EPA-453/R-94-003b
Hazardous Air Pollutant Emissions
from Process Units in the
Synthetic Organic Chemical
Manufacturing Industry-
Background Information
for Final Standards
Volume 2B: Comments on
Wastewater
Emission Standards Division
U.S. Environmental Protection Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
U.S. Environments! Protection Agency
Region 5,Libr-< ,pL-12J)
March 1994
-------
DISCLAIMER
This Report has been reviewed by the Emission Standards
Division of the Office of Air Quality Planning and Standards,
EPA, and approved for publication. Mention of trade names or
commercial products is not intended to constitute endorsement
or recommendation for use. Copies of this report are
available through the Library Services Office (MD-35), U.S.
Environmental Protection Agency, Research Triangle Park, NC
27711, or from the National Technical Information Service,,
5285 Port Royal Road, Springfield, VA 22161.
-------
ENVIRONMENTAL PROTECTION AGENCY
Background Information and Final Environmental
Impact Statement for Hazardous Air Pollutant
Emissions from Process Units in the Synthetic
Organic Chemical Manufacturing Industry
Volume 2B: Comments on Wastewater
Prepared by:
Bruce Jordan (Date)
Director, Emission Standards Division
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
1. The standards regulate emissions of organic hazardous
air pollutants (HAP's) emitted from chemical
manufacturing process units in the Synthetic Organic
Chemical Manufacturing Industry (SOCMI) and from other
processes subject to the negotiated regulation for
equipment leaks. Only those chemical manufacturing
process units that are part of major sources under
Section 112 (d) of the Clean Air Act (Act) will be
regulated. The standards will reduce emissions of 112
of the organic chemicals identified in the Act list of
189 HAP's.
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;
and the Council on Environmental Quality. Copies have
also been sent to members of the State and Territorial
Air Pollution Program Administrators; the Association
of Local Air Pollution Control Officials; EPA Regional
Administrators; and other interested parties.
3. For additional information contact:
Dr. Janet Meyer
Standards Development Branch (MD-13)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Telephone: (919) 541-5254
-------
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
5. Electronic copies of this document may be obtained from
the EPA Technology Transfer Network (TTN). The TTN is
an electronic bulletin board system which is free,
except for the normal long distance charges. To access
the HON BID:
• Set software to data bits: 8, N; stop bits: 1
• Use access number (919) 541-5742 for 1200, 2400,
or 9600 bps modems [access problems should be
directed to the system operator at (919)
541-5384].
• Specify TTN Bulletin Board: Clean Air Act
Amendments
• Select menu item: Recently Signed Rules
-------
OVERVIEW
Emission standards under section 112(d) of the Clean Air
Act (Act) apply to new and existing sources in each listed
category of hazardous air pollutant (HAP) emission sources.
This background information document (BID) provides summaries
and responses for public comments received regarding the
Hazardous Organic National Emission Standard for Hazardous Air
Pollutants (NESHAP), commonly referred to as the HON. The HON
will primarily affect the Synthetic Organic Chemical
Manufacturing, Industry (SOCMI) . However, the provisions for
equipment leaks also apply to certain polymer and resin
production processes, certain pesticide production processes,
and Certain miscellaneous processes that are subject to the
negotiated regulation for equipment leaks.
This BID comprises six volumes as follows:
• Volume 2A: Comments on Process Vents, Storage
Vessels, Transfer Operations, and Equipment Leaks
(EPA-453/R-94-003a);
• Volume 2B: Comments on Wastewater
(EPA-453/R-94-003b);
• Volume 2C: Comments on Emissions Averaging
(EPA-453/R-94-003C) ;
• Volume 2D: "" Comments on Applicability, National
Impacts, and Overlap with Other Rules
(EPA-453/R-94-003d);
• Volume 2E: Comments on Recordkeeping, Reporting,
Compliance, and Test Methods (EPA-453/R-94-003e);
and
• Volume 2F: Commenter Identification List
(EPA-453/R-94-003f).
Volume 2A is organized by emission point and contains
discussions of specific technical issues related to process
-------
vents, storage vessels, transfer operations, and equipment
leaks. Volume 2A discusses specific technical issues such as
control technology, cost analysis, emission estimates,
Group I/Group 2 determination, compliance options and
demonstrations, and monitoring.
Volume 2B addresses issues related to controlling
emissions from wastewater. Specific technical issues include
control technology, cost analysis, emission estimates,
Group I/Group 2 determination, compliance options and
demonstrations, and monitoring.
Volume 2C contains the EPA's decisions regarding
emissions averaging. Specific issues include the scope of
emissions averaging in the HON, specific provisions related to
credits and banking, and enforcement of an emissions averaging
system for the HON.
Volume 2D discusses applicability of the HON in terms of
selection of source category, selection of source, and
selection of pollutants. Volume 2D also addresses the process
for determination of the MACT floor and selection of the
specific applicability thresholds for process vents, storage
vessels, transfer racks, wastewater operations, and equipment
leaks.
Volume 2E discusses the provisions for compliance,
recordkeeping and reporting. Volume 2E also discusses issues
related to the use of EPA test methods.
Volume 2F of each volume contains a list of commenters,
their affiliations, and the EPA docket and item number
assigned to each comment.
-------
TABLE OF CONTENTS
Section Page
LIST OF TABLES xi
ACRONYM AND ABBREVIATION LIST xii
LIST OF FREQUENTLY USED TERMS xvi
1.0 INTRODUCTION 1-1
2.0 CONTROL REQUIREMENTS 2-1
2.1 REFERENCE CONTROL TECHNOLOGY 2-1
2.1.1 Clarification of the Definition of
"Reference Control Technology" .... 2-1
2.1.2 Steam Stripping as RCT 2-2
2.1.3 Comparison of Biological Treatment and
Steam Stripping 2-5
2.1.4 Use of Biological Treatment as a Control
Technology 2-13
2.1.5 Consistency of RON with Benzene Waste
NESHAP, OCPSF, and Other Rules . . . . 2-19
2.1.6 Steam Stripper Design Specifications . 2-23
2.1.6.1 Tray Efficiency 2-26
2.1.6.2 Condenser 2-29
2.1.6.3 Steam-to-Feed Ratio 2-31
2.1.7 Biological Treatment System
Specifications 2-31
2.2 OTHER CONTROL REQUIREMENTS 2-32
2.2.1 Clarification of Requirements for
Control Devices 2-37
2.2.2 Water Seal Controls 2-39
2.2.3 Definition of "Cover" 2-40
2.2.4 Submerged Fill Pipes 2-41
2.2.5 Maintenance Wastewater 2-42
2.2.6 Control of Steam Stripper Overheads . . 2-47
3.0 IMPACTS ANALYSIS 3-1
3.1 COST ANALYSIS 3-1
3.1.1 Recycling vs. Disposal of Residuals . . 3-6
3.1.2 Carbon Steel vs. Stainless Steel ... 3-7
vii
-------
TABLE OF CONTENTS, CONTINUED
Section Page
3.1.3 Heat Transfer Coefficient and Heat
Exchange System 3-8
• 3.1.4 Use of "Temporary" Tanks 3-8
3.1.5 Cost of RCRA Permitting 3-9
