
promulgated in the Federal Register (61 FR 31668) on June 20,
1996. These regulations included a list of 77 compounds for
which accident prevention and response programs are required.
Facilities storing over 1,000 pounds of C102 are subject to the
112(r) requirement of an approved accident prevention and
response plan. Section 112(r) does not give EPA authority to
require facilities to generate C102 on site. However, EPA has
16-20
-------
concluded that facilities using C1C>2 will install on site, C1C>2
generators since on-site generation has proven to be the most
cost-effective method of providing chlorine dioxide.
16.4 NEW SOURCE PERFORMANCE STANDARDS AND CONTROL TECHNOLOGY
GUIDELINES
Comment: One commenter (20,059) recommended that EPA update
the NSPS to include tighter emission limits for all criteria
pollutants and establish numeric, enforceable emission limits for
existing sources of TRS under section 111 (d) (1) . The commenter
(20,059) also requested that EPA develop a Control Technology
Guideline (CTG) for the pulp and paper industry to establish more
stringent presumptive norms for VOC RACT, as well as requiring
States with nonattainment areas to incorporate these new
requirements into their SIP.
One commenter (20,133) contended that EPA failed to update
the 1978 sulfur dioxide emission standards and TRS compound
standards. Three commenters (20,049A3, 20,082, 20,132) stated
that EPA should retain in the final rule the proposed regulations
to control the amount of TRS air emissions.
Response: The pulp and paper NESHAP were developed under
the section 112 of the Act. NESHAP are only applicable to the
compounds contained in the HAP list in section 112 (b) . The EPA
did not promulgate any sulfur dioxide emission standards for pulp
and paper mills. The EPA is not, at this time, revising the TRS
rules previously developed under section 111. However, the
pulping streams controlled under this NESHAP contain most of the
TRS compounds emitted from pulping system, and the HAP control
required by this rule will also significantly reduce TRS
emissions. The EPA estimates a reduction in TRS of 78,500 Mg/yr
as a result of this rule.
16-21
-------
The promulgated rule will achieve significant VOC emission
reductions since the technologies used to control organic HAP' s
that are subject to this rule also control VOC. The EPA
estimates a reduction in VOC emissions of 409,000 Mg/yr as a
result of this rule. All significant sources of VOC and TRS from
the pulping and bleaching systems have been captured by this
rule.
The MACT standard is a uniform, national requirement that
applies to all new and existing pulp mills. The EPA sees no need
for additional regulatory measures for TRS or VOC control for
pulp mills because any additional emission reductions would not
be significant given the reductions obtained under this rule.
Comment: One commenter (20,011) requested that process
wastewaters subject to emission control in the proposed rule
should not be additionally subject to RACT.
Response: The EPA's analysis indicates that all significant
VOC-laden pulping process condensates at pulp and paper mills
will be subject to the MACT standards. The technology used for
meeting MACT is also the best technology for VOC control. Since
all new and existing mills are subject to the rule, no additional
VOC reductions would be achieved by RACT. However, the level of
control for bleaching wastewater streams was no control and some
bleaching wastewater streams may contain significant levels of
voc . Therefore, consideration of RACT may be appropriate for
these streams.
Comment: With regard to EPA's solicitation for comments on
the potential overlap of the kraft NSPS and the proposed NESHAP
standards in the March 8, 1996 supplemental notice, two
commenters (IV-D2-7, IV-D2-15) contended that compliance with
MACT should be considered compliance with the NSPS for those
16-22
-------
sources subject to both rules and any source covered by both
rules should not have any further monitoring, recordkeeping, or
reporting obligations under the NSPS. Another commenter
(IV-D2-10) supported the concept of consolidating the NSPS and
MACT programs, noting that it should not be necessary to report
things twice and that redundant or overburdensome monitoring or
recordkeeping should be eliminated in a "common sense" rule that
would combine these requirements.
One commenter (IV-D2-4) disagreed, noting that allowing a
facility to choose compliance with the NESHAP in lieu of the NSPS
for certain process equipment is inappropriate. The commenter
(IV-D2-4) asserted that the emission units that require TRS
monitors under the NSPS (recovery furnaces, lime kilns, and
brownstock washer, evaporator, and condensate stripper systems
that are not incinerated) are not the same as those regulated
under this NESHAP. The commenter also urged caution as the NSPS
and the NESHAP standards were written to regulate different types
of pollutants.
Response: Sources that are affected by this NESHAP and the
kraft mill NSPS are the pulping system, brownstock washer, and
the steam stripper treatment system. The EPA agrees that
duplication between this NESHAP and the kraft mill NSPS should be
minimized. If an owner or operator complies with the NESHAP
requirements for these sources with one of the combustion control
options, the requirements of the NSPS would also be met. For the
reporting requirements of this NESHAP, documentation of
compliance with the combustion control option used for control of
vents from the pulping system, brownstock washer, and the steam
stripper would also satisfy the NSPS. in this case, only one set
of monitoring, recordkeeping, and reporting requirements would be
16-23
-------
required in the facility's title V permit to satisfy both the
NESHAP and NSPS requirements. However, if emissions from these
sources are controlled by a means other than combustion, such as
the clean condensate alternative, a mill will have to prove
compliance with both the NESHAP and the NSPS. In this case, a
mill would have to report HAP reductions obtained by the non-
combustion control option (to satisfy the NESHAP) and TRS
concentrations at any affected source vent (to satisfy the NSPS).
Comment: One commenter (IV-D2-15) contended that because so
many facilities have installed thermal oxidizers to meet the
standards in the NSPS, and because these controls were considered
in the MACT floor level of control, the MACT rule should include
all the operating parameters associated with NSPS controls, not
just the equipment itself.
Response: The MACT floor level of control is based on the
average of best 12 percent of the population, with respect to HAP
reduction. In the case of this NESHAP, the best-performing mills
controlled pulping system vents by combustion in power boilers,
lime kilns, and recovery furnaces, all of which achieve
98 percent HAP reduction. Therefore, the MACT floor level of
control was 98 percent control of HAP's from pulping system
vents. The operational parameters for thermal oxidizers defined
in the NSPS for control of TRS emissions may not produce the
98-percent HAP emission reduction required by this rule. For
example, the thermal oxidizer operating conditions specified in
the NSPS (1200 0.5 seconds residence time) are not sufficient
OF,
to provide the 98 percent HAP reduction that is achievable in a
boiler, lime kiln, or recovery furnace (A-92-40, IV-B-18).
Analysis of HAP destruction in thermal oxidizers show that an
operation level of 1600 and Off. 75 seconds residence time is
16-24
-------
required to meet the 98 percent HAP reduction requirement.
However, this rule allows three options for a facility to show
compliance when a thermal oxidizer is used to control HAP's in
pulping vent streams. To meet the thermal oxidizer requirements
of this rule, a facility must show that the thermal oxidizer is
operating at 1600 °F and 0.75 seconds residence time, achieving
98 percent HAP reduction, or has an outlet HAP concentration no
greater than 20 ppmv at 10 percent oxygen. Any existing thermal
oxidizer designed to comply with the minimum NSPS requirements
may have to upgrade to meet the requirements of this rule.
16.5 GENERAL PROVISIONS
Comment: One commenter (IV-D2-15) asserted that EPA should
incorporate corrections to the general provisions definitions in
the final rule. The commenter stated that these changes include
changes to definitions of affected source, malfunction, new
source, and reconstruction. The commenter (IV-D2-15) also
recommended eliminating a drafting error in the general
provisions for "part 70 permit" by defining part 70 permit to
mean "a permit issued by a State permitting authority pursuant to
a program approved by EPA under part 70 of this chapter." This
definition would recognize that permits are not issued pursuant
to part 70, but are issued by States that have permit programs
approved pursuant to part 70. The commenter (IV-D2-15)
recommended including a definition of working day in this rule
since the definition was unintentionally left out of the general
provisions to part 63. The commenter (IV-D2-15) recommended that
working day mean "any day on which the Federal government offices
are open for normal business. Saturdays, Sundays, and official
federal holidays are not working days."
16-25
-------
Response: Concurrently with this rulemaking the Agency
conducted a separate rulemaking effort to revise the general
provisions to part 63. The definitions mentioned by the
commenter relate to broad policy issues that affect all sources
subject to NESHAP. There is no basis for revising these
definitions as part of the pulp and paper NESHAP because an
individual NESHAP is not the proper mechanism for changing the
definitions in the general provisions. Any revisions to the
general provisions should be accomplished in the ongoing
litigation. When the revised general provisions are completed,
any changes will be applied to all sources subject to section 112
provisions, including pulp and paper mills.
16.6 PROJECT XL
Comment: The proposed rule did not address Project XL. One
commenter (IV-D2-14) suggested that EPA use the promulgation
process for the rule as an opportunity to build a foundation for
the implementation of a portion of Project XL. The commenter
(IV-D2-14) recommended changes to the proposed rule that would
provide the foundation for the implementation of Project XL for
affected facilities. The revised language would allow mills,
with approved Project XL Final Project Agreements (FPA),
flexibility in meeting the NESHAP. The commenter (IV-D2-14)
reasoned that the additional language would allow for the
implementation of Project XL at affected sources but does not
require EPA to agree to a Final Project Agreement in lieu of
MACT .
Response: The EPA interprets the commenter's concern to be
that the draft FPA for Project XL was just a non-binding, non-
regulatory agreement that provided the XL participant no
protection from being subject to the applicable rules. The EPA
16-26
-------
has reached agreement with the XL participant and reflected this
agreement in the draft FPA issued for public comments on
October 9, 1996. The draft FPA stated that EPA intends to
implement the agreement through a site-specific rulemaking,
permit revisions, or other appropriate legal mechanisms.
16-27
-------
-------
17.0 SCHEDULE ISSUES
17.1 RULEMAKING SCHEDULE
Comment: One commenter (20,071) indicated that EPA should
delay the final MACT standards until all data are submitted and
the impacts for RCRA and PSD/NSR are clarified. One commenter
(20,027) reasoned that EPA should have waited to receive the
industry test data before proposing the MACT standards. The
commenter (20,027) challenged that EPA is legally bound to issue
a responsible notice of proposed rulemaking and cannot plead time
pressures as a reason for issuing an irresponsible rule. [Case
law cited: NRDC v. Thomas, 805 F.2d 410, 433-437 (D.C. Cir.
1986); State of New Jersey v. Costle, 26, F.2d 1038,1042 (D.C.
Cir. 1980).] However, one commenter (20,102) cautioned that
promulgation of the rule should not be delayed for additional
testing and data gathering because it would result in a delay in
reaching the proposed limitations and additional exposure of the
public and the environment to toxic emissions.
Two commenters (20,027, 20,056) stated that EPA should
extend the rulemaking time line so that the costs and impacts of
the combustion MACT standards are considered with the impacts
from the MACT standards for non-combustion sources. One
commenter (20,018) stated that the accelerated rulemaking
schedule did not allow for development and proper analysis of
data to determine MACT regulations.
17-1
-------
Response: The December 17, 1993 proposed rule acknowledged
that more data would be collected for use in revising the
proposed rule. Additional data were collected and announcements
of data availability were published in the February 22, 1995 and
March 8, 1996 Federal Register supplemental notices. The
March 8, 1996 Federal Register supplemental notice also presented
EPA'S analysis of new data, proposed rule changes, and a
solicitation for responses on the revisions to the December 17,
1993 proposed rule. The Agency believes this consideration of
comments and new data is sufficient. There is no need to delay
the rulemaking process for further consideration of data.
In the March 8, 1996 Federal Register supplemental notice,
EPA included proposed responses to concerns raised about the
potential impacts of RCRA and PSD/NSR on compliance with the
December 17, 1993 proposed rule. The supplemental notice also
presented a strategy for streamlining the process of granting a
PCP exemption from PSD/NSR.
17.2 COMPLIANCE SCHEDULE
Comment: In the December 17, 1993 proposed rule, existing
MACT sources were given 3 years from the date of promulgation to
reach compliance. New sources (those constructing after
December 17, 1993) were required to be in compliance at the time
of startup, or upon promulgation of the final pulp and paper
rule, whichever is later.
Numerous commenters (20,001, 20,015, 20,018, 20,054A2,
20,057, 20,057A2, 20,070A1, 20,071, 20,074) on the proposed rule
expressed concern that kraft pulping mills could not meet the
3-year compliance schedule. The commenters (20,001, 20,015,
20,018, 20,054A2, 20,057, 20,057A2, 20,070A1, 20,071, 20,074)
submitted data supporting their position.
17-2
-------
Response: Based on the data received regarding the
December 17, 1993 proposal, EPA proposed (in the March 8, 1996
Federal Register supplemental notice) to extend the MACT
compliance schedule for kraft brownstock washers and oxygen
delignification units by an additional 5 years. As outlined in
the March 8, 1996 Federal Register supplemental notice, many
kraft mills are currently considering the addition of oxygen
delignification to their pulping lines by the year 2000. The
addition of oxygen delignification has been shown to have
significant environmental benefit, reducing the need for
chlorinated chemical application in the bleaching process. A
reduction in chlorinated compound use in the bleaching processes
results in reduced loadings of chlorinated pollutants to the air
and into the bleach plant effluent.
The EPA considers that the addition of oxygen
delignification would likely require redesigned brownstock
washers to improve washing efficiency before the pulp is sent to
the oxygen delignification system. The new brownstock washer
designs are more efficient, less polluting, and easier to
control. However, implementation of the new brownstock washers
and oxygen delignification systems would probably not occur
within a 3-year compliance schedule due to the cost and the need
to design and construct these systems. Given a 3-year compliance
schedule, time constraints would dictate that mills retrofit
their current washers with a vent gas collection system to
achieve compliance. Once such a collection system is installed,
mills would likely postpone or cancel installation of oxygen
delignification systems. The EPA concluded that allowing an
additional 5 years to the 3-year compliance schedule for kraft
mills would allow sufficient time for a complete evaluation of
17-3
-------
all pollution control options and provide an overall greater
benefit in terms of both air and water pollution control.
Comment: With regard to the proposed extension presented in
the March 8, 1996 supplemental notice, several commenters
(IV-D2-11, IV-D2-5, IV-D2-19, IV-D2-10, IV-D2-3, IV-D2-4,
IV-D2-2, IV-D2-15) supported the EPA's decision and rationale for
extending the compliance period for brownstock washing and oxygen
delignification vents for 5 years.
One commenter (IV-D2-8) argued that the 5-year extension for
control of brownstock washer and oxygen delignification vents
should be applied to all kraft mills instead of only those mills
that are installing oxygen delignification systems. The
commenter (IV-D2-8) noted that the decision to install an oxygen
delignification system is very difficult and mill-specific, and
that the goal of improving emissions reduction through improved
brownstock washing systems is equally applicable to mills that
determine that oxygen delignification is not an appropriate
option.
One commenter (IV-D2-8) noted that for mills where
condensates are recycled back to brownstock washers or oxygen
delignification systems, a potential compliance issue exists
because, while the vents have been given a 5-year extension,
these mills will be meeting the wastewater standards for which
compliance is required 3 years after final promulgation. This
conflict will have several negative impacts: all mills would not
be treated equally; inaction would deny relief that EPA clearly
intended to give to the regulated community; and inaction would
arbitrarily penalize many companies who already have adopted a
sound environmental approach to control of polluted streams
(recycle and reuse) in advance of Federal regulations. The
17-4
-------
commenter (IV-D2-8) suggested extending the compliance time for
wastewater streams recycled to brownstock washing and oxygen
delignification systems to coincide with the compliance time for
vents from those units.
Several commenters (IV-D2-11, IV-D2-3, IV-D2-7, IV-D2-15)
requested that the 5-year extension also be applied to weak black
liquor tanks, pre-washer knotter and screening systems, and other
HVLC vent streams because emissions from these sources will be
transported and controlled by the same HVLC collection and
incineration system as the brownstock washers. These commenters
(IV-D2-11, IV-D2-3, IV-D2-7, IV-D2-15) noted that extension of
the compliance period for all HVLC sources also allows for proper
consideration of the full range of emerging innovative control
options.
Response: The Agency reviewed the commenters' concerns and
agreed that vents included in the HVLC system should be allowed a
similar compliance time as the brownstock washing and oxygen
delignification systems. The majority of emissions and vent gas
flow from equipment associated with the HVLC vent streams occur
from the washing system and the oxygen delignification system.