3.2 EMISSION ESTIMATES 3-10
3.2.1 Emissions from Biological Treatment
Units 3-24
3.2.2 Use of Wastewater Models 3-27
3.3 OTHER ENVIRONMENTAL IMPACTS 3-30
3.4 ENERGY IMPACTS 3-31
4.0 APPLICABILITY AND GROUP I/GROUP 2 DETERMINATION . . 4-1
4.1 APPLICABILITY 4-1
4.1.1 Definition of "Residuals" 4-4
4.1.2 Definition of "Wastewater" 4-12
4.1.3 Definition of "Wastewater Stream" . . . 4-18
4.1.4 Definition of "Individual Drain System" 4-19
4.1.5 Clarification of "Point of Generation" 4-22
4.1.6 Definition of "Waste Management Unit" . 4-31
4.1.7 Solvent Use as a Feedstock 4-31
4.1.8 Wastewater Generated from Fire Fighting 4-32
4.1.9 Relationship Between Wastewater Tank and
Storage Vessel Provisions 4-32
4.1.10 Previously Installed Steam Strippers . 4-34
4.1.11 Control of Maintenance-Related
Wastewater 4-36
4.1.12 Indirect Discharges 4-39
4.1.13 Clarification of Cooling Tower System . 4-40
4.1.14 Alternative Methods for Determining
Applicability 4-42
4.1.15 Exclusion for Laboratory Waste .... 4-43
4.1.16 One Mg/yr Source-Wide Determination . . 4-43
4.1.17 Clarification of Requirements for
Containers 4-45
viii
-------
TABLE OF CONTENTS, CONTINUED
Section Page
4.2 DETERMINATION OF MOST STRINGENT STANDARDS . . . 4-48
4.2.1 Overlap with the Benzene Waste NESHAP . 4-49
4.2.2 Overlap with the Resource Conservation
and Recovery Act 4-51
4.2.3 Overlap with the Clean Water Act . . . 4-54
4.2.4 Underground Injection Wells 4-58
4.3 GROUP I/GROUP 2 DETERMINATION 4-59
4.3.1 Testing at Peak Levels for Applicability
Determination 4-63
4.3.2 Determining VOHAP Concentration .... 4-63
4.3.3 Sampling at Point of Generation .... 4-65
5.0 COMPLIANCE OPTIONS 5-1
5.1 TARGET REMOVAL EFFICIENCIES 5-1
5.2 MAINTENANCE WASTEWATER 5-10
5.3 MANAGEMENT OF RESIDUALS 5-12
5.4 AVAILABILITY OF SERVICE FIRMS 5-15
5.5 BIOLOGICAL TREATMENT 5-15
5.6 PROCESS UNIT ALTERNATIVE 5-16
6.0 COMPLIANCE DEMONSTRATIONS 6-1
6.1 BIOLOGICAL TREATMENT 6-1
6.1.1 Method 304 6-3
6.1.2 Compliance Issues 6-5
6.2 MONOD EQUATION AND ALTERNATIVE KINETICS
FORMULAS 6-6
6.3 PERFORMANCE TESTING 6-7
6.4 METHODS 25D AND 305 6-9
6.5 TESTING AT PEAK LEVELS FOR COMPLIANCE
DEMONSTRATION 6-11
6.6 USE OF MODELS TO SHOW COMPLIANCE FOR
ALTERNATIVE CONTROL TECHNOLOGY 6-12
6.7 AVAILABILITY OF COMBUSTION TECHNOLOGIES .... 6-13
6.8 USE OF EPA-APPROVED METHODS 6-13
6.9 MONITORING REQUIREMENTS FOR RECYCLED STREAMS . 6-14
IX
-------
TABLE OF CONTENTS, CONTINUED
Section Page
6.10 VENDORS 6-15
6.11 INSPECTIONS 6-15
6.12 MONITORING 6-19
6.12.1 Treatment Processes 6-19
6.12.2 Waste Management Units 6-21
6.12.3 Control Devices 6-25
6.12.4 Method 21 6-27
6.12.5 Heat Exchange Systems 6-34
6.12.5.1 Cooling Tower Systems .... 6-38
6.12.5.2 Once-Through Cooling Water . 6-45
7.0 RECORDKEEPING AND REPORTING 7-1
8.0 WORDING OF THE PROVISIONS 8-1
-------
LIST OF TABLES
Table Page
2-1 EXAMPLE OF BIOLOGICAL FATE DATA IN ACCLIMATED
BIOLOGICAL TREATMENT SYSTEMS 2-18
3-1 SUMMARY OF EPA AND COMMENTER WATER? INPUT
PARAMETERS 3-26
3-2 COMPARISON OF FE VALUES PREDICTED BY WATER? FOR
SELECTED TABLE 9 HAP'S . 3-26
4-1 WASTEWATER TANK CAPACITY AND VAPOR PRESSURE
CRITERIA 4-34
4-2 EMISSIONS FROM HEAT EXCHANGE SYSTEMS (TONS) .... 4-41
XI
-------
ACRONYM AND ABBREVIATION LIST
ACRONYM
Act
ALAPCO
ASPEN
BACT
BAT
BD
BID
BIF
CEM
CFR
CMA
CMPU
CO
CTG
CWA
DMS
DOT
DRE
EB/S
EDC
EFR
EO
E.O.
EPA
Fe
Fm
FR
Fr
FTIR
HAP
TERM
Clean Air Act
Association of Local Air Pollution
Control Officers
advanced system for process
engineering
best available control technology
best available technology
butadiene
background information document
boilers and industrial furnaces
continuous emissions monitoring
Code of Federal Regulations
Chemical Manufacturers Association
chemical manufacturing process unit
carbon monoxide
control techniques guideline
Clean Water Act
dual mechanical seal
Department of Transportation
destruction and removal efficiency
ethylbenzene/styrene
ethylene dichloride
external floating roof
ethylene oxide
Executive Order
Environmental Protection Agency
fraction emitted
fraction measured
FEDERAL REGISTER
fraction removed
Fourier transform infrared
hazardous air pollutant
xii
-------
ACRONYM AND
ACRONYM
HON
IFR
LDAR
LAER
MACT
MIBK
MR
NCS
NESHAP
NOX
NPDES
NRDC
NSPS
NSR
OCCM
OCPSF
OMB
OSHA
P.L.
PAV
POM
POTW
PRA
PRV
PSD
QIP
ABBREVIATION LIST, CONTINUED
TERM
hazardous organic national emission
standards for hazardous air
pollutants
internal floating roof
leak detection and repair
lowest achievable emission rate
maximum achievable control technology
methyl isobutyl ketone
mass removal (actual)
Notification of Compliance Status
national emission standards for
hazardous air pollutants
nitrogen oxides
National Pollutant Discharge
Elimination System
Natural Resources Defense Council
new source performance standards
new source review
Office of Air Quality Planning and
Standards Control Cost Manual
organic chemicals, plastics, and
synthetic fibers
Office of Management and Budget
Occupational Safety and Health
Administration
Public Law
product accumulator vessel
polycyclic organic matter
publicly owned treatment works
Paperwork Reduction Act
pressure relief valve
prevention of significant
deterioration
quality improvement program
Xlll
-------
ACRONYM
ACRONYM
R & D
RCRA
RCT
RIA
RMR
SARA
SIP
SMS
SOCMI
STAPPA
TAG
TACB
TCI
THC
TIC
TOC
TRE
TRI
TSDF
VHAP
VO
VOC
VOHAP
AND ABBREVIATION LIST, CONTINUED
TERM
research and development
Resource Conservation and Recovery
Act
reference control technology
Regulatory Impact Analysis
required mass removal
Superfund Amendment and
Reauthorization Act
State Implementation Plan
single mechanical seal
synthetic organic chemical
manufacturing industry
State and Territorial Air Pollution
Program Administrators
total annual cost
Texas Air Control Board
total capital investment
total hydrocarbon
total industry control
total organic compound
total resource effectiveness
toxics release inventory
treatment, storage, and disposal
facility
volatile hazardous air pollutant
volatile organics measurable by
Method 25D
volatile organic compound
volatile organic hazardous air
pollutant
ABBREVIATION
bbl
BOE
UNIT OF MEASURE
barrel
barrels of oil equivalent
xiv
-------
ACRONYM AND ABBREVIATION LIST, CONTINUED
ABBREVIATION UNIT OF MEASURE
Btu
Btu/kW-hr
°C
0F
gal
gpm
hr
kg/hr
kPa
kW-hr/yr
£/hour»m2
£pm
gal
m3
Mg
mg
mg/dscm
MW
ppb
ppm
ppmv
ppmw
psia
scm/min
TJ
yr
British thermal unit
British thermal unit per
kilowatt-hour
degrees Celsius
degrees Fahrenheit
gallon
gallons per minute
hour
kilograms per hour
kilopascals
kilowatt-hour per year
liters per hour per square meter
liters per minute
gallons
cubic meters
megagrams
milligrams
milligram per dry standard cubic
meter
megawatts
parts per billion
parts per million
parts per million by volume
parts per million by weight
pounds per square inch absolute
standard cubic meter per minute
terajoules
year
xv
-------
LIST OF FREQUENTLY USED TERMS
Act means the Clean Air Act as amended in 1990.
Administrator means the Administrator of the U. S.
Environmental Protection Agency or his or her authorized
representative (e.g., a State that has been delegated the
authority to implement the provisions of part 63).
Enhanced monitoring rule means the rule to be located in
sections 64.1 through 64.9 of part 64 of title 40 of the Code
of Federal Regulations. This rule implements section 702(b)
of title VII of the 1990 Clean Air Act Amendments. This rule
establishes the criteria and procedures that owners or
operators must satisfy in evaluating, selecting and
demonstrating enhanced monitoring, and includes appendices;
containing enhanced monitoring performance and quality
assurance requirements. The enhanced monitoring rule does not
apply to sources subject to 40 CFR part 63, and therefore does
not apply to sources subject to the HON. The proposed rule
was published in the Federal Register on October 22, 1993
(58 FR 54648) .
General Provisions means the general provisions located in
subpart A of part 63 of title 40 of the Code of Federal
Regulations. These General Provisions codify national
emission standards for hazardous air pollutants (NESHAP) for
source categories covered under section 112 of the Act as
amended November 15, 1990.
Implementing aaencv means the Administrator of the U. S.
Environmental Protection Agency or a State, federal, or other
agency that has been delegated the authority to implement the
provisions of part 63. Under section 112(1) of the Act,
States and localities may develop and submit to the
Administrator for approval a program for the implementation
and enforcement of emission standards. A program submitted by
xvi
-------
LIST OF FREQUENTLY USED TERMS, CONTINUED
the State under section 112(1) of the Act may provide for
partial or complete delegation of the Administrator's
authorities and responsibilities to implement and enforce
emission standards.
Operating permit program rule means the rule located in
sections 70.1 through 70.11 of part 70 of chapter I of
title 40 of the Code of Federal Regulations. This rule
implements section 502(b) of title V of the 1990 Clean Air Act
Amendments. Under this rule, States are required to develop,
and to submit to the EPA, programs for issuing operating
permits to major stationary sources (including major sources
of hazardous air pollutants listed in section 112 of the Act),
sources covered by New Source Performance Standards (NSPS),
sources covered by emissions standards for hazardous air
pollutants pursuant to section 112 of the Act, and affected
sources under the acid rain program. The final rule was
published in the Federal Register on July 21, 1992
(57 FR 32250).
Permitting authority means: (1) the State air pollution
control agency, local agency, other State agency, or other
agency authorized by the Administrator to carry out a permit
program under part 70; or (2) the Administrator, in the case
of EPA-implemented permit programs under part 71.
Section 112(a) rule means the rule to be located in subpart B
of part 63 of title 40 of the Code of Federal Regulations.
This rule implements section 112(g) of the 1990 Clean Air Act
Amendments. This rule will impose control technology
requirements on "constructed, reconstructed or modified" major
sources of hazardous air pollutants not already regulated by a
section 112(d) or 112(j) MACT standard.
xvii
-------
LIST OF FREQUENTLY USED TERMS, CONTINUED
Section 112(1) rule means the rule located in subpart E of.
part 63 of title 40 of the Code of Federal Regulations. Under
this rule, a State or locality may submit a program to the
Administrator to request partial or complete delegation oi: the
Administrator's authorities and responsibilities to implement
and enforce section 112 emission standards. The final rule
was published in the Federal Register on November 26, 1992;
(58 FR 62262).
Title III means title III of the 1990 Clean Air Act
Amendments. Section 112 of the Act authorizes the EPA to
establish MACT standards.
Title V means title V of the 1990 Clean Air Act Amendments;,
which authorizes the EPA to establish the operating permit
program.
Title VII means title VII of the 1990 Clean Air Act
Amendments. Section 702(b) of the Act authorizes the EPA to
establish compliance certification procedures. The part 64
enhanced monitoring rule implements section 702(b).
xviii
-------
1.0 INTRODUCTION
On December 31, 1992, the U.S. Environmental Protection
Agency (EPA) proposed the Hazardous organic National Emission
Standard for Hazardous Air Pollutants (NESHAP) for process
units in the synthetic organic chemical manufacturing
industry, commonly referred to as the HON (57 FR 62608). The
HON was proposed under the authority of section 112(d) of the
Clean Air Act. Public comments were requested on the proposed
standard and comment letters were received from industry
representatives, governmental entities, environmental groups,
and private citizens. Also, two public meetings were held,
one in Research Triangle Park, North Carolina, on February 25,
1993, and another in Baton Rouge, Louisiana, on
March 18, 1993. Five people at the North Carolina meeting and
45 people at the Louisiana meeting presented oral testimony on
the proposed NESHAP.
On August 11, 1993, the General Provisions for part 63
(58 FR 42760) were proposed. In order to allow the public to
comment on how the General Provisions relate to the Hazardous
Organic NESHAP (HON), a supplemental notice was published in
the Federal Register (October 15, 1993; 58 FR 53478). Public
comments were requested on the overlap between the General
Provisions and the HON and on some specific emissions
averaging issues. Comment letters regarding the supplemental
notice were received from 80 commenters.
The written comments that were submitted and verbal
comments made at the public hearings regarding the technical
and policy issues associated with wastewater in the proposed
rule and supplemental notice, along with responses to these
comments, are summarized in the following chapters. In
1-1
-------
chapter 2.0, the EPA addresses issues associated with control
requirements including steam stripping as the reference
control technology, the use of biological treatment as a
control technology, and clarification of other waste
management issues. Chapter 3.0 presents the impacts analysis
which includes cost analysis, emission estimates,
environments impacts, and energy impacts. In chapter 4.0,
the EPA provides information on issues related to
applicability and Group I/Group 2 determination including
clarification of definitions and discussion of overlapping
regulations. Chapter 5.0 includes discussion about compliance
options. Chapter 6.0 provides information on compliance
demonstrations, which comprises biological treatment,
performance testing, inspections, and monitoring. In
chapter 7.0, the EPA addresses recordkeeping and reporting
issues. Chapter 8.0 presents several clarifications
concerning wording of the provisions. The summary of comments
and responses serves as the basis for che revisions made to
the NESHAP between proposal and promulgation.
1-2
-------
2.0 CONTROL REQUIREMENTS
In response to commenter confusion, the EPA clarifies the
use of the terms "VOHAP concentration" and "HAP" to reflect
the proper use of the terms throughout the preamble,
regulation, and BID documents. The term "volatile organic
hazardous air pollutant concentration" or "VOHAP
concentration" means the concentration of an individually-
speciated organic HAP in a wastewater stream or a residual as
measured by Method 305. The term "VOHAP" does not refer to
the lists of organic HAP's in tables 8 and 9 of subpart G.
The wastewater provisions of the HON regulate emissions from
wastewater of those organic HAP's listed in table 8 for new
sources and in table 9 for new and existing sources. The
applicability of the requirements in the HON to wastewater
streams is based on the VOHAP concentration of the HAP's
present in the wastewater stream. The VOHAP concentration of
a compound can be calculated by multiplying the HAP
concentration of the compound by the compound-specific
fraction measured (Fm) value listed in table 34 of subpart G.
2.1 REFERENCE CONTROL TECHNOLOGY
2.1.1 Clarification of the Definition of "Reference Control
Technology"
goigfll^rj-h; one commenter (A-90-19: IV-D-73) stated that
the definition of reference control technology for wastewater
attempts to identify all reference control technologies for
collection systems. The commenter (A-90-19: IV-D-73) further
stated that there are so many options available for the
management and treatment of wastewater that it may not be
possible to reiterate them in the definition of reference
control technology. The commenter (A-90-19: IV-D-73)
2-1
-------
suggested revising the definition to clarify that the cited
technologies are examples and not all inclusive.
Response; The EPA clarifies that the definition of
reference control technology (RCT) for wastewater does not
attempt to identify all control technologies for collection
systems. The definition of RCT for wastewater does not
include collection systems. There is not a reference control
technology for collection systems; it is merely a work
practice standard.
The EPA agrees that there are a number of options
available for complying with the wastewater provisions of the
HON. However, the technologies cited in the definition of RCT
for wastewater are not examples. The reference control
technologies cited in the definition are the bases for
determining the equivalent performance of those treatment
technologies that an owner or operator may want to employ as
alternatives to the RCT. The HAP emission reduction achieved
by any alternate treatment technology must be equivalent to or
exceed the HAP emission reduction achieved by the RCT.
2.1.2 Steam Stripping as RCT
Comment; One commenter (A-90-19: IV-D-85) stated that
highly volatile compounds may evaporate before biological
treatment systems have time to work. Therefore, the commenter
(A-90-19: IV-D-85) supported steam stripping as an RCT.