Therefore, the design of the HVLC collection and transport system
would be significantly influenced by these two systems. If
different compliance times were provided for the components of
the HVLC system, an affected source would expend significant
amounts of capital to control systems required to comply in the
3-year time frame. The source would have to redesign the gas
transport and control devices 5 years later to accommodate
controlling the washing system and oxygen delignification system.
This cost could discourage the implementation of low flow washing
and oxygen delignification systems. This would serve as an
17-5
-------
obvious disincentive to installation of advanced wastewater
treatment technology since mills would be understandably
reluctant to replace a newly installed air pollution control
system. Therefore, EPA concluded that additional compliance time
is appropriate and necessary for the remaining equipment
controlled by the HVLC collection and transport system as well as
the brownstock washing system and the oxygen delignification
system (see 61 FR 9394-95, March 8, 1996) . The final rule allows
affected sources to control all the equipment in the HVLC system
at the same time, 8 years after publication of the pulp and paper
rule.
The compliance extension applies to HVLC systems at all
kraft mills. The additional design and mill modification to meet
the standards is a lengthy process. The Agency wanted to allow
sufficient time for each kraft mill to fully consider all
pollution control options. The Agency also recognized that the
pulp and paper industry will be implementing both water and air
rules essentially at the same time. Given the engineering
requirements, capital expenditures, permitting requirements, and
the resources necessary to implement both the water and air
standards, the Agency decided that all kraft mills would be given
a 5-year compliance extension for HVLC systems.
The final rule includes requirements for kraft mills to
submit a non-binding control strategy report along with the
initial notification. The purpose of the control strategy report
is to provide the Agency and the permitting authority with a
means for measuring a mill's progress towards compliance. The
control strategy report contains information such as a
description of the emission controls or process modifications
selected for compliance with the control requirements and
17-6
-------
compliance schedule. The information in the control strategy
report must be revised or updated every 2 years until the mill is
in compliance with the standards of § 63.443.
Compliance with the pulping process condensate standards in
the 3-year time frame should not pose a conflict with the
extended compliance schedule allowed for HVLC systems at kraft
mills. Many of the changes a mill will need to implement to
comply with the pulping process condensate requirements can be
considered before air pollution control systems are implemented,
since the standards do not address recycling patterns, only the
HAP content of the recycled condensates. Additionally, standards
for pollution control from pulping process condensates apply to
streams that are typically not recycled or reused in the pulping
process (namely the HVLC streams from the digester, evaporator,
and turpentine recovery systems) without prior treatment. The
control strategy of recycling uncontrolled process condensates to
controlled equipment is also an option. if a mill cannot meet
the pulping process condensates requirements using this option,
it can choose to treat condensate streams in a stream stripper or
convey condensate to a biological treatment unit.
Comment: One commenter (20,057) stated that EPA has no
basis for concluding that compliance with the December 17, 1993
proposed rule can be achieved within 3 years by all mills.
However, another commenter (20,059) argued that compliance
deadlines should be set as expeditiously as possible and that EPA
should not authorize any categorical 1-year extensions.
Several commenters (20,015, 20,018, 20,057A2, 20,054A2,
20,070A1, 20,071, 20,142, 20,146) indicated that due to the
extensive amount of changes needed and the shortage of available
engineering firms, the time period of 3 years was not sufficient
17-7
-------
for mills to perform the engineering analyses, modifications, and
permitting efforts needed to comply with the proposed rule. One
commenter (20,146) stated that the proposed wastewater
requirements would trigger PSD/NSR permitting requirements that
would make the compliance schedule unrealistic. One commenter
(20,015) suggested an extension of compliance dates along with a
phased compliance schedule allowing the air and water components
compliance dates to be staggered. One commenter (20,001)
suggested EPA extend the compliance deadlines for all existing
sources by 2 to 5 years, along with proposing interim targets for
partial compliance. Another commenter (20,074) strongly
recommended that EPA extend the compliance deadline by 2 years.
One commenter (20,027) recommended that EPA should allow more
time for compliance, particularly in view of the accelerated
schedule for promulgation of MACT for this industry. The
commenter (20,027) declared that EPA could issue a rule that
allowed more than 3 years for compliance and still require MACT
compliance within the original time frame envisioned by the
Agency (November 15, 1997). One commenter (20,071) stated that
it would realistically take up to 6 years for all facilities to
be in compliance. Two commenters (20,027, IV-D2-15) claimed that
because of the far-reaching scope of the proposed rule; its
integration with the future combustion rule; potential PSD/NSR
delays; and limits on available capital, equipment, and
expertise, EPA should grant a 1-year industry-wide compliance
extension to provide a more reasonable time frame in which mills
will be able to achieve compliance with the proposed rule.
One commenter (20,061) suggested a compliance extension of
5 years for mills making an enforceable commitment to TCF
technology. Two commenters (20,102, 20,103) suggested that EPA
17-8
-------
offer a 1-year compliance extension as an incentive to mills that
voluntarily switch to TCF processes. Two commenters (20,102,
20,103) recommended that the extension not be granted to mills
that are required by the effluent guidelines to use TCF processes
(paper-grade sulfite mills). One commenter (20,094) recommended
adjusting the compliance schedule for bleach plant chlorine
compound emission control to be in accordance with any modified
compliance schedules for TCF mills required by the effluent
guidelines.
Response: The EPA considers the 3-year compliance period
ample time for most mills to achieve compliance. For HVLC
systems at kraft mills, EPA has provided an additional 5 years
beyond the 3-year compliance time for a total of 8 years from the
date of promulgation. The EPA believes that this additional time
will be sufficient for kraft mills to completely evaluate all
pollution control options for HVLC systems and to install
pollution controls and pollution prevention processes.
With regard to the combustion MACT for pulp and paper mills,
compliance will be staggered approximately 2 years following this
rule.
If a facility realizes that it may not achieve compliance by
the specified date due to shortages of materials or services
needed to install pollution controls, it may apply for a 1-year
compliance extension. The process for receiving a compliance
extension is outlined in § 63.6 (i) of the general provisions.
The Agency reviews requests for compliance extensions on a case-
by-case basis and an extension may be granted if the Agency deems
the request acceptable.
Regarding the additional compliance period for meeting the
effluent limitation guidelines and standards, EPA's OW has
17-9
-------
included several incentive packages in the final rule for
bleaching systems at paper-grade mills which have elected to
treat wastewater to levels surpassing BAT requirements.
Incentive packages include adding oxygen delignification prior to
bleaching, implementing technologies that result in additional
reduction of process wastewater use, reducing chlorinated
bleaching chemicals use, and various bleaching system
modifications. As an incentive to make this election, EPA is not
requiring participating mills to achieve compliance with the more
stringent portions of the "Advanced Technology" BAT limitations
for 6, 11, and 16 years (for Tiers I, II, and III, respectively)
in order to afford these mills sufficient time to develop,
finance, and install the Advanced Technologies. In light of
this, the Agency is concerned that requiring bleached paper-grade
kraft and soda mills to comply in 3 years with MACT standards
based on process substitution of chlorine dioxide for elemental
chlorine would discourage these mills from electing to
participate in the Advanced Technology program. This is largely
because a mill that implements process substitution before it
installs oxygen or other extended delignification systems is
likely to construct more chlorine dioxide generating capacity
than it ultimately will need. A mill thus compelled to invest
first in process substitution may be very reluctant to abandon a
portion of that investment soon afterwards in order to
participate in the voluntary incentives program.
The EPA also believes that requiring compliance in 3 years
with a chloroform MACT standard based on baseline BAT for
bleached paper-grade kraft and soda mills would present similar
disincentives to achieving greater effluent reductions. A mill
in those circumstances will have made a substantially larger
17-10
-------
capital investment than it will need to control chloroform once
its array of advanced water technologies is installed. Also,
depending on the degree of process modifications the mill makes,
the mill may need a much smaller scrubber for the non-chloroform
chlorinated HAP' s and, in some cases, a scrubber may not be
needed at all to meet the MACT standards for chlorinated HAP
concentration limit. Thus, a mill otherwise interested in
participating in the Voluntary Advanced Technology Incentives
Program will find itself diverting capital to environmental
controls that it ultimately will not need, instead of employing
that capital to make more advanced process modifications that
will benefit both the water and the air.
Under these unusual circumstances where imposition of MACT
requirements could likely result in foregoing substantial cross-
media environmental benefits, EPA believes that a two-stage MACT
compliance scheme is justified for existing sources at bleached
paper-grade kraft and soda mills that enroll in the water
Voluntary Advanced Technology Incentives Program (see 61 FR 9394
for a similar argument relating to compliance with MACT for
washers and oxygen delignification systems). The first stage is
an interim MACT of no backsliding--which reflects the current
level of air emissions control. The second stage requires
compliance with revised MACT based on baseline BAT requirements
for all parameters for bleached paper-grade kraft and soda mills.
(The second stage in effect revises MACT to reflect the control
technologies which will be available at this later date. See Act
section 112(d)(6). ) The no-backsliding provisions apply to the
period from 60 days from publication in the Federal Register
until compliance with the second-stage MACT standards is required
6 years from publication in the Federal Register. This two-step
17-11
-------
alternative is available only to bleached paper-grade kraft and
soda mills actually making the binding decision to comply with
Tier I, II, or III water limitations.
The EPA believes that providing these mills an additional
3 years to comply with MACT (i.e., baseline BAT requirements for
all parameters) is an appropriate and logical outgrowth of the
discussions set forth in the March 8, 1996 Federal Register
supplemental notice (61 FR 9393) and the July 15, 1996 Federal
Register supplemental effluent guidelines notice (61 FR 36835-
58). In the March 8, 1996 Federal Register supplemental notice,
EPA solicited comments on its preliminary findings that MACT for
chloroform air emissions should be compliance with baseline BAT.
Commenters agreed with this preliminary determination. In the
July 15, 1996 Federal Register notice, EPA set forth its vision
of more stringent BAT for mills that voluntarily enter the
Advanced Technologies Incentives program. As part of that
voluntary program under the water standards, EPA is promulgating
a requirement that mills in Tiers II and III, at a minimum, meet
all the limitations promulgated as baseline BAT no later than 6
years after publication in the Federal Register. Thus, more
stringent air emission controls than stage one MACT will likewise
be available at this time since compliance with these interim BAT
limitations will result in compliance with MACT. For Tier II and
Tier III mills, this means that the second stage MACT requirement
is compliance with the baseline BAT limitations by 6 years from
date of publication in the Federal Register. The same is the
case for Tier I mills, even though under the water regulation
Tier I mills will be required to achieve more stringent
limitations at that time. The EPA is defining MACT to be the
baseline BAT limitations even in this situation because
17-12
-------
compliance with the more stringent absorbable organic halides
limitations and other requirements unique to Tier I are
unnecessary to control chloroform emissions at these mills.
The EPA further believes that most plants likely to elect to
comply with a tier option already control air emissions of
chlorinated HAP's (both chloroform and other chlorinated HAP's)
through application of the MACT technologies (process
substitution for chloroform and caustic scrubbing for the
remaining chlorinated HAP's). Thus, there will be some control
of the emissions from these bleaching operations during the time
preceding compliance with the second stage of MACT. To ensure
that there is no lessening of existing controls, EPA also is
promulgating a no backsliding requirement as an interim MACT --
reflecting current control levels. During the extended
compliance period, mills may not increase their application rates
of chlorine or hypochlorite above the average rates determined
for the 3-month period prior to 60 days after promulgation in the
Federal Register.
The EPA notes that an affected bleached paper-grade mill
must comply with the MACT requirements no later than 3 years from
publication in the Federal Register, even if the mill's existing
CWA NPDES permit does not yet reflect the corresponding effluent
limitation guidelines and standards because its existing terms
have not expired or it has been administratively extended. Put
another way, even if a mill's existing NPDES permit serves as a
shield (until reissuance) against imposition of new limits based
on new effluent limitations guidelines (see CWA section 402(k)),
the MACT requirement for bleached paper-grade mills to control
chloroform emissions through compliance with all parameter
requirements in the effluent limitation guidelines and standards
17-13
-------
take effect to satisfy the requirements of the Act. Similarly,
if a bleached paper-grade mill's NPDES permit is reissued sooner
than the expiration of the 3-year compliance schedule authorized
for the chloroform MACT requirements and calls for immediate
compliance with the BAT limitations, that deadline would prevail.
The same principles will apply when effluent limitations
guidelines and MACT standards are promulgated for dissolving-
grade mills.
Comment: One commenter (20,027) stated that the benefits of
the December 17, 1993 proposed rule (i.e., orderly, coordinated
approach to air and water pollution control) will not be realized
unless the compliance schedules established in the December 17,
1993 proposed rule apply uniformly to all pulp and paper sources.
Response: The benefit of the December 17, 1993 proposed
rule is that regulatory requirements and compliance dates would
be known in advance. This advance knowledge provides a mill with
the opportunity to plan and choose the appropriate method of
compliance that satisfies all rules and at the lowest cost for
each mill. Additionally, the proposed rule contains extended
compliance schedules for specified pulping area systems and
incentive programs for bleaching systems to provide increased
flexibility for mills to develop and implement compliance
strategies.
Comment: One commenter (20,083) recommended extending the
compliance deadline by 3 years for sources subject to different
definitions of "new source" under the Act and CWA. The result is
that, if a source begins construction after the December 17, 1993
proposed rule but before promulgation, the source is "new" under
Act but not under CWA. Consequently, the source would have to
begin planning for compliance with the air requirements before
17-14
-------
promulgation, yet the final rule might impose water requirements
that would make those plans and expenditures worthless. One
commenter (20,057A2) expressed that the proposed rule should not
be applicable to construction or reconstruction during the period
prior to the final promulgation.
Response: The EPA proposed a broad definition of "source"
for the proposed rule in order to reduce or eliminate the number
of sources which would be defined differently by the Act and CWA.
If "source" is defined to include all pulping processes, all
bleaching processes, and all associated wastewater streams at
mills, there will be far fewer instances in which a source will
be constructed or reconstructed between proposal and promulgation
than if a "source" is defined to be an individual piece of
process equipment. If source is defined broadly, a piece of
equipment that is added will not constitute a "new source" in
most situations, but instead will be considered a change to an
existing source. Such changes would be required to comply with
the existing source standards at some period of time after
promulgation of the standards, when all requirements of the
guidelines are known.
Comment: With regard to the March 8, 1996 Federal Register
supplemental notice, two commenters (IV-D2-18, IV-D2-17)
supported EPA's OW suggestion to specify the application of BAT
as the compliance mechanism for bleaching wastewater in place of
numerical emission limits. The commenters suggested, however,
that under such an approach the compliance date for dissolving
mills would need to be deferred, because the BAT for these mills
will not be established by the time the December 17, 1993
proposed rule is promulgated. One commenter (IV-D2-17) noted
that there is significant environmental benefit for withdrawing
17-15
-------
and reserving dissolving-grade MACT until the ongoing technology
development for BAT is completed. Namely, if EPA were to set
MACT for chloroform now, based on current process technology, the
MACT floor level of control would be no control with no
reasonable add-on control technology.
Response: The MACT floor level of control for all bleaching
systems is 99 percent reduction of chlorinated HAP's using
caustic scrubbing and process modifications (100 percent chlorine
dioxide substitution and elimination of hypochlorite as a
bleaching agent) . The technology basis for BAT under the CWA are
at least equivalent to the MACT requirements. Since BAT and MACT
are essentially the same, EPA therefore proposed in the March 8,
1996 supplemental notice that chloroform emissions be controlled
by complying with the BAT requirements. No adverse comments were
received to this proposal.
As stated in the July 15, 1996 Federal Register notice
(61 FR 36835), EPA is evaluating new data on the technical
feasibility of reducing hypochlorite usage and implementing high
levels of chlorine dioxide substitution on a range of dissolving-
grade pulp products. Therefore, EPA is deferring issuing
effluent limitation guidelines and standards for dissolving-grade
mills until the comments and data can be fully evaluated. The EPA
expects to promulgate final effluent limitation guidelines and
standards for dissolving-grade subcategories at a later date.
The EPA has decided to delay establishing these MACT
standards for chloroform and for other chlorinated HAP's for
dissolving-grade bleaching operations until promulgation of
effluent limitation guidelines and standards for those
operations, for the following reasons. With respect to the MACT
standard for chloroform, first, as explained above and in the
17-16
-------
March 8, 1996 Federal Register notice, the control technology
basis for the effluent limitation guidelines and standards and
the MACT requirements will be the same. Second, at present, the
Agency is unsure what level of chlorine substitution and
hypochlorite use is achievable for dissolving-grade mills. Thus,
although EPA has a reasonably good idea what the technology basis
of MACT and effluent limitation guidelines and standards is
likely to be for dissolving-grade mills, the precise level of the
standards remains to be determined. Consequently, at present,
EPA is unable to establish what the MACT floor would be for
chloroform emissions from bleaching systems at these mills, and
there is no conceivable beyond-the-floor technology to consider.
The EPA will make these determinations based on data being
developed, and then promulgate for these mills effluent
limitation guidelines and standards and, concurrently, MACT
standards based on those effluent limitation guidelines and
standards. Covered mills would therefore be required to comply
with the MACT standards reflecting performance of the effluent
limitation guidelines and standards no later than 3 years after
the effective date of those standards, pursuant to
section 112 (i) (3) (A) of the Act.
The basis for delaying MACT requirements for chlorinated
HAP's other than chloroform (again, from dissolving-grade bleach
operations only) differs somewhat. As noted above, the
technology basis for control of these HAP's is use of a caustic
scrubber. However, when plants substitute chlorine dioxide for
chlorine and eliminate hypochlorite (in order to control
chloroform emissions and discharges to water, as explained
above), a different scrubber will be needed that can adequately
control both the chlorine dioxide emissions for worker safety
17-17
-------
reasons and the emissions of chlorinated, non-chloroform HAP's.
The Agency's concern (shared by the commenters who addressed this
question) is that immediate control of the non-chloroform
chlorinated HAP's could easily result in plants having to install
and then replace a caustic scrubber system in a few years due to
promulgation of effluent limitation guidelines and standards and
MACT requirements for chloroform. This result is an
inappropriate utilization of scarce pollution control resources.
Comment: One commenter (IV-D2-15) contended that if an
alternative technology will produce the same or greater emission
reductions than specified in the rule, EPA should work with the
source to develop a reasonable compliance timetable. If after
the technology is installed and operating normally, it does not
achieve the reductions that were predicted, or if operation of
the technology does not turn out to be practicable, the source
should be required to revert back to the original control
requirements. In that event, the source should negotiate a new
compliance date with the Administrator. The source should be
required to comply with the original requirements as soon as
practicable under the circumstances, but in no event later than
8 years after the technology has been found to be inadequate or
impracticable.
Response: If an affected source wishes to establish an
alternative means of emission reduction, the affected source can
apply for and obtain approval for this alternate compliance
method through the procedures outlined in § 63.6(g) of the
general provisions. The EPA maintains that the 3-year compliance
schedule outlined in the rule is reasonable and further adds that
all kraft mills have a 5-year extension for compliance on HVLC
systems. If an alternate compliance method does not reach the
17-18
-------
standards set in the rule, and if the affected source cannot
reach compliance by the target date, the affected source may
apply for a 1-year extension of the compliance date under the
procedures outlined in section 112(i) (3) of the Act.
17-19
-------
-------
18.0 MECHANICAL PULPING MILL, SECONDARY FIBER
PULPING MILL, NON-WOOD FIBER PULPING MILL,
AND PAPERMAKING SYSTEM (MACT III) COMMENTS
Comment: Several commenters (IV-D2-1, IV-D2-4, IV-D2-5,
IV-D2-7, IV-D2-8, IV-D2-9, IV-D2-10, IV-D2-12) stated that they
supported the March 8, 1996 standards proposed by EPA for
MACT III sources (i.e., floor for pulping systems at these mills
and papermaking systems at all mills is no control and for
bleaching systems at these mills that use chlorine or chlorine
dioxide, control is caustic scrubbing).
Three commenters (IV-D2-1, IV-D2-5, IV-D2-8) agree with
EPA's findings that the floor level of control for papermaking
systems is no control. The same commenters (IV-D2-1, IV-D2-5,
IV-D2-8) also supported further examination of HAP emissions
attributable to the use of papermaking system additives in order
to determine if these emissions are a major source of HAP.
Three commenters (IV-D2-2, IV-D2-3, IV-D2-6) disagreed with
EPA's findings that the floor level of control for MACT III
sources (mechanical mills, secondary fiber mills, non-wood fiber
mills, and papermaking systems) was no control. Due to the lack
of available data, the commenters (IV-D2-2, IV-D2-3, IV-D2-6)
stated that EPA should wait for the conclusion of the MACT III
testing program sponsored by industry before promulgating a final
rule. The commenters (IV-D2-2, IV-D2-3, IV-D2-6) stated that
18-1
-------
they believed there were not enough data to substantiate a
finding of no control for MACT III sources.
Response: The information gathered during the MACT
development process indicates that there are no air pollution
control devices in place at mechanical mills, secondary fiber
mills, and non-wood fiber mills except at elemental chlorine and
chlorine dioxide bleaching stages. This information also
indicated that no air pollution control devices are currently in
place on papermaking systems at any mill (A-95-31, II-B-1).
Through site visits, working with stakeholders, and reviewing the
results of the industry-sponsored MACT III testing program, EPA
maintains that the floor level of control for these sources is no
control except for elemental chlorine and chlorine dioxide
bleaching stages. The EPA has also concluded that going beyond
the floor level of control requiring controls for MACT III
sources would be cost-prohibitive given the estimated reduction
in HAP's (A-95-31, IV-B-5, IV-B-6, IV-B-7, IV-B-8). Therefore,
EPA decided to move forward with the MACT standard for the
MACT III sources and promulgate in conjunction with standards for
MACT I sources. The EPA believes this is a sound decision that
will result in time and money savings for the Agency and
stakeholders, and that no environmental benefit would be gained
by delaying promulgation of the standards for MACT III sources.
Comment: One commenter (IV-D2-10) requested that EPA give
MACT III sources additional time in which to comply if the
promulgated rule is more stringent than the proposed rule.
Response: The proposed standard for MACT III mills was no
control except for chlorine bleaching stages. For chlorine
bleaching stages, EPA concluded that scrubbers were already used
to control chlorine and HC1 for process and worker safety
18-2
-------
reasons. The final standard does not contain any additional
control requirements for MACT III mills than stated at proposal.
Therefore, EPA maintains that the promulgated standard is no more
stringent than the proposed standard and additional compliance
time is not necessary.
Comment: Several commenters (IV-D2-2, IV-D2-3, IV-D2-6)
requested that EPA provide information on the industry sampling
and testing protocol and any data collected through the industry
test program.
Response: The information provided on the industry test
program is available in the Pulp and Paper MACT III docket
(A-95-31, IV-J-3 through IV-J-13). The sampling and testing
protocol provided by the industry are entries II-D-5 and II-D-11.
Comment: Two commenters (IV-D2-2, IV-D2-3) requested that
EPA publish the final MACT III rule in a separate Federal
Register notice. The commenters (IV-D2-2, IV-D2-3) stated that
the inclusion of MACT III sources at the end of the MACT I
sources has inadvertently limited public comment on the proposed
rule because State and local organizations without MACT I sources
would not look for a MACT III rule in the document in which the
rule was proposed.
Response: The EPA concluded that the most efficient way to
address MACT III sources was to propose standards for this source
group in combination with the announcement in the March 8, 1996
Federal Register supplemental notice. Both actions were stated
at the beginning of the notice after the title "National Emission
Standards for Hazardous Air Pollutants for Source Category: Pulp
and Paper Production," followed by a summary of the action
describing what was addressed in the notice. A separate section
for the MACT III mills, section X, "Standards for Mechanical
18-3
-------
Mills, Secondary Fiber Mills, Nonwood Fiber Mills, and Paper
Machines," was provided. Consequently, EPA maintains that
adequate notice was qiven for comment response on the proposed
MACT III rule. Durinq subsequent work qroup meetinqs which
included State and local representatives, notification was qiven
that EPA intends to combine the MACT III promulqation with the
final pulp and paper rule.
Comment: One commenter (IV-D2-9) questioned whether or not
the MACT standards for these sources were warranted. The
commenter (IV-D2-9) further stated that due to the low emissions
from their mill they believed that no MACT standards were
necessary.
Response: All cateqories and subcateqories of major sources
that are listed pursuant to section 112 (c) of the Act must be
evaluated for possible NESHAP. Since MACT III sources are major
sources of a listed source cateqory, EPA is legally bound to
determine MACT standards pursuant to section 112(d) of the Act.
Comment: Two commenters (IV-D2-8, IV-D2-10) confirmed that
MACT III sources that bleach with elemental chlorine are already
using scrubbers for chlorine emission control. Both commenters
(iv-~2-8, IV-D2-10) indicated that requlatory controls for
elemental chlorine bleachinq systems in addition to scrubbers
such as incineration, would not be appropriate. One commenter
(IV-D2-10) interpreted the term "chlorine bleachinq" to
exclusively represent the use of elemental chlorine as a
bleachinq aqent, and "non-chlorine bleachinq" to represent any
bleachinq aqent other than elemental chlorine. One commenter
(IV-D2-8) aqreed that bleach plants at MACT III sources that are
collocated with MACT I sources will be subject to the MACT I
controls if any equipment is common to both process lines.
18-4
-------
Response: For collocated and stand-alone MACT III mills
that have elemental chlorine or chlorine dioxide bleaching
stages, these stages shall reduce the total chlorinated HAP mass
in the vent stream entering the control device (scrubber) by
99 percent or more by weight, achieve a treatment device
(scrubber) outlet concentration of 10 ppmv or less of total
chlorinated HAP (other than chloroform), achieve an outlet mass
emissions limit of 0.001 kg of total chlorinated HAP (other than
chloroform) per Mg ODP. Information available to EPA (A-95-31,
IV-B-5) indicated that MACT III mills bleaching with elemental
chlorine or chlorine dioxide already employ scrubbers.
Information from industry (A-95-31, IV-B-5) also indicated that
the majority of MACT III mills that have bleaching systems use
hypochlorite and are not controlled. Furthermore, available test
data show that HAP emissions from hypochlorite bleaching stages
are not large. Therefore, EPA has concluded that control
requirements for hypochlorite bleaching stages at MACT III mills
are not warranted. MACT III mills with hypochlorite bleaching
stages are exempt from any bleaching control requirements. In
addition, MACT III mills that use TCF bleaching are also exempt
from any bleaching control requirements.
Comment: One commenter (IV-D2-3) stated that they supported
capture and combustion of LVHC gas streams and requested that EPA
investigate the feasibility of controlling these streams.
Response: Based on available information, EPA believes that
LVHC streams do not exist at MACT III mills. Information
available to EPA (A-95-31, IV-B-7) indicate the HVLC streams at
MACT III mills are not controlled and that HAP emissions from
these sources are low. Therefore, EPA concluded that little
18-5
-------
environmental benefit would be gained by controlling the HVLC
streams at MACT III mills.
Comment: One commenter (IV-D2-14) stated that the
definition for non-wood pulping should include other sources of
non-wood fiber and not just flax. The commenter (IV-D2-14) also
recommended a detailed process description for non-wood pulping.
Response: The EPA agrees with the recommendation to adjust
the definition for non-wood pulping. The EPA has defined
non-wood pulping as the production of pulp from fiber sources
other than trees. The non-wood fiber sources include, but are
not limited to, bagasse, cereal straw, cotton, flax straw, hemp,
jute, kenaf, and leaf fibers. Since the promulgated standard for
non-wood pulping is no control, EPA determined that a detailed
process description was not needed.
18-6
-------
19.0 MISCELLANEOUS COMMENTS
19.1 INFORMATION OMITTED FROM THE DOCKET
Comment: One commenter (20,027) indicated that information
submitted to EPA by Mead regarding process equipment, economic
impacts, and non-water quality environmental impacts associated
with soda pulping was not included in the docket at the time of
proposal. The commenter (20,072A1) included the Mead Corporation
information in their comments on the proposed rule. In addition,
information submitted by Weyerhauser Corporation and NCASI was
omitted from the docket at the time of proposal. The American
Forest Products Association (AF&PA) has resubmitted this
information as appendices MACT 7, 8, and 9 (20,027A7, 20,027A8,
20,027A9).
Response: All of the data and public comments regarding the
proposed MACT standards and the final MACT standards that were
not claimed confidential were submitted to the public docket
maintained by EPA's OW.
19.2 ADDITIONS OR CORRECTIONS TO THE PROPOSED RULE
Comment: One commenter (20,056) indicated the following
corrections to the December 17, 1993 proposed rule: (1) the
units of the de minimis level for the bleaching system should be
standard cubic meters per minute (not standard cubic feet);
(2) § 63.444(a) (5) referred to item (1) (iii) which does not
exist; (3) in § 63.444 (f) (1), the "or" should be substituted for
19-1
-------
"and" in the following excerpt: "... knowledge of the process,
and mass balance information..." so that process and mass balance
information may be used.
Response: Valid editorial mistakes identified by commenters
have been corrected.
Comment: One commenter (20,102) stated that procurement
guidelines should be established to encourage the reduction of
both HAP generation at the production facility and the quantity
of solid waste produced after use of the paper product.
Response: The EPA does not have the authority under the Act
to establish procurement guidelines.
19.3 OTHER COMMENTS
Comment: One commenter (20,091) stated that verification of
compliance by the regulatory agencies responsible for pollution
control is essential for public confidence and environmental
protection. The commenter (20,091) argued that this area of the
system needed improvement and suggested more frequent unannounced
inspections (with sampling and testing) should be conducted by
State and Federal agencies.
Response: All facilities will have to perform tests for
compliance or acceptably document emission control. Notification
is required prior to performance tests so that the tests can be
attended by a State or Federal representative. Test results must
be submitted to the regulatory agency to demonstrate compliance.
Facilities will have to monitor operational parameters,
established through performance testing, on an ongoing basis to
prove compliance. Facilities must submit semi-annual reports of
compliance status, and report any infractions. The EPA believes
that the monitoring, recordkeeping, and reporting requirements of
this rule provide acceptable assurance of compliance.
19-2
-------
Comment: One commenter (20,089) stressed that EPA failed to
provide adequate public notice and opportunity for public comment
on its general MACT determination for most segments of the
regulated community. The commenter (20,089) stated that most of
EPA requests for general comments on the MACT floor have been
contained in industry-specific rulemakings which do not affect
large portions of the regulated community. The commenter
(20,089) indicated that the notice of the rulemaking will likely
reach only limited industry segments since EPA addresses general
comments on the MACT floor determination within rulemakings for
specific MACT categories.
Response: The EPA requested comments on a discussion of the
statutory interpretation and determination of the MACT floor for
HON and other sources on March 9, 1994 (59 FR 11018) . Comments
were received from industries, trade associations, environmental
groups, State and local agencies, and labor unions. On June 6,
1994 (59 FR 29196), EPA published a discussion of the statutory
interpretation and determination of the MACT floor. While this
notice established general interpretation, it also stressed that
EPA has certain areas of discretion within the statutory
framework to determine how best to set the MACT floor for each
source category considering the data available for each category.
The notice also envisioned that as additional MACT standards are
developed, they may raise new issues pertaining to the MACT floor
(although no such issues are present in this rule, and the Agency
accordingly applied the interpretation out in the June 6 notice).
As stated in the June 6, 1994 notice, to properly consider the
specifics of each source category, EPA will solicit and fully
consider comments on individual MACT standards, including
comments on interpretation of section 112 regarding MACT floor
19-3
-------
determination. All section 112 standards are proposed in the
Federal Register and anyone is free to submit comments on any
proposed rule. A revision of EPA procedures for requesting
public comment on MACT floor determinations is not planned.
Comment: One commenter (20,085) expressed an interest in
establishing health monitoring or health surveillance programs to
determine the effectiveness of pollution control programs.
Response: This activity is not covered under the scope of
this NESHAP. The purpose of the testing and monitoring
requirements of the NESHAP must be to ensure compliance with the
emission limits in the rule. The EPA has no authorization under
section 112 to monitor public health because health data are not
needed to determine compliance with the MACT standards.
Comment: One commenter (20,078) suggested that EPA conduct
a pilot level regulatory/permitting process using the proposed
rule to identify whether or not the chosen limits can be
implemented.
Response: The limits specified in this rule have been set
from actual emission levels obtained at the best performing
(12 percent) mills in the nation. MACT standards are set in this
manner to insure that the specified limits can be actually met in
practice.