Response: The EPA agrees that volatile HAP compounds may
be emitted to the atmosphere before reaching the biological
treatment unit. However, the EPA recognizes that biological
treatment units can achieve high levels of HAP destruction if
operated properly. Therefore, the EPA maintains the
requirement for suppression of HAP emissions from the
collection system down to the treatment process, such as a
steam stripper or a properly operated biological treatment
unit. Once volatile compounds reach the biological treatment
unit, a performance test using Method 304A, Method 304B, or
other methods described in appendix C of part 63, is required
to ensure that the biological treatment unit is working
2-2
-------
properly and that the biological treatment unit is achieving
the required destruction efficiency.
^oiment:• Several commenters (A-90-19: IV-D-9; IV-D-45
and IV-F-7.7; IV-D-57; IV-D-70; IV-D-85; IV-D-118; IV-D-124;
IV-D-125; IV-F-7.39 and IV-F-12) supported the use of steam
strippers and suppression system components such as covers,
and other control devices to limit air emissions from
wastewater streams. One commenter (A-90-19: IV-F.39 and
IV-F-12) stated that steam stripping is an improvement over
biological treatment. Another commenter (A-90-19: IV-D-57)
stated that the list of HAP's for process wastewater in
tables 8 and 9 of subpart G is correct based on both the RCT
of steam stripping and the volatility of the chemicals.
Several commenters (A-90-19: IV-D-9; IV-D-118; IV-D-124;
IV-D-125) claimed that the provisions which allow biological
treatment as a substitute to steam stripping weaken the
regulation.
Response: While the EPA agrees with the commenter that
steam stripping and suppressed collection units provide good
control of HAP emissions, the EPA does not agree that allowing
biological treatment weakens the regulation. The wastewater
collection system must be suppressed down to the treatment
process that is used to achieve compliance, including
biological treatment, which meets the treatment provisions of
§63.138. Additionally, the biological treatment unit must
achieve an organic HAP emission reduction equivalent to steam
stripping.
Coiqffleip-h; one commenter (A-90-19: IV-D-75) claimed that
steam stripping should not be RCT for wastewater, because it
does not meet the 12-percent criteria. The commenter
(A-90-19: IV-D-75) stated that the primary application of
steam stripping is for product recovery and recycle, and not
for control. The commenter (A-90-19: IV-D-75) claimed that
the EPA could also designate biological treatment as a
\
reference control technology if it is used by 12 percent of
sources and achieves the required efficiency.
2-3
-------
Response; The amount of emissions reduction achieved by
biological treatment, even for biologically degradable
compounds, will vary widely among different facilities due to
the wide range in operating and design parameters which define
a biological treatment system. The parameters which affect
the emission rate of volatile organic compounds include, but
are not limited to, the biological degradation rate, surface
area, amount of aeration, hydraulic residence time, and the
active biomass concentration. Therefore, the performance of
individual biological treatment systems with respect to
volatile organic compound emission reduction will also vary
greatly. For these reasons, the EPA determined that
biological treatment would not be appropriate as the RCT.
Furthermore, it was not possible to predict the performance of
biological treatment units without site-specific data, and
therefore, the EPA selected steam stripping as the basis of
the standard.
The EPA is aware that many SOCMI facilities use
biological treatment units as part of their wastewater
treatment systems. However, because biological treatment
systems are typically located at or near the end of the
wastewater treatment process, many of the volatile regulated
compounds are emitted to the atmosphere prior to reaching the
biological treatment unit. Additionally, not all of the
regulated compounds are significantly biodegradable.
Steam stripping was selected as the reference control
technology (RCT) because it is the most universally applicable
treatment technology for removing volatile organic compounds
from wastewater.
It is assumed that by the term M12-percent criteria" that
the commenter was referring to the requirement in
section 112(d) of the Act that MACT standards for existing
sources must be at least as stringent as the best-performing
12 percent of existing sources. The MACT standard for
controlling HAP emissions from HON wastewater collection and
treatment systems was based on the best control technology
2-4
-------
that was universally applicable to all emission points in the
SOCMI.
2.1.3 Comparison of Biological Treatment and Steam Stripping
qmrcqneiTfr; Four commenters (A-90-19: IV-D-62; IV-D-63;
IV-F-1.2 and IV-F-4); (A-90-23: IV-D-17) stated that
biological treatment is at least as effective at minimizing
emissions as the design stripper and should therefore be
included as an RCT.
Response: The EPA agrees that some HAP's regulated under
the wastewater provisions of the HON can be biologically
degraded at a level equivalent to or exceeding the removal
efficiency achieved by steam stripping. However, this will
depend on the site-specific design and operating parameters of
the biological treatment system. Hence, facilities must
demonstrate that their biological treatment system achieves a
volatile organic HAP emissions reduction equivalent to steam
stripping. The EPA has added an additional biological
treatment option to the final regulation. Under this option,
the owner or operator may biologically treat all process
wastewater. Compliance is achieved by demonstrating
95-percent biodegradation of total mass of HAP's listed on
table 9 of subpart G. Facilities complying with this option
must comply with §63.133 through §63.137 for all process
wastewater streams. However, facilities do not have to comply
with either the applicability determination requirements of
§63.144 or the Group I/Group 2 determinations.
Comment; One commenter (A-90-19: IV-D-32) provided data
depicting a typical configuration of an activated sludge
system that would effectively treat biodegradable HAP's when a
suppressed collection and treatment system is also used. The
commenter (A-90-19: IV-D-32) stated that many of the same
assumptions that the EPA used in the development of the
BACT/LAER document were used by the commenter in developing
the typical activated sludge unit configuration.
The commenter (A-90-19: IV-D-32) provided treatment
efficiency estimates for biological treatment units that were
derived using WATER?. One commenter (A-90-19: IV-D-92) cited
2-5
-------
a study that was completed by the CMA using WATER? that
indicates that biological treatment has a significantly higher
removal rate than steam stripping. The commenter (A-90-19:
IV-D-92) claimed that biological treatment produces limited
air emissions and has a 99-percent removal efficiency for each
HAP in proposed strippability groups A, B, and C.
Two commenters (A-90-19: IV-D-92) (A-90-23: IV-D-9)
stated that biological treatment effectively removes HAP's and
generates a low level of air emissions.
One commenter (A-90-19: IV-D-92) included a report
funded by the EPA which indicates that biological treatment is
an effective way to treat HAP's in wastewater.
One commenter (A-90-23: IV-D-9) stated that a design
steam stripper should not be designated as the RCT for
wastewater because it is infeasible for batch processes. The
commenter (A-90-23: IV-D-9) claimed facilities would have a
difficult or impossible task determining which streams are
subject to the rule because pharmaceutical batch processes
generate numerous wastewater streams. The commenter (A-90-23:
IV-D-9) claimed that batch processes produce wastewaters which
are variable in composition and concentration, and
consequently make the use of a single steam stripper design
impossible. The commenter (A-90-23: IV-D-9) claimed that
achieving the required 95 percent strippability for low
concentration streams would be difficult.
Response: In section 3.2.2 of this BID volume, the EPA
explains why claims by the commenter are based on a flawed
analysis. The biological treatment system described by the
commenter does not achieve 99 percent removal of strippability
group A, B, and C compounds according to WATER? results. As
stated in other responses to comments regarding biological
treatment, the level of biodegradation achieved will vary
among different facilities. The EPA agrees that, under some
conditions, biological treatment will achieve HAP emissions
reduction equal to or exceeding that obtained by steam
stripping. For this reason, the EPA has included biological
treatment options in the proposed and final rule.
2-6
-------
Steam stripping is feasible for batch processes.
Wastewater can be stored until enough of it is accumulated for
treatment. Batch steam strippers are currently available from
vendors. The commenter provided no data to demonstrate that
achieving the required removal efficiencies at low
concentrations will be difficult. Nor did the commenter
(A-90-23: IV-D-9) define "low" concentration. The
applicability threshold of the regulation was chosen to
prevent the inclusion of low concentration streams. The
estimates made by the EPA indicate that the removal
efficiencies required in the HON are achievable for Group 1
wastewater streams.