19-4
-------
20.0 ENVIRONMENTAL AND COST IMPACTS
20.1 INTRODUCTION
This section summarizes the environmental and cost impacts
of the final rule. This section also discusses the major
revisions to the environmental impacts and associated analyses at
proposal. A complete description of the methodology for
estimating impacts was presented in the 1993 BID (A-92-40,
II-A-35) . The 1993 BID, along with this section, documents EPA's
conclusions concerning demonstrated control technologies, HAP
emissions, control costs, and other impacts upon which the final
rule is based.
The final rule covers chemical and semi-chemical wood
pulping and bleaching processes and papermaking systems at the
following types of mills:
chemical and semi-chemical wood pulp mills;
integrated mills (mills that combine on-site production
of both pulp and paper);
mechanical wood pulping mills;
secondary fiber deinking and non-deinking mills; and
non-wood pulping mills.
20-1
-------
Such mills typically fall under standard industrial
classification (SIC) codes 2611 and 2621 for pulp and integrated
mills, respectively.
The only processes regulated at non-chemical mills
(mechanical wood pulping mills, secondary fiber deinking and non-
deinking mills, and non-wood mills) are chlorine and chlorine
dioxide stages in bleaching operations (see section 16.0). As a
result, most of the analyses in this chapter center on chemical
and semi-chemical pulping and bleaching processes because these
are the processes that are most affected by the final rule and
because the pulping and bleaching processes occur at the same
mill. Unless otherwise specified, references to pulp mills or
pulping or bleaching processes should be interpreted to mean only
wood pulping and bleaching processes at chemical and semi-
chemical pulp mills and integrated mills. The only regulated
process condensates are from the kraft pulping process.
This section presents an overview of the revisions made to
the impacts analyses performed at proposal. Where the full
rationale for specific revisions can be briefly presented, this
chapter presents the relevant information. For some of the more
complex technical issues, this section summarizes the technical
approach, explains the assumptions, presents the results, and
refers the reader to the documents contained in the public docket
for the detailed technical analyses.
20.1.1 Section Organization
Section 20.2 characterizes the pulp and paper industry and
includes process descriptions, the emissions estimation approach
and estimated baseline emissions. Section 20.3 discusses
applicable control technologies. Section 20.4 presents the
20-2
-------
technical approach for estimating the impacts of the final rule.
Section 20.5 relates the approach taken to estimate control costs
and section 20.6 documents the development of the data base used
to estimate national environmental and cost impacts for the pulp
and paper industry. Section 20.7 summarizes the cost and
environmental impacts of all MACT standards on the pulp and paper
industry, before and after the effluent guidelines limitations
have been implemented.
20.2 PROCESS DESCRIPTIONS AND EMISSIONS ESTIMATES
This section characterizes mill processes and baseline
emissions. Section 20.2.1 describes the nationwide distribution
of pulp and paper mills in the U.S, section 20.2.2 describes
changes to the emissions estimation approach used since proposal,
and section 20.2.3 presents the baseline emissions estimates and
control technology assumptions.
20.2.1 Industry Characterization
The pulp and paper industry includes facilities that
manufacture pulp, paper, or other products from pulp. Converting
operations, such as the production of paperboard products
(e.g., containers and boxes) and coating or laminating, are not
included in this assessment.
There are approximately 566 operating pulp and paper mills
in the United States. This estimate was determined from
responses to a 1990 EPA OW survey. (The responses to the survey
are considered Confidential Business Information.) This number
reflects both chemical and non-chemical mills. Many of these
pulp and paper mills operate more than one type of pulping
process; for example, a mill may produce pulp using a chemical
process (e.g., kraft or sulfite) and a mechanical or semi-
20-3
-------
chemical process. Industry correspondence and literature
sources, reflecting closures and changes that have occurred since
the initial survey, were used to adjust the total number of
chemical and semi-chemical mills. As of fall 1996, there were
156 total mills operating some combination of kraft, semi-
chemical, sulfite, or soda pulping processes. Table 20-1 shows
the distribution of the 156 mills in each State by type of
chemical or semi-chemical pulping process used. The States with
the highest concentration of chemical pulp mills are Washington,
Alabama and Georgia. Of the 156 mills, 112 are kraft mills, 16
are semi-chemical mills, 2 are soda mills, 15 are sulfite mills,
10 are co-located kraft and semi-chemical mills, and 1 is a co-
located kraft and sulfite mill.
Other sources of information used to characterize mills at
proposal included the 1992 voluntary NCASI survey, site visits,
and literature sources (such as the Lockwood Post's Directory).
After proposal, EPA received comments and new information from
the industry (the March 8, 1996 Federal Register notice presents
a listing of the new data) . As stated above, the OW survey has
been continuously updated (with information on mill names,
closures, production capacities, bleaching sequences, and number
of process systems) as has the Lockwood Post's Directory (with
information on mill names, pulping processes, and production
capacities). The latest updates for the OW survey and Lockwood
Post Directory (A-92-40, IV-J-87) occurred in 1996.
20.2.2 System Approach to Emissions Estimation
20.2.2.1 System Approach Issues. At proposal, EPA
developed emission factors for each type of individual emission
point typically found at pulp and paper mills. To estimate
20-4
-------
TABLE 20-1. DISTRIBUTION OF CHEMICAL AND SEMI-CHEMICAL PULP
PROCESSES IN THE UNITED STATESa
State Kraft Semi-chemical Soda Sulfite
Alabama 14 2
Alaska 1
Arizona 1
Arkansas 7
California 2
Florida 7 l
Georgia 12 2
Idaho 1
Indiana 1
Iowa 1
Kentucky 2 l
Louisiana 10 3
Maine 7 1
Maryland 1
Michigan 3 3
Minnesota 2
Mississippi 6
Montana 1
New Hampshire 1 1
New York 1 2
North Carolina 5
Ohio 1 2
Oklahoma 1
Oregon 7 2
Pennsylvania 3
South Carolina 6 1
Tennessee 2 1
Texas 6
Virginia 4 2
Washington 63 4
Wisconsin 41 5
Total 123 26 2 15
aln this table, mills with more than one pulp process are counted once for
each pulp process (e.g., a mill with kraft and semi-chemical processes is
listed in both kraft and semi-chemical columns). Of the 156 total mills,
112 are kraft mills, 16 are semi-chemical mills, 2 are soda mills, 15 are
sulfite mills, 10 are co-located kraft and semi-chemical mills, and 1 is a
co-located kraft and sulfite mill. Note that the sulfite mill in Alaska is
closing.
Source: 1990 EPA OW Survey, adjusted as of Fall 1996
20-5
-------
emissions, emission points (e.g., the digester, knotter, and
washer) were grouped based on operating parameters believed to
affect emissions. Their emission factors were averaged, and then
assigned to a pulping line model process unit. After receiving
additional test data following proposal, EPA adopted an emissions
estimation approach based on mill systems. A mill system is a
collection of equipment and ancillary tanks and piping that
performs a discrete operation (e.g. the pulp washing system
consists of pulp washer, filtrate tank, and foam tank) . Test
data from systems where the complete system was evaluated
(i.e., all the emission points in the system were tested) were
analyzed on a system basis rather than on an emission point
basis, and emission factors for each system were developed.
Emissions were then estimated for each mill based on which
systems were present, according to survey results.
The EPA has concluded, after assessing the additional test
data and industry concerns regarding the emission point approach,
that the mill system approach is a better tool for analyzing the
data and yields results that more accurately reflect the actual
emissions from the industry. Details of industry comment and EPA
response on this issue are contained in chapter 2.0, Industry
Characterization. Details of the system approach to estimating
mill emissions, including estimated emission factors for each
mill system, are contained in the revised Chemical Pulping
Emission Factor Development Document (A-92-40, IV-A-8). A
summary of the rationale for using the mill system approach is
contained in the March 8, 1996 Federal Register supplemental
notice.
20-6
-------
20.2.2.2 Pulping Processes. The proposal BID describes the
pulping process and provides a general overview of pulping
technologies and the types of equipment common to the industry.
As a basis for the creation and selection of model process units,
the document also notes which operating parameters influenced air
emissions. Several assumptions and conclusions have been revised
based on public comment and data submitted since proposal.
Specifically, EPA has modified the following assumptions and
conclusions since the proposal BID:
At proposal, two model emission points had been
assigned to digester blow gases on the assumption that
digester blow gas emissions differ between batch and
continuous digesters. (Specifically, batch digesters
release gases in surges when the digester blows its
entire load into a blow tank; continuous digester
emissions are released at a constant rate.) Digester
blow gases are now included in the "digester system and
evaporators" mill system for both types of digestion
because EPA's analysis of the data did not show a
significant difference in the quantity of emissions as
a function of digester type.
The proposal BID suggested that brownstock washer
emissions are a function of pulp production, type of
digestion (batch or continuous), type of wood pulp
(softwood or hardwood), and point of shower water
application. New data, however, do not support
establishing different emission factors for washer
emissions on any basis other than washer type and HAP
concentration in the shower water. The EPA's final
analysis includes emission factors for low-flow washer
systems (e.g., belt presses and diffusion washers) and
high-flow washer systems (i.e., rotary vacuum drum
washers). Based on the data, rotary vacuum drum washer
systems are differentiated by the HAP concentration in
the shower water.
The proposal BID noted a relationship between wood type
and emission rates for semi-chemical pulping processes.
20-7
-------
The final analysis does not differentiate mill systems
by wood type as the data do not support a significant
difference in emission rates based on wood type for
semi-chemical pulping processes.
20.2.2.3 Bleaching Processes. At proposal, EPA developed
emission factors for each bleaching stage in the bleaching
process. Each bleaching sequence (i.e., series of bleaching
stages) was assigned emission factors based on the type of stages
present.
Since proposal, EPA concluded that emissions from bleaching
processes are more a function of mill operating parameters and
equipment rather than bleaching sequence. The EPA statistically
analyzed all the emission data from bleaching processes and the
associated mill parameters (presence of oxygen delignification,
bleaching sequence, degree of chlorine dioxide substitution, use
of hypochlorite, wood type, inlet methanol concentration in the
bleaching process shower water, and air flow rate of bleach plant
vents). The results of the statistical analyses indicated that
only the presence of a hypochlorite stage in the bleaching
sequence and the degree of chlorine dioxide substitution
significantly affect the level of chlorinated HAP emissions. The
EPA was unable to detect a significant difference in emissions as
a result of bleach sequence (apart from the presence of a
hypochlorite stage or any of the other parameters). The EPA
developed three emission factors to represent total bleach plant
air emissions: one for bleach plants with a hypochlorite stage,
one for bleach plants without a hypochlorite stage and with a low
degree of chlorine dioxide substitution (60 percent or less), and
one for bleach plants without a hypochlorite stage and with a
high degree of chlorine dioxide substitution (greater than
-------
60 percent) . Details of the analysis of bleach plant air
emissions are contained in the revised Chemical Pulping Emission
Factor Development Document (A-92-40, IV-A-8). The emission
factors were assigned to each bleaching process based on which
case fit each bleach sequence at a mill.
20.2.3 Baseline Emissions
Baseline emissions are the uncontrolled emission estimates
adjusted for the effects of current State and Federal
regulations, as well as additional controls known to be currently
in place based on the MACT survey. The revised estimation of the
baseline level of control for kraft, semi-chemical, soda and
sulfite pulping processes and for bleaching processes is
documented in detail in a memorandum contained in the public
docket (A-92-40, IV-B-16). The memorandum also presents the
percent of systems that are controlled at each type of mill.
Two of the most significant revisions to the baseline level
of control since proposal that affect the baseline emissions are
the following:
Control of all kraft LVHC vents (digester, evaporator
and turpentine recovery system vents and steam stripper
overheads) is now considered to be included in the
baseline level of control.
Control of only chlorinated HAP through caustic
scrubbers at all bleach plants (and no control of
non-chlorinated HAP) is now considered to be included
in the baseline level of control.
Additionally, OW is promulgating BAT controls that would
affect pulping and bleaching processes. Under one OW control
option (Option A) , the BAT controls would require 100 percent
substitution of chlorine with chlorine dioxide and elimination of
hypochlorite as a bleaching agent at all paper-grade bleaching
20-9
-------
processes. Option A would also require mills to replace
vibratory screens with lower-emitting pressure screens and add a
low- air flow washing stage to the washing system. These changes
decrease the emissions from the pulping area. A second OW option
(Option B) would require bleached paper-grade kraft mills to
apply oxygen delignification in addition to the requirements of
Option A. The addition of an oxygen delignification system will
increase the concentration of methanol in process water recycled
to the decker system. Baseline emissions are increased because
of additional oxygen delignification units and higher emissions
from the decker system using dirtier (i.e., higher HAP
concentration) process water from oxygen delignification
filtrates. The BAT requirements have not yet been established
for dissolving-grade bleaching processes.
At proposal, data available to estimate HAP emissions from
pulping and bleaching processes were limited. These data
included a field test program of air and liquid samples from four
kraft and one sulfite mills (referred to as EPA 5-mi11 study)
(A-92-40, II-A-17 a through d) and some limited industry data
(see the proposal BID) . In their comments to the proposed rule,
industry representatives maintained that these data were
insufficient to accurately characterize emissions. Following
proposal, industry commenters supplied EPA with additional test
data from kraft, sulfite, semi-chemical, and soda mills. The EPA
evaluated and incorporated the data into its analyses. The
revised emission factors for mill systems are contained in the
revised Chemical Pulping Emission Factor Development Document (A-
92-40, IV-A-8).
20-10
-------
Uncontrolled emissions from pulping vent streams are
calculated by multiplying the Ib/ODTP for each pollutant in each
equipment system (e.g., pulp washing system), the pulp capacity
(ODTP/day) for each equipment system at each mill, and the hours
of operation per year. Baseline emissions from pulping vent
streams are calculated by applying the emission reduction
efficiency of existing control devices (e.g., 98 percent
reduction for combustion devices) associated with each equipment
system to the uncontrolled emissions from each system. Baseline
emissions of chlorinated HAP from bleaching vent streams are
calculated by applying the emission factors specific to the
process conditions at each bleaching process (i.e., C1C>2
substitution level and hypochlorite use) . Baseline emissions of
non-chlorinated HAP do not change because of process conditions.
Emissions are calculated by multiplying the non-chlorinated HAP
emission factors, the pulping capacity of the mill, and the
operating hours.
Uncontrolled emissions from pulping process condensates are
calculated by multiplying the mass of each pollutant in the
condensate by the fraction emitted (Fe) values developed from
wastewater treatment models. The models are discussed in the
revised Chemical Pulping Emission Factor Development Document
(A-92-40, IV-A-8). Baseline emissions from pulping process
condensates are calculated by applying the reduction efficiency
of existing control devices (e.g., steam stripping combined with
overhead vent combustion) associated with the condensate streams.
Table 20-2 presents uncontrolled and baseline emissions for
an example mill. Table 20-3 summarizes estimated national
baseline emissions from the pulp and paper industry (before and
20-11
-------
after OWs BAT options are applied), and includes estimates for
total HAP, total VOC, TRS, and the 25 highest emitted compounds.
As shown in the table, methanol is the largest constituent
contributing to total HAP emissions.
The analysis does not include air emissions from mechanical
wood pulping mills, secondary fiber deinking and non-deinking
mills, non-wood pulping mills, paper machines, or chemical
recovery at chemical and semi-chemical mills. Air emissions of
HAP's from mechanical wood pulping mills, secondary fiber
deinking and non-deinking mills, non-wood pulping mills and paper
machines are discussed in the September 29, 1995 presumptive MACT
report for non-chemical and other pulp and paper (MACT III) mills
(A-95-31, II-B-1).
20.3 EMISSION CONTROL TECHNIQUES
This section discusses the assumptions made regarding
control techniques applied to reduce HAP emissions from pulping,
bleaching, and pulping process condensates. The MACT emission
control technologies have design criteria and operating
parameters (e.g., combustion control device temperature and
residence time) that were determined for proposal. Based on
comments and subsequent evaluation, the Agency revised some of
the assumptions previously presented. Section 20.3.1 discusses
vent controls and section 20.3.2 discusses pulping process
condensate controls. Section 20.3.1 also presents the theory and
assumptions behind the clean condensate alternative, an
alternative emission control strategy that was not described in
the proposal BID.
20-12
-------
TABLE 20-2. EMISSIONS FROM AFFECTED SOURCES FOR AN EXAMPLE MILL
(1,000 tons of oven-dried pulp per day kraft mill)
Affected Sourcesa
Digester and evaporator^
Knotters
Screens
Pulp washing
Rotary vacuum drum washers
LOW air flow washers
Deckers
Oxygen delignification
Weak black liquor storage
Pulping wastewater
Total - pulping area
Bleaching system6' f
Bleaching wastewater
Total - bleaching area
Mill total
Existing controls
99.9% (combustion)
None
None
None
None
None
None
None
Biotreatment
Scrubber
None
Low Recycle "Open"
Mill (Mg/yr)b
HAPC
2.2
2.6
73
81
20
20
66
12
115
390
65
64
130
520
High Recycle "Closed"
Mill (Mg/yr)b
HAPC
2.2
0.48
1.3
280
20
31
225
12
115
690
46
24
70
760
aSystems listed are assumed to exist at the example mill.
^Emission factors taken from Chemical Pulping Emission Factor Development Document (A-92-40, IV-A-E).
Emissions (Mg/yr) = emission factor (Ib/ODTP) * capacity (1000 ton/day) * (350 day/yr) *
(1 Mg/1.1 ton) * (1 ton/2000 Ib) .
^Total HAP is calculated by summing emissions from all HAP species.
^Combustion control efficiency of 99.9 percent assumed in the emissions presented.
eNo uncontrolled emission factor available. All data is post scrubber. Emissions reflect the presence
of a scrubber.
^Assumed no hypochlorite stage and no chlorine dioxide substitution for the open mill. Assumed
100 percent substitution for the closed mill.
20-13
-------
TABLE 20-3. SUMMARY OF ESTIMATED NATIONAL BASELINE EMISSIONS3
Major Pollutants
HAP VOC
Total HAP
Total VOC
TRS
Selected Compounds
a-pinene V
methanol / /
terpenes /
b-pinene V
dimethyl sulfide V
p-cymene V
dimethyl disulfide V
hydrogen sulfide
methyl mercaptan •/ •/
o-cresol V V
acetaldehyde V" -f
cumene V" V
chloroform V -f
methyl ethyl ketone •f •f
1,2,4-trichlorobenzene / V
ethyl benzene / /
formaldehyde V V
phenol / V"
carbon tetrachloride / /
o-xylene / •f
1, 1, 1-trichloroethane / V
methylene chloride V •f
propionaldehyde / /
chlorine V
aSummary of emissions from chemical and semi-chemical pulping
largest constituents of the total emissions (plus chlorine) ,
Baseline Emissions (Mg/yr)
TRS Current After OW Option A
209,000
826,000
145,000
267, 000
139,000
134, 000
84,500
V" 64,100
31, 800
/ 20,600
V 43,900
/ 16,300
9, 800
8,140
7,520
10,800
4,230
3,340
4,590
2,800
2, 790
2,500
2,200
350
1,020
1, 850
120
, bleaching, and condensate operations.
sorted by descending baseline emissions
198,000
814,000
142,000
253,000
135,000
134,000
78,600
75,300
31,800
24,300
23,500
19,400
10,600
8,940
8,030
4,020
4,700
3,750
2,930
3,160
2, 720
2,740
1,430
1,350
1,200
1,170
25
Selected
after OW
After OW Option B
232,000
872,000
144,000
257, 000
159,000
142,000
85, 600
72,800
31,800
28,300
21,100
22, 200
14,600
11, 100
7, 980
4,020
5,640
4,940
2,860
3, 890
3,340
3, 040
1,430
1,390
1, 220
1,390
24
compounds are the 23
Option B.
20-14
-------
20.3.1 Applicable Control Techniques for Vents
This section discusses the applicable control technologies
for pulping and bleaching process vents. The control
technologies can be categorized into two types of control, add-on
control devices applied to an emission point or condensate stream
to reduce HAP, and process modifications or substitutions that
affect the formation or generation of HAP compounds. Vent
control devices typically found in the industry include
combustion control devices (i.e., lime kilns, power boilers,
recovery furnaces, thermal oxidizers, and flares), scrubbers, and
condensers. Pulping process modifications include extended
cooking, oxygen delignification, use of pressure screens, and
improved pulp washing. Bleaching process modifications include
chlorine dioxide substitution, elimination of hypochlorite, using
oxygen or peroxide in extraction stages, split chlorine addition
and ozonation.
All of these technologies are described in the proposal BID
(and are not discussed in this document), but the approach for
estimating the effect of process modifications and the removal
efficiencies of scrubbing have been updated based on the Agency's
evaluation of public comments and additional data. These changes
are discussed below.
20.3.1.1 Process Modifications. Process modifications
affect the generation of HAP compounds, and therefore, the amount
of HAP's that can be emitted. Process modifications are
accounted for through assigning different emission factors to a
facility; one emission factor represents conditions before the
process modification and another emission factor represents -,-IQO
conditions following the modification (i.e., before and after
using pressure screens, improved washers, 100 percent
substitution, elimination of hypochlorite). The environmental
impact of the process modifications is estimated as the
20-15
-------
difference in emissions obtained from applying the two emission
factors.
The emission factors used at proposal were based on a
collection of emission points for individual pieces of equipment.
In the mill system analysis for the final rule, process
modifications are still taken into account through separate
emission factors, but the impact of process modifications is
estimated on a mill system basis rather than on an equipment-
specific basis. The development of mill system emission factors
that vary based on process modifications is detailed in the
revised Chemical Pulping Emission Factor Development Document
(A-92-40, IV-A-8).
20.3.1.2 Removal Efficiencies for Gas Scrubber. Commenters
stated that the non-chlorinated HAP removal efficiency for bleach
plant vent scrubbers was overstated, especially for methanol.
The commenters stated that the scrubbers in the bleach plant were
designed and operated only for removal of chlorinated compounds,
primarily chlorine and chlorine dioxide. Industry testing of air
vent scrubbers at bleach plants (A-92-40, II-I-24) , supported
99 percent removal of chlorinated HAP (measured as chlorine) and
no reduction of non-chlorinated compounds. The EPA agreed with
the commenters that bleaching systems achieve at least 99 percent
control of chlorinated HAP's, but do not reduce non-chlorinated
HAP's. Detailed responses to comments on this topic are
presented in chapters 4.0 and 6.0 of this document. The final
impacts analysis reflects this updated information.
20.3.1.3 Clean Condensate Alternative. This section
briefly describes the conceptual basis for the CCA, a pollution
prevention option allowed in the final rule for compliance with
the kraft mill air standards for HVLC system vents specified in
63.443. The CCA compliance option was not included in the
proposal.
20-16
-------
The concept behind the CCA is that a portion of the HAP
emissions from a process unit that receives recycled or reused
process water is attributable to the HAP concentration in that
water. By reducing the HAP concentration in water that is used
in open or vented process equipment, less HAP will be available
to be volatilized to the atmosphere.
The EPA's evaluation of emissions from pulp washing systems
supports this concept since the emission factor for pulp washing
systems that received process water with a relatively high HAP
concentration was greater than the emission factor developed for
pulp washing systems that used low HAP concentration process
water (A-92-40, item IV-A-8).
The industry submitted a CCA preliminary engineering study
to EPA (A-92-40, items IV-D1-59 and 92). The control technique
presented in the study is based on biological treatment.
Although no mills currently have this technology in place, the
industry speculates that the CCA is capable of reducing the HAP
concentration in process waters down to the 100 ppmw range.
Industry asserts that the CCA can achieve HAP reductions equal to
or greater than would have been achieved by implementing the MACT
add-on controls on the HVLC system vents. The emission
reductions would come from process equipment in the HVLC system
(e.g., pulp washing system) as well as other process areas that
are not addressed by the MACT standard (e.g., causticizing
system).
20.3.2 Applicable Control Technologies for Pulping Process
Condensates
This section addresses the technical changes to the
applicable control technologies for pulping process condensate
emission points. The use of steam stripping systems and
biological treatment systems (combined with hard-piping) are the
two pulping process condensate control technologies used by the
pulp and paper industry. In addition, volume reduction options
20-17
-------
for condensate streams to be controlled have been added to the
analysis since proposal.
20.3.2.1 Steam Stripping Systems. The HAP removal
efficiency has been revised to reflect the comments and operating
data received following proposal (A-92-40, IV-B-10). Based on
analyses of the comments and performance data, the removal
efficiency for total HAP and methanol has been increased from 90
to 92 percent. The EPA has also determined that the hydrophilic
properties of methanol relative to the other HAP compounds
indicate that a 92 percent removal of methanol constitutes at
least 92 percent removal of total HAP.
Mass removal and steam stripper system outlet concentration
options were determined based on the same set of performance data
used to develop the 92 percent removal of total HAP (methanol) .
Therefore, the following control options are found to be
equivalent to the 92 percent removal achieved by steam stripping:
for bleached pulp mills, pulping process condensates treated to
remove 4.6 kilograms or more of total HAP per megagram of ODP or
achieving a total HAP outlet concentration of 330 ppmw from the
steam stripper system; and for unbleached pulp mills, pulping
process condensates treated to remove 2.9 kg/Mg ODP or more or
achieving a total HAP outlet concentration of 210 ppmw from the
steam stripper system.
20.3.2.2 Biological Treatment Systems. In the proposed
rule, a biological treatment system that achieved 90 percent
reduction in total HAP was specified as one of the control
options for pulping process condensates. A closed-collection
system had to be used to convey the pulping process condensates
to the biological treatment system (i.e., hard-piping).
In the final rule, biological treatment systems are retained
as a compliance option. However, the total HAP destruction
efficiency has been increased from 90 percent to 92 percent to
reflect the revisions made to the steam stripper system
20-18
-------
performance requirements. Additionally, the closed-collection
system requirements for tanks, containers, surface impoundments,
and drain system in the proposed rule were revised. In the final
rule, the individual drain systems must meet the requirements
specified in referenced §§ 63.960, 63.961, 63.962, and 63.964 of
subpart RR of part 63.
20.3.2.3 Condensate Segregation. The final rule contains
provisions for allowing mills the option of minimizing the volume
of digester, turpentine recovery, and evaporator system
condensates sent to treatment in the steam stripping or
biological treatment systems. Condensate segregation is
typically achieved using multistage condensation techniques on
the vent stream gases or vapors. Industry commented that mills
would perform condensate segregation in order to generate a low
volume, high HAP concentration stream that would be sent to
treatment and a high volume, low HAP concentration stream that
could be sent to the mill's hot water tank for distribution to
other process areas (e.g., pulp washing system). This practice
will reduce the energy cost associated with steam stripping (more
concentrated, lower volume stream sent to treatment) and reduce
the demand for fresh water in the mill process.
Based on industry data received since proposal, the mills
that use this practice can achieve a 65/35 percent mass split
(A-92-40, item IV-B-24). This means that 65 percent of the total
HAP mass is contained in the LVHC stream. In addition to
achieving the percent mass split, the final rule contains an
option for achieving the segregation option requirements based on
sending a minimum HAP mass to treatment from the segregated
digester, turpentine recovery, and evaporator system condensates
and the LVHC and HVLC collection system condensates.
20.4 CONTROL OPTIONS AND ENVIRONMENTAL IMPACTS
Section 20.4.1 presents the control options that were
analyzed to estimate national impacts of the final rule.
20-19
-------
Section 20.4.2 discusses emissions reductions and secondary
environmental impacts.
20.4.1 Control Options
This section presents the options analyzed to estimate the
national impacts of the final rule. The rationale for choosing
the MACT floor or going beyond the floor is discussed in the
preambles to the proposed and final rules, and in chapter 4.0 of
this document. The final rule specifies a MACT technology to be
used to control emissions. The final rule also allows mills to
use other control technologies, including control devices and
process modifications or chemical substitutions, if they achieve
equivalent control to the MACT technology. For purposes of
estimating costs and environmental impacts, the Agency selected
control technologies that would represent how mills could comply
with the final rule. Table 20-4 presents the options considered
for existing pulping sources in the national impacts analysis.
Table 20-5 presents the options considered for new pulping
sources. The first option shows the option selected for the
final rule. Additional options above the sources covered by the
final rule, in order of cost-effectiveness, are also presented.
Subsequent tables only present costs and impacts for the selected
option.
For bleaching systems, the MACT floor is control of
chlorinated HAP's (by 99 percent) using a caustic scrubber.
Information supplied by commenters to the proposed rule and
industry survey responses indicate that all bleach plants use
scrubbing technologies to reduce chlorinated HAP emissions. As
stated in the proposal preamble, EPA analyzed more stringent
controls, such as combustion of scrubber vent gases or combustion
of bleaching vent gases followed by a scrubber. These more
stringent options were determined to be unreasonable considering
20-20
-------
TABLE 20-4
CONTROL OPTIONS FOR EXISTING PULPING SOURCES
Subcategory
Option
Equipment System
Kraft
1 Pulp washing, oxygen delignification, high
emitting deckers, high emitting knotters and
screens, and steam stripping condensates
2 Option 1 systems, and weak liquor storage tanks
3 Option 2 systems, and low emitting deckers
4 Option 3 systems, and low emitting knotters and
screens
Sulf ite
Semi -chemical
Soda
1
2
1
2
1
2
Digester, evaporator, end stock washer
Option 1 systems, and weak liquor tanks and
strong liquor tanks
Digesters and evaporators
Option 1 systems, and pulp washing
Digesters and evaporators
Option 1 systems, and pulp washing
TABLE 20-5. CONTROL OPTIONS FOR NEW PULPING SOURCES
Subcategory
Kraft
Sulf ite
Semi -chemical
Soda
Option
1
2
1
2
1
2
1
2
Equipment System
Pulp washing, oxygen delignif ication, all
deckers, all knotter and screens, weak liquor
storage tanks, and steam stripping condensates
Option 1 systems, and pulp storage tanks
Digester, evaporator, red stock washer
Option 1 systems, and weak liquor tanks and
strong liquor tanks
Digesters and evaporators, pulp washing system
Option 1 systems, and pulp storage tanks
Digesters and Evaporators, pulp washing systems
Option 1 systems, and pulp storage tanks
20-21
-------
the cost and environmental impacts. The EPA's baseline analysis
of air impacts on the bleaching system are after OWs BAT options
are implemented (see section 20.2.2). No additional control
technologies or options in the bleaching systems were identified.
20.4.2 Environmental Impacts
This section discusses the methodology used to estimate
national air, water, energy, and other environmental impacts of
the final rule. For the final rule, impacts were estimated for
each individual mill and summed to provide the national estimate.
Mill-by-mill variations in costs and impacts are a function of
mill-specific design, equipment, and operating parameters, which
are based on the site specific mill data obtained from the 1992
voluntary MACT survey and the OW survey, and updated from
comments and information provided in response to the proposed
rule. Section 20.6 (Data Base System for Estimating National
Impacts) presents the procedure for assigning default process
operations and equipment to mills where site specific information
was unavailable.
20.4.2.1 Primary Air Impacts The primary air impacts
include the reduction of HAP, VOC, and TRS emissions directly
attributed to applying the control options. Emission reductions
for kraft, soda, and semi-chemical pulping vents are calculated
by applying the reduction efficiency of combustion devices
(98 percent) to the baseline emissions for systems not already
controlling emissions using a combustion device. Emission
reductions for sulfite mills are estimated based on the reduction
from baseline necessary to meet the sulfite pulping emission
limits.
20-22
-------
Table 20-6 presents primary air impacts for an example mill.
Table 20-7 presents the national primary air impacts (baseline
and emission reductions) of MACT controls, by mill type and mill
area (pulping vents, pulping wastewater, bleaching vents, and
bleaching wastewater) for methanol, total HAP, total VOC, and TRS
after OW s BAT Option A has been applied. Table 20-8 presents
similar information after OWs BAT Option B has been applied.
20.4.2.2 Energy Impacts. Additional energy is required for
the control of vent streams and condensate streams. This energy
may take the form of electricity, steam, or fuel. Table 20-9
lists the areas of the mills where energy is consumed to meet the
MACT standard.