Comment; Three commenters (A-90-19: IV-D-61; IV-D-92;
IV-D-108) asserted that steam stripping is not justified as an
RCT and that biological treatment should be specified as an
RCT for several reasons. Three commenters (A-90-19: IV-D-32;
IV-D-92), (A-90-23: IV-D-20) claimed that the EPA did not
provide any scientific analysis of HAP removal from steam
stripping in the proposal BID volume IB. The commenter
(A-90-19: IV-D-92) stated that the strippability groups and
target removal efficiencies in table 9 of the proposed rule
are not consistent with the laws of chemistry and
thermodynamics, and that the use of Henry's law constants must
include consideration of solubility. Two commenters (A-90-19:
IV-D-92; IV-D-108) claimed that water soluble HAP's cannot be
removed from wastewater by steam stripping, but the
concentrations of these HAP's can be reduced by biological
treatment. One commenter (A-90-19: IV-D-92) added that the
EPA's inaccurate emissions estimates for biological treatment
units and steam strippers led the EPA to propose the wrong
RCT. One commenter (A-90-19: IV-D-32) requested that the EPA
submit supporting documentation for estimating target removal
efficiencies to the docket prior to promulgation of the final
rule.
Response: It is assumed that the commenter is referring
to the basis for the HAP removal efficiencies achieved by the
design steam stripper. The commenter is correct that this was
2-7
-------
not discussed in the proposal BID. However, the EPA has
documented the RCT performance estimates, for both the
proposed and final regulation. For the proposed rule, steam
stripper performance was documented in a memorandum titled
"Approach for Estimating Emission Reductions of Hazardous Air
Pollutants from Wastewater Streams in the HON," (Docket
No. A-90-23, Item II-B-5). For the final rule, steam stripper
performance is documented in a memorandum titled "Estimating
Steam Stripper Performance and Size," to M.T. Kissell, U.S.
Environmental Protection Agency from C. Bagley, Radian
Corporation, August 24, 1993.
The commenter did not indicate what laws of chemistry and
thermodynamics the target removal efficiency groups and
removal efficiencies are inconsistent with.
It is unclear what the commenter means by stating that
Henry's law constants must consider solubility. The Henry's
law constant describes the proportional relationship between
the concentration of a compound dissolved in a liquid, and the
pressure of that compound in the gas phase above the liquid.
The Henry's law constant of a compound in solution in water is
a function of the temperature and pressure of the system, and
the solubility of the compound.
Water-soluble HAP's are removed by steam stripping. It
is true that if a condenser is present, steam may condense out
with the soluble organics. However, steam strippers are
typically operated at less than 1 kilogram of steam per
kilogram of wastewater. Therefore, water soluble compounds
are concentrated in the overhead stream.
The commenter's statement that the EPA's emission
estimates for biological treatment units are inaccurate is
based on an analysis submitted by another commenter (A-90-19:
IV-D-32). As described in section 3.2.2 of this BID volume,
an input error in the referenced analysis led to erroneous
results. Therefore, the EPA concludes that emissions from
biological treatment units are not overestimated and that the
EPA has not proposed the wrong RCT.
2-8
-------
r QTmnenti One commenter (A-90-19: IV-D-92) stated that
some HAP's are current food sources for microbes in biological
treatment systems. One commenter (A-90-23: IV-D-4) stated
that the EPA's requirement to steam strip methanol would
remove a necessary food source and require the addition of a
different nutrient at the biological treatment plant. The
commenter (A-90-23: IV-D-4) stated that the methanol is
necessary for the control of nitrogen-containing pollutants in
the facility's wastewater streams.
Response; The EPA clarifies that steam stripping is only
one option for complying with the wastewater provisions of the
HON. Any treatment technology may be used, including
biological treatment, provided that the emissions reductions
achieved are equivalent to that achieved by steam stripping.
The EPA also notes that the target removal efficiency for
methanol is relatively low.
Comment; One commenter (A-90-19: IV-D-97) stated that
steam stripping is not a control, but rather a separation
process which requires thermal treatment of concentrated
organic streams. The commenter (A-90-19: IV-D-97) contended
that fuel and organic stream treatment will generate NOX and
C02 pollution, which would not be generated by biological
treatment.
Response; The EPA agrees that steam stripping is a
separation process, which can be used to remove volatile
organic compounds from wastewater. By separating the volatile
organic compounds from the wastewater, their emissions to the
atmosphere could be reduced. The recovered organics from
steam stripper overheads can be recycled, burned as fuel, or
incinerated. Therefore, steam stripping acts as a control
device.
The EPA's analysis shows that secondary emissions
resulting from steam production for steam stripping are
approximately 100 times lower than the resulting HAP-emission
reductions. Therefore, an overall environmental benefit is
achieved. The EPA's analysis is documented in the memorandum
"Secondary Impact Factors used in the Framework for Steam
2-9
-------
Stripping of F/astevater, " to M.T. Kissell, U.S. Environmental
Protection Agency, from K. Pelt, Radian Corporation,
February 1, 1994.
r-mmpient-• Two commenters (A-90-19: IV-D-108); (A-90-23:
IV-D-9), who expressed a preference for the use of biological
treatment, indicated that biological treatment converts HAP's
to nonhazardous constituents, while steam stripping merely
concentrates the hazardous constituents and requires further
treatment. One commenter (A-90-19: IV-D-108) summarized the
results from several studies which seem to indicate that
biodegradation rather than volatilization is the principal
fate of many HAP's in biological treatment units unless the
HAP is highly volatile and resistant to biodegradation.
Response: The data summarized by the commenter
demonstrate the wide range of biodegradation that can be
achieved. For example, the data show that the percent of
methylene chloride biodegraded is reported to be from 43 to
97.1 percent and the amount volatilized is reported to be from
2 to 43 percent. Other compounds for which data are
summarized by the commenter exhibit similarly wide ranges for
the percent biodegraded and/or volatilized.
The EPA agrees that a degree of biodegradation can be
achieved for many organic compounds, but that the degree of
biodegradation for any given compound achieved will vary
widely, depending on the site-specific biological treatment
unit design and operating parameters. Furthermore, the
recovered organics from steam stripper overheads can be
recycled, or burned as fuel, in addition to being incinerated.
Comment; One commenter (A-90-19: IV-D-85) stated that
the EPA should not allow facilities to use biological
treatment to demonstrate equivalence with steam stripping when
treating streams that contain low volatility compounds because
a small percentage of the highly volatile compounds, which
would be removed by steam stripping, will volatilize into the
air during biological treatment.
Response; The EPA allows any technology to be used for
treatment, provided that organic HAP emissions reductions can
2-10
-------
be demonstrated to be equivalent to steam stripping.
Therefore, every compound must have an emission reduction in
the biological treatment unit that is equivalent to the
emission reduction in a steam stripper. For a biological
treatment system, the owner or operator must demonstrate
equivalence per §63.138(b)(1)(iii)(C), §63.138(c)(1)(iii)(D),
or §63.138(e). Because of the site-specific variability in
performance of biological treatment units, it would not be
appropriate to exempt such units from the equivalency
demonstration.
comment; Several commenters (A-90-19: IV-F-1.2 and
IV-F-4; IV-D-113; IV-D-77; IV-D-108) stated that biological
treatment is more cost effective than steam stripping and that
the cost of disposing of consumed biomass generated by
biological treatment units is less than the cost of disposing
of residuals generated by steam strippers.
One commenter (A-90-19: IV-D-71) stated that biological
treatment should be encouraged, where appropriate, since it
can perform the necessary emissions reductions with little
negative impact on the environment.
Response; The commenters (A-90-19: IV-D-113; IV-F-1.2
and IV-F-4) present no emission or cost data to substantiate
their statement that biological treatment is more cost-
effective than steam stripping. The EPA agrees that
biological treatment units may be more cost-effective to
operate than a steam stripper when the biological treatment
unit is demonstrated to achieve mass removal of HAP's equal to
or exceeding that achieved by steam stripping. However, this
depends on the cost (if any) of modifying the biological unit
to achieve the required mass removal. Without data, the EPA
cannot respond more fully.
The EPA has added an additional biological treatment
option to the final regulation. Under this option, the owner
or operator may biologically treat all process wastewater.
Compliance is achieved by demonstrating 95-percent
biodegradation of total mass of table 9 HAP's. Facilities
2-11
-------
complying with this option must comply with §63.133 through
§63.137 for all process wastewater streams.
The cost of disposing of residuals generated by steam
stripping depends on the disposal method. The highest
disposal cost can be expected to occur for off-site
incineration. However, facilities may choose to incinerate
residuals in on-site boilers or to recycle residuals to the
process. Multiple disposal methods, such as landfarming,
composting, and on- and off-site incineration, are also
utilized for disposal of consumed biomass. Depending on the
disposal method, residual disposal cost may or may not exceed
the cost of consumed biomass disposal.
The EPA agrees that a properly operated biological
treatment unit may achieve emission reduction equivalent to
that achieved by steam stripping for some organic HAP's with
little negative impact on the environment. For this reason,
the EPA included the required mass removal provisions in
§63.138(b)(1)(iii)(C), §63.138(c)(1)(iii)(D), or §63.138(e) of
the final rule.
Comment; Two commenters (A-90-19: IV-F-1.2 and IV-lf-4:
IV-D-108) contended that biological treatment was better for
the environment than steam stripping because it uses less
energy and does not generate additional pollution from fuel
combustion.
Response: The EPA agrees that biological treatment of
wastewater consumes less energy than steam stripping and does
not generate the secondary criteria pollutant emissions
associated with the burning of fuel required to generate steam
for steam stripping. However, the HAP emissions reduction
achieved by biological treatment will vary widely among
different facilities. Additionally, not all of the regulated
compounds are significantly biodegradable. Steam stripping is
the most universally applicable treatment technology for
removing volatile organic compounds from wastewater. The
secondary impacts analysis conducted by the EPA shows that the
secondary criteria pollutant impacts of the HON are 100 times
less than the HAP emission reduction resulting from steam
2-rl2
-------
stripping of wastewater. Therefore, the EPA concludes that
steam stripping of wastewater is an environmentally acceptable
treatment technology.
The secondary impacts estimated by the EPA are based on
the assumption that all facilities will treat all the Group 1
streams using steam stripping. However, some facilities may
choose other treatment technologies, including biological
treatment, which use less energy and generate less secondary
criteria pollutant emissions than steam stripping. Therefore,
the EPA's estimate of secondary impacts represents
conservative estimates of the secondary impacts associated
with the HON.
2.1.4 Use of Biological Treatment as a Control Technology
Comment; Several commenters (A-90-19: IV-D-58; IV-D-77;
IV-D-92; IV-F-1.2 and IV-F-4) (A-90-23: IV-D-4; IV-D-9;
IV-D-20) stated that the most common type of wastewater
treatment currently employed by existing SOCMI sources is
biological treatment.
Several commenters (A-90-19: IV-D-32; IV-D-34; IV-D-55;
IV-D-62; IV-D-63; IV-D-67; IV-D-77; IV-D-79; IV-D-86; IV-D-92;
IV-D-97; IV-D-113; IV-F-1.2 and IV-F-4) (A-90-23: IV-D-4;
IV-D-171; IV-D-20) asserted that the EPA's decision should
include biological treatment as an RCT in the final rule.
One commenter (A-90-19: IV-D-71), who supported
biological treatment as an RCT, stated that biological
treatment systems are being used successfully for control of
wastewater and for the remediation of groundwater and soils.
Three commenters (A-90-19: IV-D-32; IV-D-86; IV-D-113)
recommended the use of biological treatment as an RCT option
because it will provide treatment for organic HAP's in many
wastewater streams, such as Group 2 streams, that are not
subject to the control requirements of the proposed rule.
Two commenters (A-90-19: IV-D-77; IV-D-79) stated that
the EPA's assessment of MACT was flawed by the omission of
biological treatment as an RCT.
Response; Steam stripping was selected as the RCT
because it is the most universally applicable treatment
2-13
-------
technology for removing volatile organic compounds from
wastewater.
Typically, the biological treatment system at a SOCMI
facility is at or near the end of the wastewater treatment
process. Therefore, many of the regulated compounds could be
emitted to the atmosphere prior to reaching the biological
treatment unit due to their volatility if there is no emission
suppression up to the biological treatment unit.
Additionally, not all of the regulated compounds are
significantly biodegradable. The amount of emissions
reduction achieved by biological treatment, even for
biologically degradable compounds, will vary widely among
different facilities due to the wide range in operating and
design parameters which define a biological treatment system.
The parameters which affect the emission rate of volatile
organic compounds include, but are not limited to, the
biological degradation rate, surface area, amount of aeration,
hydraulic residence time, and the active biomass
concentration. These parameters will vary widely among
facilities. Therefore, the performance of individual
biological treatment systems with respect to volatile organic
compound emission reduction will also vary greatly.
The EPA is aware that many SOCMI facilities employ
biological treatment units as part of their wastewater
treatment systems. However, for the reasons discussed above,
the EPA did not select biological treatment as the reference
control technology.
It would be inappropriate for the EPA to define the
wastewater RCT as biological treatment based on the fact that
biological treatment units will treat organic HAP's other than
those regulated under the HON. Steam strippers will also
remove organic HAP's not regulated by the HON wastewater
provisions, to the extent that such compounds exist in Group 1
streams.
Comment; One commenter (A-90-19: IV-D-32) stated that
the wastewater MACT floor should include biological treatment
with suppressed sewer systems.
2-14
-------
Response: in order for the MACT floor to reflect
biological treatment with suppressed sewer systems, the EPA
would have to demonstrate that this level of control currently
exists at the average of the top 12 percent of SOCHI
facilities. In the preamble to the proposed HON regulation,
the EPA requested specific information which would support an
analysis to determine if biological treatment with suppressed
sewer systems in fact reflects the MACT floor for wastewater
in the SOCMI. Although the EPA did receive information during
the comment period, the data were not sufficient to indicate a
MACT floor based on biological treatment with suppressed sewer
systems.
However, information provided by the commenter (A-90-19:
IV-D-32) was used to revise estimates of volatile organic HAP
emissions from wastewater. These revisions included
accounting for some suppression of the wastewater collection
and treatment system, whereas in the original analysis, the
EPA had assumed the wastewater collection and treatment system
was unsuppressed. The EPA clarifies that biological treatment
units were included in the original analysis.
Comment; Several commenters (A-90-19: IV-D-32; IV-D-75)
(A-90-23: IV-D-2) suggested the EPA incorporate a combination
of an emissions-suppressed collection system with a biological
treatment unit as an RCT in the HON for those chemicals that
are effectively biodegradable. One commenter (A-90-19:
IV-D-97) stated that biological treatment should be the
primary RCT choice, followed by steam or air stripping for
those HAP's that are nonbiodegradable.
The commenters (A-90-19: IV-D-32; IV-D-75) (A-90-23:
IV-D-2) referred to the following references which indicate
that biological treatment is an effective means of
biodegrading a wide range of chemical compounds:
1. A table from an EPA publication entitled "Estimating
Releases and Waste Treatment Efficiencies for the
Toxic Chemical Release Inventory Form," which
indicates that most nonchlorinated organics have
high percent removal and low volatilization in an
acclimated biological treatment system,
2-15
-------
2. The NETAC data base, associated with the University
of Pittsburgh and maintained for the biological
treatment industry, which indicates that there are
microbes that can remove chlorinated compounds, and
3. A nomograph plotting the percent biodegraded for
IS compounds as a function of the compounds' Henry's
law constants and biokinetic rate constants, K^.
Several commenters (A-90-19: IV-D-32; IV-D-33; IV-D-58;
IV-D-71; IV-D-75) asserted that many of the HAP's listed in
the regulation are not volatile and cannot be removed by steam
stripping but can be removed using biological treatment. Two
commenters (A-90-19: IV-D-32; IV-D-75) suggested that the EPA
should specify biological treatment as RCT for all VOHAP's
with Henry's law constants less than 10"4 atm-m3/mole, which
are biodegraded at an efficiency of 98 percent or greater.
Response; The EPA agrees that biological treatment can
play an important role in the reduction of organic HAP
emissions from wastewater, and has therefore included
provisions for biological treatment as a compliance option in
the wastewater provisions of the HON. jSteam stripping was
selected as the wastewater RCT because it is the most
universally applicable treatment method for removing volatile
organic HAP's from wastewater and thereby preventing their
release to the atmosphere. Steam stripping effectively
removes all the compounds regulated under the wastewater
provisions of the HON, including those which are not readily
biologically degradable. The development of the design steam
stripper which is the basis for the performance standards for
the wastewater provisions of the HON is documented in the
memorandum "Estimating Steam Stripper Performance and Size,"
to M.T. Kissell, U.S. Environmental Protection Agency from C.