For pulping vent streams, the amount of electricity required
to operate equipment (e.g., fans, pumps) and auxiliary fuel
needed to combust vent streams were determined from algorithms
contained in the OCCM (A-92-40, II-A-4), in the proposal BID
(A-92-40, II-A-35), and in supporting memoranda (A-92-40,
IV-B-13, IV-B-28). The amount of electricity required to operate
fans or pumps is estimated from the horsepower required to
provide motive force to transport vent and condensate streams to
control devices. Electricity demand was assumed to be met using
off-site power generation facilities. Electricity demand was
converted to equivalent fuel requirements assuming off-site power
When a combustion device is used to control HAP emissions,
auxiliary fuel may be required to sustain combustion. At
proposal, vent streams were assumed to be combusted in existing
combustion devices to estimate the effect on fuel usage
requirements. Following proposal, EPA determined that some mills
could use existing devices for combusting vent streams, while
20-23
-------
TABLE 20-6. PRIMARY AIR IMPACTS FOR AN EXAMPLE MILL
(1,000 tons of oven-dried pulp per day kraft mill)
Baseline HAP
emissions after
OW Option A
(Mg/yr) a
2.2
0.48
1.3
Control
Option
None
Combustion
Combustion
Emission reduction
(percent)
HAP VOC TRS
None None None
98 98 98
Emission reduction
(Mg/yr)
HAP
None
0.47
1.3
VOC
None
0.60
1.6
TRS
None
0.11
0.31
Affected sources
Digester and evaporator
Knotters
Screens
Pulp washing
Rotary vacuum drum washers
LOW air flow washers
Deckers
Oxygen delignification
Weak black liquor storage
Pulping wastewater
Total - pulping area
Bleaching system
Bleaching wastewater
Total - bleaching area
Mill total
aBaseline HAP emissions assumes control of digester and evaporator vents and OW Option A process changes
(including no hypochlorite state and 100 percent C102 substitution in the bleaching system).
"The baseline HAP emissions of 115 Mg/yr corresponds to a mill using biological treatment with a HAP
FE equal to 6.3 percent.
280
20
31
225
12
115b
687
46
24
70
757
Combustion
Combustion
Combustion
Combustion
None
None
None
None
98
98
98
98
None
None
None
None
98
98
98
98
None
None
None
None
98
98
98
98
None
None
None
None
274
20
31
220
None
None
550
None
None
550
810
62
300
63
None
None
1200
None
None
1200
190
14
35
20
None
None
260
None
None
260
20-24
-------
TABLE 20-7. NATIONAL EMISSIONS AND REDUCTIONS FROM THE MACT,
AFTER APPLYING OW OPTION A (Mg/yr)a
Pollutant
INDUSTRY TOTAL
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
KRAFT
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
SEMICHEM
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
SODA
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
Pulping
Vents
Baseline Reductions
after OW from the
Option A MACT
165,537
778,444
140,807
110,036
420
6,926
157,235
765,226
139,969
103,631
399
6,802
2,197
4,942
838
1,564
3
29
678
1,903
-
597
1
1
124,382
392,901
75,493
88,600
267
5,193
118,524
386,176
75,493
83,590
250
5,168
864
909
0
607
0
0
596
1,402
-
541
1
1
Pulping Condensates
Baseline Reductions
after OW from the
Option A MACT
12,178 3,880
11,841 3,880
-
11,262 3,861
-
-
10,116 3,880
10,092 3,880
-
10,022 3,861
-
-
550
550
-
550
-
-
103
103
-
103
-
-
Bleaching Vents
Baseline Reductions
after OW from the
Option A MACT
6,689
10,699
1,647
5,016
500
580
5,824
9,688
1,647
4,643
404
469
-
-
-
-
-
-
57
72
-
49
2
2
Bleaching Condensates
Baseline Reductions
after OW from the
Option A MACT
13,554
13,335
-
8,585
3,104
3,601
12,976
12,735
-
8,074
3,090
3,584
-
-
-
-
-
-
97
97
-
86
8
9
Industry
Total
Baseline Reductions
after OW from the
Option A MACT
197,958
814,319
142,454
135,005
4,024
11,224
186,151
797,741
141,616
126,370
3, 893
10,855
2,747
5,492
838
2,220
3
29
935
2,176
-
835
11
12
128,262
396,782
75,493
92,461
267
5,193
122,404
390, 056
75,947
87,451
250
5,168
864
909
0
607
0
0
596
1,402
-
541
1
1
20-25
-------
Pollutant
TABLE 20-7. NATIONAL EMISSIONS AND REDUCTIONS FROM THE MACT,
AFTER APPLYING OW OPTION A (Mg/yr)a (Continued)
Pulping Vents
Pulping Condensates
Bleaching Vents
Bleaching Condensates
Industry Total
Baseline Reductions Baseline Reductions Baseline Reductions Baseline Reductions Baseline Reductions
after OW from the after OW from the after OW from the after OW from the after OW from the
Option A MACT Option A MACT Option A MACT Option A MACT Option A MACT
SULFITE
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
5,428
6,373
-
4,245
17
94
4,398
4,414
-
3,862
17
24
1,409
1,096
-
587
-
-
808
939
-
324
93
108
481
502
-
424
6
7
8,126
8,910
-
5,580
116
208
4,398
4,414
-
3,862
17
24
' A "-" in the column represents zero emissions, while a zero in the column represents a result less than 0.5 Mg/yr. Baseline
emissions represent the baseline emissions after implementing the OW Option A. Reductions represents the amount of emissions
reduction from applying the MACT requirements.
20-26
-------
TABLE 20-8. NATIONAL EMISSIONS AND REDUCTIONS FROM THE MACT,
AFTER APPLYING OW OPTION B (Mg/yr)a
Pollutant
INDUSTRY TOTAL
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
KRAFT
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
SEMICHEMICAL
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
SODA
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
Pulping
Vents
Baseline Reductions
after OW from the
Option B MACT
199,866
836,521
142, 822
133,665
420
9,274
191,312
822,712
141,984
127,031
398
9,100
2,197
4,942
838
1,564
3
29
929
2,495
-
825
2
51
157,576
449,050
77,468
114,790
267
7,379
151,718
442,325
77,468
107, 047
248
7,304
864
909
-
607
0
0
596
1,402
-
541
2
51
Pulping Condensates Bleaching Vents Bleaching Condensates Industry Total
Baseline Reductions Baseline Reductions Baseline Reductions Baseline
after OW from the after OW from the after OW from the after OW
Option B MACT Option B MACT Option B MACT Option B
12,179
11,842
-
11,263
-
-
10,117
10,093
-
10, 023
-
-
550
550
-
550
-
-
103
103
-
103
-
-
4,081 6,689
4,081 10,699
1,647
4,062 5,016
500
580
4,081 5,824
4,081 9,688
1,647
4,062 4,643
404
469
-
-
-
-
-
-
57
72
-
49
2
2
13,554
13,335
-
8,585
3,104
3,601
12,976
12,735
-
8,074
3,090
3,584
-
-
-
-
-
-
97
97
-
86
8
9
232,287
872,398
144,469
158,634
4,024
13,454
220,228
855,228
143,631
149,772
3,893
13,153
2,747
5,492
838
2,220
3
29
1,186
2,767
-
1,063
12
63
Reductions
from the
MACT
161, 657
453,117
77,468
116,120
267
7,379
155,799
446,406
77, 922
110,381
248
7,304
864
909
0
607
0
0
596
1,402
-
541
2
51
20-27
-------
TABLE 20-8. NATIONAL EMISSIONS AND REDUCTIONS FROM THE MACT,
AFTER APPLYING OW OPTION B (Mg/yr)a (Continued)
Pollutant
SULFITE
Total HAP
Total VOC
TRS
Methanol
Chloroform
Total Chlorinated HAP
Pulping
Baseline
after OW
Option B
5,428
6,373
-
4,245
17
94
Vents
Reductions
from the
MACT
4,398
4,414
-
3,862
17
24
Pulping Condensates
Baseline
after OW
Option B
1,409
1,096
-
587
-
-
Reductions
from the
MACT
-
-
-
-
-
-
Bleaching Vents
Baseline
after OW
Option B
808
939
-
324
93
108
Reductions
from the
MACT
-
-
-
-
-
-
Bleaching
Baseline
after OW
Option B
481
502
-
424
6
7
Condensates
Reductions
from the
MACT
-
-
-
-
-
-
Industry
Total
Baseline Reductions
after OW
Option B
8,126
8,910
-
5,580
116
209
from the
MACT
4,398
4,414
-
3,862
17
24
* A "-" in the column represents zero emissions, while a zero in the column represents a result less than 0.5 Mg/yr. Baseline
emissions represent the baseline emissions after implementing the OW Option B. Reductions represents the amount of emissions reduction
from applying the MACT requirements.
20-28
-------
TABLE 20-9.
SUMMARY OF TYPES OF ENERGY REQUIREMENTS OF MACT AND OW STANDARDS
(by process area)
Area
Location of
Energy Energy Provider
Source
Requiring
Energy
Description of
Source
Fuel
Pulping vents Auxiliary fuel On-site
Electricity
Off-site from
power grid
Combustion
device
Fans
Pumps
Incinerator
Existing boiler
Motive force for
vents from emission
Soint to control
evice
Motive force for
condensate stream
from condenser to
tanks
Methane
Mixture of fuels
Residual oil
Residual oil
Pulping
conaensates
Bleaching
System
Steam
Electricity
On-site from
boilers
Off-site from
power grid
Steam stripper
Fans
Used to remove
pollutants from
pulping process
conaensates stream
Motive force for
vents from emission
Mixture of fuels
Residual oil
Pumps
Chlorate
production
device
Motive force for
condensate stream
from condenser to
tanks
Chemical
manufacturing,
energy supplied to
the chemical
manufacturer by off-
site electrical
producer
Mixture of fuels
Residual oil
20-29
-------
others would need to construct a stand-alone incinerator
(A-92-40, IV-E-93) .
For stand-alone incinerators, EPA assumed that natural gas
would be used as the auxiliary fuel. For existing boilers, a
mixture of hog fuel (i.e., wood waste) (60 percent), oil
(10 percent), natural gas (10 percent), and coal (20 percent) was
assumed based on fuel usage information supplied by the pulp and
paper industry in responses to the OW survey (A-92-40, IV-B-28).
The fuel energy required to combust vent streams was calculated
in the incinerator and boiler design algorithms from the OCCM.
For boilers, the fuel energy was converted to the mass of hog
fuel, oil, natural gas, and coal needed using the fuel splits
presented above and the heating value of each fuel (4,500 Btu per
Ib of coal; 18,000 Btu per Ib of oil; 1,000 Btu per standard
cubic foot of natural gas; and 13,000 Btu per Ib of hog fuel).
(A-92-40, IV-J-78)
For kraft pulping condensates, increased steam is required
for stripping HAP-laden condensate streams. Steam demand was
converted to equivalent fuel requirements based on the same
composite of fuels used in the existing boiler assumption.
Table 20-10 presents energy impacts for an example mill.
Table 20-11 presents national energy requirements.
20.4.2.3 Secondary Air Impacts. The secondary air impacts
evaluated are the increases in criteria pollutant emissions ^^2>
CO NO ^' anc^ VOC) resulting from: (1) combustion of
compounds in vent streams and (2) increased burning of fuel used
as auxiliary fuel or for steam or electricity generation used for
powering equipment.
20-30
-------
TABLE 20-10. ENERGY IMPACTS FOR AN EXAMPLE MILLa
(1,000 tons of oven-dried pulp per day kraft mill)
Area
After OW Option A
MACT I electricity
MACT I auxiliary fuel
MACT I steam
MACT II electricity
OW on-site electricity
OW off-site electricity
OW steam
OW wastewater
TOTAL
After OW Option B
MACT I electricity
MACT I auxiliary fuel
MACT I steam
MACT II electricity
OW on-site electricity
OW off-site electricity
OW steam
OW wastewater
TOTAL
Energy
Pulping vents
57,745
20,319
--
2,610
10,941
--
(15,221)
--
76,394
66,582
23,577
--
2,610
60,487
--
(16,052)
--
137,204
requirements
Pulping
condensates
--
--
135,977
--
--
--
--
(28,584)
107,393
--
--
135,977
--
--
--
--
(26,614)
109,363
(Million Btu/yr)
Bleaching
vents Total
57,745
20,319
135,977
2,610
10,941
51,945 51,945
(15,221)
(28,584)
51,945 235,732
66,582
23,577
135,977
2,610
60,487
(95,622) (95,622)
(16,052)
(26,614)
(95,622) 150,945
aSources being controlled are defined in Table 20-2. Table shows requirements, so negative values (in
parenthesis) are net energy credits. A "--" in the table indicates no energy impacts for the given area.
20-31
-------
TABLE 20-11.
NATIONAL ENERGY IMPACTS AFTER MACT AND OW OPTIONS
(Million Btu/yr)a
Energy demand
description Pulping vents Pulping condensates Bleaching vents
After MACT and OW Option A
Electricity
Auxiliary fuel
Steam
Total
After MACT and OW Option B
Electricity
Auxiliary fuel
Steam
Total
9
2
(2,
10
18
3
(2,
19
,830
,989
161,
,658
,162
,466
279,
,350
,643 (4,058,136) 7,374,715
,103
000) 19,112,019
,746 15,053,883 7,374,715
,959 (3,788,493) (13,575,681)
,194
000) 19,112,019
,153 15,333,526 (13,575,681)
Total
13,
2,
16,
33,
3,
16,
21,
147,
989,
957,
087,
808,
466,
833,
107,
222
123
193
344
785
214
019
998
aEnergy impacts include those associated with applying the MACT (I, II, and III) and with applying the OW
options. The table reports impacts. Negative results, shown in parenthesis, are therefore net energy
credits. A "--" in the table indicates no energy impacts for the
given area.
20-32
-------
The combustion of vent gases results in secondary emissions
of sulfur dioxide. The proposal BID presented secondary
emissions of sulfur dioxide based on combustion of TRS compounds
only (i.e., hydrogen sulfide, dimethyl disulfide, dimethyl
sulfide, and methyl mercaptan). For the final rule, secondary
emissions of sulfur dioxide were estimated by assuming complete
stoichiometric combustion of all sulfur containing compounds
(i.e., the TRS compounds, and carbon disulfide and carbonyl
sulfide) to the combustion end products of water, C1C>2, and
sulfur dioxide. For example, 1 kilogram of hydrogen sulfide
oxidizes to form 1.88 kilograms of sulfur dioxide. The new
methodology increases the sulfur dioxide emissions calculated for
the final rule.
The impacts calculated for the final rule may over-estimate
sulfur dioxide emissions because mills may be able to use
existing sulfur dioxide controls to reduce sulfur dioxide
emissions. However, EPA does not have sufficient information on
the number and effectiveness of these controls, so no reductions
were taken. Also, mills may use other control options that may
not increase sulfur dioxide emissions, such as the clean
condensate alternative or low-emitting equipment.
Criteria pollutants are also emitted from fuel used to
generate electricity and steam, and from the burning of auxiliary
fuel to combust vent streams. Areas where energy is consumed are
presented in table 20-9. Criteria pollutant emissions were
calculated from the amount of fuel required (as discussed in
section 20.4.2.2) and criteria pollutant emission factors
(usually in pound of pollutant per ton or gallon of fuel)
presented in previous EPA studies for combustion of each fuel.
(A-92-40, IV-J-77)
20-33
-------
Scrubbing of bleaching system vent streams was assumed to
generate no secondary air emissions because all bleached mills
are assumed to be already operating a scrubber, and no additional
control techniques were applied to the bleaching system.
Table 20-12 presents secondary air impacts for an example
mill. Tables 20-13 and 20-14 present national secondary air
emissions from applying the MACT requirements and OW Option A and
Option B, respectively, to the current baseline.
20.4.2.4 Water and Other Impacts. No significant revisions
were made to the assumptions or conclusions regarding water
impacts and other impacts (i.e., noise, visual, odor, and solid
waste).
20.5 ESTIMATED CONTROL COSTS
This section presents the national cost of the final rule
and the changes made to the costing methodology. Section 20.5.1
discusses the assumptions used for sizing and estimating the
costs of control technologies; section 20.5.2 presents estimated
national costs.
20.5.1 Control Cost Methodology.
The national costs are estimated by calculating the cost of
each control option applicable to each mill and summing the mill-
specific results to obtain a national total.
The OCCM was used to size and cost equipment in the proposed
and final rules. In general, most of the inputs to the OCCM
design and cost algorithms did not change from proposal. Some of
the global changes were to the interest rate used to estimate
capital recovery (7 percent was used in the final rule instead of
10 percent) and labor and utility rates were updated.
The EPA has assumed some different control technologies and
equipment for the final rule, which required a revision of the
costs. These changes are discussed below.