Bagley, Radian Corporation, August 24, 1993.
The table referenced by one commenter (A-90-23: IV-D-21)
from the-EPA publication "Estimating Releases and Waste
Treatment Efficiencies for the Toxic Chemical Release
Inventory Form," does not fully indicate that most chlorinated
compounds have high percent removals in acclimated biological
treatment systems. The document states that a number of
2-16
-------
design and operational factors will affect the fate of the
listed compounds at any given treatment plant and the data
should be used only as a rough approximation.
Table 2-1 of this BID volume summarizes some of the
contents of the referenced table for eight of the chlorinated
compounds regulated by the wastewater provisions of the HON.
Each of these compounds has a steam stripper removal
efficiency in excess of 99 percent, based on the design and
operating parameters of the design steam stripper. This
exceeds the percent biodegradation values in table 2-1 of this
BID volume for all the compounds listed. However, facilities
may choose to use biological treatment if it can be
demonstrated to meet the applicable requirements in
§63.138(b)(1)(iii)(C), (c)(1)(iii)(D), or (e).
Based on a comment received from one commenter, the EPA
obtained data from the National Environmental Technology
Applications Corporation (NETAC) data base. This corporation
provides profiles and descriptions of technical principles,
applications, operating features, and innovative technologies
for waste treatment technologies. However, this information
is provided for a fee, and the commenter did not include data.
The EPA encourages sources to use all available information at
their disposal, but cannot comment further on this particular
source without specific data.
The EPA has also reviewed the third data set submitted by
commenters; which is nomograph that presents data for
18 compounds. The fraction biodegraded values used in the
analysis are based on a faulty estimation of the percent of
biodegradation and are discussed in more detail in
section 3.2.2 of this BID volume. Therefore, the revisions
suggested by the commenters have not been implemented. The
EPA clarifies that, based on these and additional comments
received, seven compounds were deleted from regulation under
2-17
-------
Q
W
1
H
J
U
u
<
z
H
<
s-
a w
s:
W H
EH EH
<< W
fa SH
j W
M
< EH
u a
H H
8E
s<
O W
H «
03 EH
fa J
O^j
<
O
W M
iJ O
Oi Q
S iJ
s °
>< H
U CO
•
H
1
(N
TABLE
•a
0)
•d
2
0>
0)
•8
,_J
^^n
oa
<^
0
-P
T3
0)
C (U
o cr
•H -a
•P s
•H rH
4J cn
M
(0
a<
<*p
o
•p
"8
m
N
•H ..
-3
4j •<
A)
rH
0
c^P
T3
Compoui
H o o co _ n ir»
^t in ^- H TO CM CM
m t o H c* vo i
O U 0) rH -H JC
k jJ O .C >H -P
o a) M o o a) a
rH EH O -H H O C
JC 1 -H r4 £ U 0
O i U O
(M H *J H C -H rH
* » 0) «. -H ^ fi
H H EH H > EH U
0)
(fl
(0
Q)
rH
0)
rH
«J
u
•H
§
w
u
0
•H
x
o
EH
0)
£
4J
b
o
H
H
2-18
-------
subparts F and G, as discussed in section 3.2 of this BID
volume.
2.1.5 Consistency of HON with Benzene Waste NESHAP. OCPSF.
and Other Rules
gnm]mep^; Several commenters (A-90-19: IV-F-1.6 and
IV-F-6; IV-D-67; IV-D-55; IV-D-32; IV-D-62; IV-D-63; IV-D-58;
IV-D-92) (A-90-19: IV-D-20) suggested that by adding
biological treatment as an RCT to treat all of the organic
HAP's addressed in the rule, the HON would be consistent with
the Benzene Waste NESHAP requirements.
Response: Although the EPA has not included biological
treatment as the RCT in the final rule, biological treatment
remains an allowable method of treatment using mass removal
calculations to show compliance. With regard to the
consistency with the Benzene Waste NESHAP, the EPA has allowed
the use of biological treatment as an alternative compliance
option to treat benzene waste that is subject to the Benzene
Waste NESHAP because the EPA determined that biological
treatment systems would sufficiently biodegrade benzene in
dilute wastewater streams. However, the HON includes many
more regulated HAP's in addition to benzene. The EPA
continues to allow the use of biological treatment units as
treatment for HAP's regulated by the HON as long as the
biological treatment systems achieve the RMR of HAP's from the
wastewater stream as determined by §63.145(h)(2). The EPA has
added a compliance option to the final rule that allows for
the treatment of all process wastewater streams in a
biological treatment unit. An owner or operator who elects to
use this option in §63.138(e) must ensure that the biological
treatment unit achieves 95-percent HAP removal from the
wastewater streams as specified in §63.145(i).
Comment; Two commenters (A-90-19: IV-D-67; IV-D-58)
suggested listing ranges for the operating parameters of a
biological treatment unit, as in the Benzene Waste NESHAP.
One commenter (A-90-19: IV-D-67) also suggested that the EPA
should provide a maximum inlet concentration for each
individual pollutant entering a biological treatment system to
2-19
-------
minimize risk of volatilization (i.e., 10 ppmw as in the
Benzene Waste NESHAP) .
Response : The Benzene Waste NESHAP regulates only a
single chemical. However, the HON regulates 75 additional
HAP's, and the EPA was unable to determine a set of operating
parameters for biological treatment units which would ensure
HAP emission reductions of all regulated compounds at levels
equivalent to steam stripping. Therefore, the EPA has not
listed operating ranges for biological treatment units in the
HON.
Comment; Three commenters (A-90-19: IV-D-55; IV-D-58;
IV-F-1.6 and IV-F-6) stated that many facilities are currently
using biological treatment units to comply with Benzene Waste
NESHAP requirements. Several commenters (A-90-19: IV-D-55;
IV-D-58; IV-D-62; IV-F-1.6 and IV-F-6) are concerned that many
compliance uncertainties and complexities will result when
SOCMI and non-SOCMI wastewater streams (e.g., refinery
wastewater streams) are combined in the collection system or
the biological treatment unit.
Response: The commenter does not provide details
concerning the types of uncertainties and complexities that
could result from treating wastewater streams that are subject
to both the HON and the Benzene Waste NESHAP. However, the
final regulation addresses overlap with other regulations for
wastewater at §63. 110 (e) . In the final rule, after the dates
of compliance specified in §63. 100 (k) of subpart F, streams
subject to the HON and to the Benzene Waste NESHAP must comply
with both regulations. The EPA cannot anticipate every site-
specific situation, but has developed the regulation to
provide flexibility in the methods available for compliance.
One commenter (A-90-19: IV-D-89) claimed that
the type of steam stripper required in the Benzene Waste
NESHAP may not be acceptable under the proposed HON. Another
commenter (A-90-19: IV-D-92) stated that refineries which
have chosen to comply with the Benzene Waste NESHAP using
biological treatment will be required to adopt additional,
2-20
-------
expensive treatment methods to comply with both the HON and
the future petroleum refinery MACT standards.
Response; An existing steam stripper that was installed
to comply with the Benzene Waste NESHAP can still be used to
comply with the HON. For an existing source, any treatment
process can be used to demonstrate compliance with the HON
wastewater provisions as long as the treatment process
complies with any of the required treatment options in
§63.138.
Comment; Two commenters (A-90-19: IV-D-64; IV-D-86)
supported biological treatment as the RCT for SOCMI facilities
and stated that (1) biological treatment is the most common
technology used by direct dischargers in the OCPSF wastewater
category, and (2) companies have made significant expenditures
to meet OCPSF limitations using biological treatment.
Response; Owners and operators may use existing
biological treatment units to meet the HON wastewater
treatment requirements by demonstrating the required level of
biod'egradation. In cases where biological treatment cannot
meet the required biodegradation, facilities may need to
install additional treatment equipment such as a steam
stripper. In the final rule for OCPSF effluent guidelines
(52 FR 42561), the EPA stated that facilities should consider
incorporating steam stripping as the treatment method for
meeting the effluent guidelines because subsequent air
emission regulations may require steam stripping.
Comment; One commenter (A-90-19: IV-D-85) indicated
that the EPA had strongly recommended in the Federal Register
(52 FR 42561) that facilities incorporate steam stripping to
comply with the OCPSF effluent guidelines to avoid costly
retrofit requirements that may subsequently be imposed under
the Act. Therefore, the commenter (A-90-19: IV-D-85)
contended that the EPA should not allow facilities to comply
with the HON through the use of treatment that is less
effective than steam stripping.