20-34
-------
TABLE 20-12. SECONDARY AIR IMPACTS FOR AN EXAMPLE MILL (Mg/yr)a
(1,000 tons of oven-dried pulp per day kraft mill)
CO
Area; After OW Option A
Pulping vents
Pulping condensates
Bleaching
Total
Area; After OW Option B
Pulping vents
Pulping condensates
Bleaching
Total
MACT I electric
MACT I fueld
MACT II electric
OW electric on-site
OW steam
Vent combustion6
MACT I steam
OW wastewater
OW electric off-site
MACT I electric
MACT I fueld
MACT II electric
OW electric on-site
OW steam
Vent combustion6
MACT I steam
OW wastewater
OW electric off-site
0
8
0
0
(6
--
57
(0
0
61
57
10
0
0
(6
--
1
(0
(1
61
.90
.62
.04
.17
.46)
.69
.44)
.80
.69
.00
.04
.94
.81)
.02
.41)
.48)
NOX
9
3
0
1
(2
--
20
(4
8
37
20
3
0
10
(2
--
11
(4
(16
23
PMb
.78
.03
.44
.85
.27)
.28
.84)
.80
.28
.51
.44
.24
.40)
.28
-51)
.19)
0.
0,
0.
0.
(0,
--
0.
(0,
0.
0
0,
0,
0.
0 ,
(0.
--
0.
(0.
(0.
0.
.01
.08
.00
.00
.06)
.56
.00)
.00
.59
.56
.09
.00
.00
.06)
.01
.00)
.01)
.59
S02
19
7
0
3
(5
588
49
(9
17
672
49
8
0
20
(5
604
22
(9
(32
659
.54
.39
.88
.71
.53)
.77
.43
.68)
.57
.43
.57
.88
.46
.83)
.16
.52
.00)
-34)
VOCC
0.
0
0
0
(0,
20.
0.
(0,
0.
21
0,
0.
0.
0.
(0.
23 .
0.
(0.
(0.
24
.05
.08
.00
.01
.06)
.77
.54
.03)
.05
.54
.09
.00
.06
,06)
.76
.06
.02)
08)
aSources being controlled are defined in Table 20-2. Table shows emissions, so negative values, (in parenthesis)
indicate reductions. A "--" in the table represents no secondary air impacts, while a "0.00" in the table indicates
less than 0.005 Mg/yr.
"PM emissions are assumed to be reduced by 90 percent using existing electrostatic precipitators on existing combustion
devices.
CVOC generated from incomplete combustion.
"^Calculated using EPA derived emission factors for fuel types (A-92-40, IV-J-77), fuel splits from the OW industry
survey, and energy in table 20-10.
eCalculated assuming stoichiometric conversion of all sulfur containing compounds to S02, water, and C02-
20-35
-------
TABLE 20-13. SECONDARY AIR EMISSIONS AND ENERGY IMPACTS FROM APPLYING
THE MACT AND OW OPTION A TO THE CURRENT BASELINE3
Secondary Pollution
Impacts
KRAFT
Energy (106 Btu/yr)
S02 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mg/yr)
SEMICHEM
Energy (106 Btu/yr)
S02 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mg/yr)
SODA
Energy (106 Btu/yr)
S02 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mg/yr)
SULFITE
Energy (106 Btu/yr)
SO2 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mg/yr)
Pulping
Vents
10,731,190
86,397
1,803
458
2,775
4
(296,658)
(103)
(50)
20
18
0
55,124
20
8
19
25
0
169,090
57
29
3
0
0
Pulping Bleaching Bleaching
Compensates Vents Condensates
15,094,464 7,300,968
5,593 2,469
2,170 1,237
8,047 113
228 7
79 0 -
-
-
-
-
-
-
(40,581) 73,747
(14) 25
(7) 12
(1) 11
(0) 0
(0) 0
-
-
-
-
-
-
Kraft Total
33,126,622
94,459
5,210
8,607
3,010
83
(296,658)
(103)
(50)
20
18
0
88,290
31
14
29
25
0
169,090
57
29
3
0
0
20-36
-------
TABLE 20-13. SECONDARY AIR EMISSIONS AND ENERGY IMPACTS FROM APPLYING
THE MACT AND OW OPTION A TO THE CURRENT BASELINE (Mg/yr)a (Continued)
Secondary Pollution
Impacts
INDUSTRY TOTAL
Energy (106 Btu/yr)
SO2 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mq/vr)
Pulping
Vents
10,658,746
86,371
1,791
499
2,818
4
Pulping
Condensates
15,053,883
5,579
2, 163
8,046
228
79
Bleaching Bleaching
Vents Condensates
7,374,715
2,494
1,249
124
7
0
Kraft Total
33,087,343
94,444
5,203
8,659
3,053
83
a Results shown are emissions and impacts resulting from applying OW Option A and the MACT requirements to the current
baseline. Results are emissions or impacts, therefore negative values indicate a reduction of emission or a net
energy credit. A "-" in the table represents a value of zero, while a zero in the table represents a result less than
0.5 Mg/yr (or 0.5 MMBtu/yr).
20-37
-------
TABLE 20-14. SECONDARY AIR EMISSIONS AND ENERGY IMPACTS FROM APPLYING
THE MACT AND OW OPTION B TO THE CURRENT BASELINE51
Secondary Pollution
Impacts
KRAFT
Energy (106 Btu/yr)
S02 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mg/yr)
SEMICHEM
Energy (106 Btu/yr)
S02 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mg/yr)
SODA
Energy (106 Btu/yr)
S02 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mg/yr)
SULFITE
Energy (106 Btu/yr)
SO2 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mg/yr)
Pulping
Vents
19,284,308
91,460
3,243
731
3,204
6
(261,242)
(90)
(44)
24
18
0
132,123
46
21
21
25
0
194,964
66
33
3
0
0
Pulping
Condensates
15,371,311
5,688
2,216
8,041
229
79
-
-
-
-
-
-
(37, 785)
(13)
(6)
(1)
(0)
(0)
-
-
-
-
-
-
Bleaching Bleaching
Vents Condensates
(13,439,924)
(4,546)
(2,276)
(208)
(12)
(1)
-
-
-
-
-
-
(135,757)
(46)
(23)
(2)
(1)
(0)
-
-
-
-
-
-
Kraft Total
21,215,695
92,602
3,183
8,564
3,421
84
(261,242)
(90)
(44)
24
18
0
(41,419)
(13)
(9)
18
24
0
194,964
66
33
3
0
0
20-38
-------
TABLE 20-14. SECONDARY AIR EMISSIONS AND ENERGY IMPACTS FROM APPLYING
THE MACT AND OW OPTION B TO THE CURRENT BASELINE (Mg/yr)a (Continued)
Secondary Pollution
Impacts
INDUSTRY TOTAL
Energy (106 Btu/yr)
S02 (Mg/yr)
NOx (Mg/yr)
CO (Mg/yr)
VOC (Mg/yr)
PM (Mq/yr)
Pulping Pulping
Vents Condensates
19,350,153
91,481
3,253
779
3,247
6
15,333,526
5,675
2,210
8,040
229
79
Bleaching Bleaching
Vents Condensates
(13,575,681)
(4,592)
(2,299)
(210)
(13)
(1)
Kraft Total
21, 107, 998
92,564
3,164
8,610
3,463
84
a Results shown are emissions and impacts resulting from applying OW Option B and the MACT requirements to the current
baseline. Results are emissions or impacts, therefore negative values indicate a reduction of emission or a net energy
credit. A "-" in the table represents a value of zero, while a zero in the table represents a result less than
0.5 Mg/yr (or 0.5 million Btu/yr).
20-39
-------
20.5.1.1 Enclosures and Vent Gas Conveyance Systems. The
algorithms and assumptions used for estimating the cost of
equipment enclosures, as presented in the proposal BID, have not
been revised. However, assumptions regarding ductwork used for
the conveyance of vent streams from either discrete emission
points or from enclosures to the control devices have been
revised (A-92-40, IV-B-13). Specifically, the following design
assumptions affecting cost have been revised since proposal:
Based on comments received following proposal, the duct
length from the emission points and enclosures to
existing combustion devices has been increased from
1,000 feet to 1,500 feet.
At proposal, the equipment comprising the ductwork
system was assumed to include ductwork and elbows, fan,
knock-out drum(s), flame arrester (s), rupture discs,
supports, and insulation. Based on comments to the
proposed rule, EPA also included the following
additional equipment: condenser, condensate storage
tank(s), and sampling port(s).
Costs for ductwork at bleach plants were not evaluated
because mills are already controlling the chlorinated
vents. Analysis of non-chlorinated vents were dropped
as an option as discussed earlier.
20.5.1.2 Control Technology Costs for Vents. For the
thermal oxidizer system, heat recovery is a key variable
affecting capital costs. At proposal, the model mill that was
used to calculate costs was assumed to combust pulping vents in
an existing combustion device (e.g., lime kiln or power boiler).
Therefore, no heat recovery for a thermal oxidizer was assumed.
After proposal, EPA surveyed several mills on the capacity of
existing combustion devices to combust additional vent streams
(A-92-40, IV-E-85) . The results of the survey indicated that
20-40
-------
two-thirds may have the capacity in their existing combustion
devices, while one-third would construct a stand-alone
incinerator. This ratio was used to estimate national impacts.
Because EPA did not have sufficient information to assign the
control scenarios to each mill, costs were calculated assuming
one-third of the costs for controlling pulping systems were from
a stand-alone incinerator and two-thirds of the cost was from
routing vent streams to an existing boiler. A 95 percent heat
recovery was assumed in developing the final thermal oxidizer
system costs. Costs associated with a thermal oxidizer were
calculated using algorithms on the OCCM. These algorithms were
previously used to cost thermal oxidizers to control bleaching
system vents in the proposal BID.
20.5.1.3 Control Technology Costs for Pulping Process
Condensates. This section describes the steam stripper design
considerations. It also provides the design parameters affecting
cost and the general methodology used to develop capital and
annual costs for steam-stripping and for hard-piping condensate
streams to wastewater treatment systems.
Steam stripping costs. No revisions were made since
proposal to the general methodology used to develop capital and
annual costs for steam stripper systems, but some specific
revisions were made to the steam stripper design assumptions used
in the costing methodology. As discussed in section 20.3.2.1,
the removal efficiency of methanol was increased from 90 to 92
percent.
The volumetric flow rate of condensate sent to the stripper
system was also revised. Based on industry data submitted
following proposal (A-92-40, item IV-B-g) , the flow rate was
20-41
-------
decreased from approximately 1 GPM per ADTP to 0.20 GPM per ADTP.
The revised value is more accurately reflects the volumes of
condensate treated in the existing steam stripper systems.
The steam stripper column tray efficiency was reduced from
67 percent to 50 percent. This revision was made based on
industry comments indicating that a 50 percent efficiency more
accurately reflects the operation of steam strippers in the pulp
industry due to plugging of tray openings associated with the
fiber content of pulping process condensates.
Steam stripper cost credits were developed for the methanol
rectification process and the reduced amount of BOD sent to
biological treatment system because of the operation of the steam
stripper. The methanol rectification credit is based on costs
savings associated with replacing fossil fuels used for power
generation with the concentrated methanol condensates derived
from the steam stripper vent gases (A-92-40, IV-B-17) The
biological treatment system cost credit was developed based on
information submitted by industry following proposal IA-92-40,
IV-B-25).
Hard-piping to biological treatment system costs. The cost
of biological treatment was not estimated at proposal. Following
proposal, several companies submitted estimates for the costs
associated with hard-piping pulping process condensates to a
biological treatment system. These costs were normalized to a
dollar per ton of pulp produced basis ($1,230 per PDTP total
capital investment, $197 per ODTP total annual cost). The
normalized cost factor was then used to estimate the cost of
hard-piping for other mills. (A-92-40, IV-B-25)
20-42
-------
20.5.2 National Costs for the Control Options
Tables 20-15 and 20-16 present summaries of total capital
investment and total annual cost by mill type for the control
options chosen in the final rule (and options beyond the chosen
option) after OW Option A and B have been implemented.
Tables 20-15 and 20-16 present costs for controlling existing
sources only. A summary and discussion of total capital
investment and total annual costs for controlling new sources is
contained in the new source costing memorandum (A-92-40,
IV-B-100).
20.6 DATA BASE SYSTEM FOR ESTIMATING NATIONAL IMPACTS
This section summarizes the changes to the pulp and paper
NESHAP data base that is used to estimate national impacts of the
final rule. This section only presents changes to the original
data base discussed in chapter 6 of the proposal BID.
Section 20.6.1 discusses revisions to the data base.
Section 20.6.2 presents the calculation of impacts and
section 20.6.3 discusses revisions to the national impacts
estimation methodology and analyses performed to incorporate
effluent guidelines regulations, which are being promulgated
simultaneously with the NESHAP.
20.6.1 Data Base Revisions
20.6.1-1 Data Base Structure. At proposal, EPA developed
model pulping and bleaching process units to represent the
variety of emission points in a pulp mill. To estimate national
impacts, the models were assigned to each mill based on As
discussed in section 20.2.1, EPA used a number of sources to
develop a data base characterizing pulp and paper mills. When
information from the various sources conflicted, the following
20-43
-------
TABLE 20-15.
SUMMARY OF NATIONAL COST AND COST-EFFECTIVENESS OF MACT CONTROL OPTIONS AFTER
OW OPTION A (selected option is underlined)
Existing Source Control Option
KRAFT
(1) Washing system, OD, deckersc, knotters and
screensd, steam stripping combined with
hardpiping
(2) Option (1) equipment, plus WBLST
(3) Option (2) equipment, plus low emitting
deckers
(4) Option (3) equipment, plus low emitting
knotters and screens
SEMI -CHEMICAL
(1) Digesters and evaporators
(2) Digesters and evaporators, washing system
SODA
(1) Digesters and evaporators
(2) Digesters and evaporators, washing system
SULFITE
(1) Digester, evaporator, red stock washer
(2) Digester, evaporator, red stock washer,
weak liquor, strong liquor
Total (for selected options)
Total (for recordkeeping and reporting)
Grand Total
Capital
Investment
(106 $)
452
473
488
508
11
25
2.4
5.0
23
32
488
8.3
496
Annual
Cost
(10s $/yr)
118
125
130
138
1 . 0
4. 8
0.2
1.3
4.8
6.4
124
6.9
130
Baseline
Emissions3
(103 Mg/yr)
186
186
186
186
2.9
2.9
1.1
1.1
7.7
7.7
198
--
198
Emission
Reduction"
(103
Mg/yr)
123
125
125
126
0.86
1.6
0.6
0.8
4.4
4.6
128
--
128
Emission
Reduction
(percent)
66%
67%
67%
68%
30%
55%
55%
73%
57%
60%
65%
--
65%
Average
Cost
Effectiveness
($/Mg)
959
1,000
1, 038
1,099
1,215
3,000
333
1,625
1,095
1,396
963
--
1,016
Incremental
Cost
Effectiveness
($/Mg)
-_
3,500
10, 833
22,438
--
5, 075
--
5,500
--
8,000
--
--
alndustry total baseline emissions from affected sources, assuming OW Option A has been implemented.
'-'Emission reductions reflects the change from the post-OW Option A baseline to the post-MACT residual emissions.
cHigh-emitting deckers are controlled.
dHigh-emitting knotters and screens are controlled.
20-44
-------
TABLE 20-16.
SUMMARY OF NATIONAL COST AND COST-EFFECTIVENESS OF MACT CONTROL OPTIONS
AFTER OW OPTION B (selected option is underlined)
Existing Source Control Option
KRAFT
(1) Washing system, OD, deckers0, knotters and
screens'1, steam, stripping combined with hardpiping
(2) Option (1) equipment, plus WBLST
(3) Option (2) equipment, plus low emitting
knotters and screens
SEMICHEMICAL
(1) Digesters and evaporators
(2) Digesters and evaporators, washing system
SODA
(1) Digesters and evaporators
(2) Digesters and evaporators, washing system
SULFITE
(1) Digester, evaporator, red stock washer
(2) Digester, evaporator, red stock washer, weak
liquor, strong liquor
Total (for selected options)
Total (for recordkeeping and reporting)
Capital
Investment
(106 $)
588
609
629
11
25
2.4
5.0
23
32
624
8.3
Annual
Cost
(106 $/yr)
149
156
164
1.0
4.8
0.2
1.3
4.8
6.4
155
6.9
Baseline
Emissions*
(103 Mg/yr)
220
220
220
2.9
2.9
1.4
1.4
7.7
7.7
232
--
Emission
Reduction13
(103 Mg/yr)
156
158
158
0.86
1.6
0.6
0.8
4.4
4.6
162
--
Emission
Reduction
(percent)
71%
72%
72%
30%
55%
43%
57%
57%
60%
70%
--
Average
Cost
Effectiveness
($/Mg)
955
987
1, 036
1,215
3,000
333
1,625
1,095
1,396
958
--
Incremental
Cost
Effectiveness
($/Mg)
--
3,500
22,438
--
5,075
--
5,500
--
8, 000
--
--
Grand Total
632
162
232
162
70%
1,001
alndustry total baseline emissions from affected sources, assuming OW Option B has been implemented.
^Emission reductions reflects the change from the post-OW Option B baseline to the post-MACT residual emissions.
cHigh-emitting deckers are controlled. After OW Option B, all deckers are assumed to be high-emitting deckers because of condensate recirculation
associated with adding an oxygen delignification unit.
^High-emitting knotters and screens are controlled.
20-45
-------
hierarchy was used. The OW survey was determined to be the most
current source of information, followed by the comments and
industry information supplied after proposal, the latest edition
of the Lockwood Post Directory, site visits, and the 1992 NCASI
survey.
Secondary emissions from control devices were also re-
evaluated as discussed in section 20.4. The major changes
include assuming all sulfur containing compounds were combusted;
and emissions would be controlled using an existing boiler or
incinerator, so secondary impacts were calculated from the fuel
burned.
20.6.1.2 Default Values. At proposal, the data base
structure was based on model process units. Since each mill was
assigned a specific pulping and bleaching model, default values
were not necessary to fill in data gaps for each facility.
Following proposal, the data base was revised to estimate
national impacts based on the actual equipment systems at each
facility. The data base contains complete information on mill
production capacity, bleaching sequences, the number of oxygen
delignification systems, etc.; however, complete pulping system
equipment and control information was not available for every
mill. In the absence of mill-specific data, default information
was used. A detailed discussion of the defaults is presented in
a supporting memorandum (A-92-40, IV-B-28).
In general, information on the vent stream characteristics
(e.g., emissions, vent gas flow rate, temperature, and moisture
content) for each mill was not available. Therefore, average
characteristics were developed from test data. The average
emission factors are documented in the revised Chemical Pulping
20-46
-------
Emission Factor Development Document (A-92-40, IV-A-8), and the
average vent stream characteristics are detailed in a separate
memorandum (A-92-40, IV-B-28).
If no emission controls were reported in the surveys or in
other data then generally none were assigned. The exceptions
were the control status of the LVHC system and assignment of
enclosures to pulp washers. Since a NSPS exists requiring LVHC
vent streams at kraft mills to be controlled, the assumption was
made that all LVHC systems at kraft mills were already
controlled. As documented in the proposal BID, pulp washers
constructed after 1978 were assumed to be enclosed.
Information was not available to describe the pulping
condensate characteristics (e.g., flow rate, HAP concentration,
recycle/reuse patterns, etc.) at each mill. Condensate
characteristics, per unit capacity, were developed based on the
information contained in the NCASI condensate study (A-92-40,
IV-A-8). These characteristics were used for all pulping types.
20.6.2 Calculation of Impacts
For the final rule, national impacts were calculated using
methodologies discussed earlier in this subsection and in
sections 20.1 through 20.5. The structure of the data base
remained relatively unchanged since proposal (see figure 6-1 in
the proposal BID). The only major changes are that actual system
data was used rather than using model units, and default values
were used to fill in data gaps. Therefore, the industry
characterization did not include model mills.
Baseline emissions and emission reductions resulting from
control options were calculated for each equipment system and
were summed for each system, each mill, and for all mills
20-47
-------
combined. Control costs were calculated by mill area (pulping
vents, pulping wastewater, bleaching vents) and summed for the
total mill, instead of by equipment system since add-on controls
may be applied to multiple systems (i.e., multiple vents were
assumed to be routed to control devices through a common header).
As discussed in section 20.5, after proposal EPA determined
that for kraft, soda, and semi-chemical mills, some could use
existing combustion devices while others would need to construct
stand-alone incinerators to control vent streams. Because EPA
did not have sufficient information to assign the control
scenario to each mill, total costs were calculated by adding two-
thirds of the costs for routing to an existing device and one-
third of the cost for constructing a dedicated incinerator. The
control costs for sulfite mills were based on reducing the
temperature of the vent gas streams before they are routed to the
acid plant scrubber or nuisance scrubber to increase the removal
efficiency.
For kraft pulping condensates, baseline emissions were
estimated based on the type of treatment process (i.e.,
biological treatment system or steam stripping) in place at
existing mills. For mills where the configuration of the
existing biological treatment systems were available, emissions
were based on the configuration of the equipment and the model
that calculates air emissions from wastewater systems. Emission
reductions were estimated based on hard-piping pulping process
condensates to well-operated biological treatment systems. Steam
stripping emission reductions were applied to mills with existing
biological treatment systems not arranged to meet the final rule
-------
without major reconstruction. For control purposes a new steam
stripper was costed.
For mills where no information was available regarding the
configuration of the existing biological treatment systems and
there was not existing steam stripper, the impacts were estimated
by applying average factors, on a ton of pulp production basis.
The average factors were calculated by dividing the total
impacts, estimated for the mills with sufficient configuration
data, by the total production for those mills.
20.6.3 Effects of OW Changes on Impacts
As discussed in section 20.2.3, OW is promulgating effluent
guidelines that change pulping and bleaching processes. The
baseline equipment configurations and assumptions were adjusted
to represent the mill after the two OW options where enacted.
The major adjustments that affect emissions and costing are:
Use of low water flow washers (emitting less air
emissions) at specific mills,
More recycling condensates of in the pulping mill which
increases air emissions for affected equipment,
Use of oxygen delignification (which increases the
number of emission sources to be controlled, and
increases the HAP concentration in process water
recycled into other systems), and
Changes in the use of chlorinated compounds in bleach
areas (which reduces chlorinated air emissions).
The effect the changes have on emissions and costs is presented
in supporting memoranda (A-92-40, IV-B-28). Impacts presented in
this section are after OW Options A and B have been implemented,
and differences in environmental and energy impacts is shown on
table 20-17.
20-49
-------
20.7 SUMMARY OF ENVIRONMENTAL AND COST IMPACTS OF THE MACT
STANDARDS
Table 20-17 also summarizes the total impacts of the MACT
standards after OWs effluent guidelines (Options A and B) have
been implemented. The table presents the current baseline
emissions, baseline emissions after OW Options A and B are
implemented, and the MACT II baseline emissions (see MACT II BID
for details on the MACT II estimates). The table also summarizes
emission reductions, energy consumption, secondary impact
generation, and costs resulting from applying MACT controls to
the MACT I and MACT II sources.
20-50
-------
TABLE 20-17.
TOTAL MILLS AND EMISSIONS; PULP AND PAPER NATIONWIDE
ENVIRONMENTAL AND COST IMPACTS
IMPACTS
MACT I & III Baseline
Emissions
(Before MACT Is Applied)a'c
Current
(for
affected
sources)
After OW
Opt. A
After OW
Opt. B
MACT II
Baseline
Emissions
(before
MACT is
applied)c
MACT I & III
Emissions (After MACT
is applied)b
-------
TABLE 20-17. TOTAL MILLS AND EMISSIONS; PULP AND PAPER NATIONWIDE
ENVIRONMENTAL AND COST IMPACTS (Continued)
IMPACTS
MACT I & III Baseline
Emissions
(Before MACT Is Applied) a'c
Current
(for
affected
sources)
After OW
Opt. A
After OW
Opt. B
MACT II
Baseline
Emissions
(before
MACT is
applied) c
CO - (840) (1,100) 248,400
NOX - 500 (1,820) 120, IOC
S02 - 860 (3,800) 102,600
MACT I & III
Emissions (After MACI
is applied) b/ c
With OW
Opt. A
With OW
Opt. B
MACT II
Emissions
(After
MACT is
applied) c
Emissions After
MACT I, II, & IIIC
With OW
Opt. A
With OW
Opt. B
8,660 8,610 190,7001 199,400 199,300
5,230 3,200 120,600 126,000 124,000
94,500 92,600 102,500 197,000 195,000
Energy (million Btu/Yr) :
- Electric - 4,870,000 (8,770,000)
- Steam - (2,161,000) (2,279,000)
- Fuel -
- Total - 2,710,000 (11,050,000)
MACT Control, Reporting, and Recordkeeping Costs:
Capital (Million $) -
Annual (Million $/Yr.) -
Cost-effectiveness :
$ / Mg of HAP
$ / Mg of VOC ...
13,320, 000
16,950,000
2,975,000
33,250,000
496
130
1,020
327
982,000
16,833,000
3,452, 000
21, 300,000
633
162
1,000
358
(173,000)
-
14,300
(158,000)
258
42
16,400
1,290
13,150,000
16, 950, 000
2,989,300
33,100,000
756
172
1,300
400
809,000
16, 833, 000
3,466,300
21,100,000
893
207
1,200
420
aParticulate, CO, nitrogen oxide, and sulfur dioxide emissions are from secondary impacts of OW options.
k Particulate, CO, nitrogen oxide, and sulfur dioxide emissions are increases from baseline due to OW options and after
MACT is applied.
cNumber in parenthesis indicates a decrease in pollutant or energy use.
20-52
-------
REFERENCES
Docket Citation
A-92-40, II-A-35
A-92-40, IV-J-87
A-92-40, IV-A-*
A-92-40, IV-B-16
A-92-40, II-A-17 a
through d
Description
Pulp, Paper, and Paperboard Industry--
Background Information for Proposed
Air Emission Standards. Manufacturing
Processes at Kraft, Sulfite, Soda, and
Semi-Chemical Mills. Publication No.
EPA-453/R-93-050a. October 1993
1996 Lockwood-Post's Directory of the
Pulp, Paper, and Allied Trades.
Miller Freeman Publications, pp.
37-152.
Chemical Pulping Emission Factor
Development Document. Revised Draft.
Prepared by Eastern Research Group for
U.S. EPA. September 1997.
D.B. Greene, ERG, to Penny Lassiter,
EPA:WCPG. August 18, 1997. Addendum
to the baseline level of control
analysis.
Testing of Non-Combustion Processes in
a Pulp an Paper Facility. Champion
International, Cantonment, Florida.
Final Draft Report. Entropy
Environmentalists, Inc. Prepared for
the U.S. Environmental Protection
Agency, Research Triangle Park, North
Carolina. August 1992.
20-53
-------
A-95-31, II-B-1 J.R. Farmer, EPA, to Distribution.
August 22, 1980. Thermal incinerators
and flares.
A-92-40, IV-B-26 R.V. Oommen, ERG, to Penny Lassiter,
EPA:WCPG. October 7, 1997.
Memorandum presenting outputs from the
pulp and paper NESHAP database.
A-92-40, II-I-24 Bleach Plant Chlorine and Chlorine
Dioxide Emissions and Their Control.
National Council of the Paper Industry
for Air and Stream Improvement
(NCASI) . Technical Bulletin No. 626.
September 1991. 84pp.
A-92-40, IV-B-24 D.B. Greene, ERG, to Penny Lassiter,
EPA:WCPG. September 10, 1997.
Memorandum documenting the
supplemental analysis to confirm the
percent mass split requirements
developed for condensate segregation.
A-92-40, II-A-4 OAQPS Control Cost Manual. Fourth
Edition. Publication No. EPA-450/3-0-
-006. January 1990. pp. 2-25 thru 2-
29, 3-1 thru 3-66, 7-38 and 7-39, an
9-1 thru 9-76.
20-54
-------
A-92-40, IV-B-13 R.V. Oommen, Radian Corporation, to
Penny Lassiter, EPA:WCPG. February
26, 1996. Revisions to the approach
used for the design and cost estimates
for equipment associated with
transporting, treating, and
controlling emissions from vents at
pulp and paper mills.
A-92-40, IV-B-28 R.V. Oomen, ERG, to Penny Lassiter,
EPA:WCPG. September 25, 1997.
Characterization of the industry
population affected by the pulp and
paper NESHAP.
A-92-40, IV-E-85 D. Greene, ERG, to P. Lassiter,
EPA:WCPG. August 26, 1997. Summary
of the October 21 and 30, 1996
teleconference with AF&PA
representatives to discuss biological
treatment system issues.
A-92-40, IV-J-78 Steam - Its Generation and Use.
Fortieth Edition. The Babcox and
Wilcox Company. 1992. pp. 8-9, 8-15,
8-17, and 8-18.
A-92-40, IV-B-9 D.B. Greene, Radian Corporation, to
Penny Lassiter, EPA:WCPG. February
22, 1996. Memorandum presenting the
development of kraft pulping
condensate characteristics.
20-55
-------
A-92-40, IV-B-25 D.B. Greene, ERG, to Penny Lassiter,
EPA:WCPG. September 10, 1997.
Memorandum documenting the development
of the. Cost factors to hardpipe
condensates to a biological treatment
system.
A-92-40, IV-B-100 R.V. Oommen, ERG, to Penny Lassiter,
EPA:WCPG. August 18, 1997.
Memorandum presenting the analysis of
control requirements for new pulping
sources.
A-92-40, IV-B-17 D.B. Greene, ERG, to Penny Lassiter,
EPA:WCPG. September 10, 1997.
Memorandum presenting the development
of the pulping process condensates
treatment costing methodology.
A-92-40, IV-D1-59 J.E. Pinkerton, NCASI, Inc., to P.
Lassiter, EPA:WCPG. March 17, 1995.
Description of Assumptions Used to
Develop Comparison Between the
Industry's Condensate Pretreatment
Option and the EPA's Proposed MACT
Standards.
A-92-40, IV-D1-92 Kraft Mill Condensate Treatment System
Conceptual Design and Order-of-
Magnitude Costs Estimate. Prepared by
CH2M Hill for NCASI. December 1994.
20-56
-------
A-92-40, IV-B-10 D.B. Greene, Radian Corporation, to
Penny Lassiter, EPA:WCPG. February
29, 1996. Steam stripper performance
parameters for bleached and unbleached
kraft pulping mills.
A-92-40, IV-J-77 Compilation of Air Pollutant Emission
Factors (AP-42). Volume I:
Stationary Point and Area Sources.
Fourth Edition. Office of Air Quality
Planning and Standards, Research
Triangle Park, North Carolina.
January 1995. pp. 1.1-3, 1.1-9,
1.1-20, 1.3-9, 1.3-15, 1.4-4, through
1.4-6, 1.6-5, 1.6-7, and 1.6-8.
20-57
-------
-------
TECHNICAL REPORT DATA
(please read instructions on the reverse before completing)
1. REPORT NO.
EPA-453/R-93-050b
4. TITLE AND SUBTITLE
Pulp, Paper, and Paperboai
Information for Promulgate
Manufacturing Processes at
Semi - Chemical , Mechanical,
Fiber Mills, Final EIS
2.
-d Industry - Background
>d Air Emission Standards,
. Kraft, Sulfite, Soda,
and Secondary and Non-wood
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND
Office of Air Quality Pi
U.S. Environmental Prote
Research Triangle Park,
ADDRESS
anning and Standards
ction Agency
North Carolina 27711
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Air and Radiation
U.S. Environmental Protection Agency
Washington, DC 20460
3. RECIPIENTS ACCESSION NO.
5. REPORT DATE
October 1997
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
EPA /200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
National emission standards for hazardous air pollutants (NESHAP) are being
promulgated for the pulp and paper industry under authority of Section 112 (d) of the
Clean Air Act as amended in 1990. This background information document provides
technical information and analyses used in the development of the promulgated pulp and
paper NESHAP, and contains responses to comments from the proposed rule. This
document covers air emission controls for wood pulping and bleaching processes at pulp
mills and integrated mills (i.e., mills that combine on-site production of both pulp
and paper) . Effluent guideline limitations for pulp and paper mills are being
developed concurrently under the Clean Water Act . Technical information used for the
development of effluent guideline limitations is in separate documents.
17.
a. DESCRIPTORS
Air pollution
Pollution control
Pulp and paper mills
Hazardous air pollutant
National impacts
Paper mills
18. DISTRIBUTION STATEMENT
Unlimited
KEY WORDS AND DOCUMENT ANALYSIS
b. IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air pollution control 13b
19. SECURITY CLASS (This Report) 21. NO. OF PAGES
UNCLASSIFIED 498
20. SECURITY CLASS (This Page) 22 PRICE
UNCLASSIFIED
EPA Form 2220-1 (Rev. a-77) PREVIOUS EDITION IS OBSOLETE
------- |