Response: The EPA clarifies that steam stripping is an
effective method of treatment to reduce HAP emissions from
2^21
-------
wastewater. However, the EPA continues to allow other
treatment processes that can achieve equivalent HAP emission
reductions in order to allow flexibility in compliance and the
opportunity for sources to use existing control equipment:.
eoimnent• One commenter (A-90-19: .IV-D-92) stated that
the OCPSF Effluent Guidelines and Standards cite biological
treatment as the best available technology for SOCMI and as a
control technology associated with the NSPS for new SOCMI
sources, and allows the use of biological treatment to comply
with pre-treatment requirements. The commenter (A-90-19:
IV-D-92) also claimed that subpart I (Direct Discharge Point
Sources That Use End-of-Pipe Biological Treatment) of these
standards gives effluent limits in the low parts per billion
range for many of the chemicals subject to the HON. The
commenter (A-90-19: IV-D-92) cited these other regulations as
support for including biological treatment as the RCT in the
HON.
Response: The EPA continues to allow treatment processes
other than steam strippers to meet HAP emission reduction
requirements. Facilities that have previously installed
biological treatment systems to comply with other regulations
may continue to use them to comply with the HON as long as
they meet the required HAP emission reductions.
Comment; One commenter (A-90-23: IV-D-9) expressed
concern that future pharmaceutical NPDES effluent guidelines
may conflict with the type of steam stripping designated by
the HON.
Response; The commenter does not provide any details
about how the two regulations may conflict. Neither the HPDES
effluent guidelines nor the HON specify that a steam stripper
must be used to achieve compliance, therefore, without more
information, no conflict has been identified.
Comment; One commenter (A-90-19: IV-D-92) reported that
SOCMI facilities holding NPDES permits are required to conduct
"whole effluent biological testing" to determine if the
effluent from the facilities causes mortality or morbidity in
"exquisitely sensitive" biological organisms (e.g., 7-day old
2-22
-------
fat head minnows). The commenter (A-90-19: IV-D-92) stated
that this type of testing was completed and the results do not
indicate that the chemicals on the HAP list in the HON are
toxic to these organisms.
Response; Although facilities holding NPDES permits may
be required to conduct whole effluent biological testing, the
results of such tests do not necessarily indicate compliance
with the HON. Furthermore, the tests required for facilities
holding NPDES permits are designed to determine the toxicity
of the effluent and do not determine anything about air
emissions. Additionally, toxicity does not always correlate
with concentration or quantity. For example, a large amount
of one compound may not be as toxic as a small amount of
another compound.
2.1.6 Steam Stripper Design Specifications
Comment: Two commenters (A-90-19: IV-D-104; IV-D-108)
argued that the EPA should not specify the design of a steam
stripper. One commenter (A-90-23: IV-D-9) claimed that the
EPA's design steam stripper does not reflect practical
performance. Several commenters (A-90-19: IV-D-75; IV-D-104;
IV-D-108) contended that the design of the steam stripper does
not adequately consider site-specific conditions such as steam
quality, the specific mix of chemicals in the wastewater,
quantity of wastewater, wastewater salinity, total carbon, and
the method by which individual strippabilities were
determined. Two commenters (A-90-19: IV-D-75; IV-D-108) said
that these variables must be considered to optimize
effectiveness and cost and are best addressed on a
site-specific basis. Two commenters (A-90-19: IV-D-104;
IV-D-108) recommended that the agency allow an alternative
steam stripper design which meets the performance criteria of
the RCT.
Response; The provisions in §63.138(g) of the final rule
for a design steam stripper is one option for complying with
the wastewater treatment provisions of the HON. Other control
technologies, including steam strippers with alternative
designs, are also allowed as long as they meet the performance
2-23
-------
criteria of the RCT (e.g., outlet concentration or required
mass removal). However, a steam stripper meeting the design
criteria in §63.138(g) does not require a compliance
demonstration, whereas a compliance demonstration is required
for any alternative design. Owners and operators are required
to comply with the appropriate monitoring, reporting, and
record keeping provisions regardless of whether the treatment
device complies with the design steam stripper provisions in
§63.138(g) or is an alternative design.
Steam stripper performance was determined using the
Kremser equation. The Kremser equation can be used to
determine the removal efficiency of a steam stripper at a
given steam-to-feed ratio, number of theoretical trays, and
compound-specific Henry's law constants. The commenters
(A-90-19: IV-D-104; IV-D-108) did not propose a method for
accounting for the site-specific conditions specified in their
comments, or suggest how these parameters might affect the
design of the steam stripper.
Comment: Several commenters (A-90-19: IV-D-32; IV-D-64;
IV-D-108) (A-90-23: IV-D-20) stated that the EPA's design of
the RCT steam stripper should include only the minimum design
specifications needed to achieve the performance standard.
Two commenters (A-90-19: IV-D-32; IV-D-108) requested that
the EPA require a performance standard rather than an
equipment standard to be achieved by the RCT steam stripper.
The commenters (A-90-19: IV-D-32; IV-D-108) suggested that
the EPA could implement a performance-oriented standard by
including a minimum number of equilibrium stages and a HAP
removal target rather than extensive design specifications.
Response; The RCT for wastewater is not limited to an
equipment standard. The RCT for wastewater includes the
following options: (1) an equipment standard in §63.138(g);
(2) performance standards in §63.138(b)(1)(ii)(A) or (C) ,
§63.138(b)(1)(iii)(A) or (C), §63.138(c)(1)(ii)(B), (C), or
(D), and §63.138(c)(1)(iii)(B), (C), or (D); or (3) work
practice standards in §63.138(b)(1)(i) or (c)(1)(i). The EPA
selected this approach for defining RCT in order to provide
2-24
-------
flexibility while ensuring equivalent levels of control. The
design steam stripper specified in §63.138(g) is only one of
several compliance options. Facilities may use existing steam
strippers not meeting the design specifications of §63.138(g),
but must demonstrate the existing steam stripper is equivalent
to the RCT by demonstrating compliance with the applicable
performance standards. For example, an existing facility
using an existing steam stripper to treat a combination of
Group 1 wastewater streams has a choice of demonstrating
compliance with §63.138(c)(1)(iii)(B), (C), or (D).
A combination of a standard including a minimum number of
equilibrium stages and a HAP removal target would not serve to
augment or simplify the current wastewater provisions of the
HON. The current wastewater provisions provide a compliance
option of meeting a target HAP removal. Combining this with a
required number of theoretical plates would be more
restrictive than the current provisions for meeting the target
HAP removal which does not require a set number of theoretical
plates.
Comment; One commenter (A-90-19: IV-D-73) alleged that
only the removal efficiency should be specified for the RCT
and that all other design parameters should be specified by
the owner or operator on a case-by-case basis using a process
simulation model and the required HAP removal. The commenter
(A-90-19: IV-D-73) was specifically concerned with specifying
one steam-to-feed mass ratio and cited a report which shows
different steam-to-feed ratios that can be used to achieve a
99-percent removal in columns operating at different pressures
for a benzene-water mixture.
Response: The current wastewater provisions of the HON
allow owners or operators to meet the target HAP removal
requirements using a steam stripper with different operating
and design parameters than those specified in §63.138(g) of
the final rule. Owners and operators may also use treatment
technologies other than steam stripping to meet target HAP
removal requirements. However, any treatment device which
differs from the design and operating requirements of
2-25
-------
§63.138(g) is subject to a compliance demonstration by the
procedures in either paragraph §63.138(j)(l) or (j)(2).
-------
universally applicable (i.e., packed columns are more
susceptible than tray columns to problems as a result of
wastewater particulate loading).
Comment: One commenter (A-90-23: IV-D-17) disagreed
with the concept of defining a minimum number of theoretical
trays for the design steam stripper. The commenter (A-90-23:
IV-D-17) claimed that column dynamics and vapor-liquid
equilibrium based on actual data from a steam stripper must be
known in order to determine the number of theoretical stages.
The commenter (A-90-23: IV-D-17) then stated that scale-up is
difficult even if actual data is known because of dispersion
caused by radial and axial mixing.
Response; The EPA agrees with the commenter regarding
defining the number of trays, and has revised §63.138(g) of
the final rule to specify the number of actual trays.
However, actual steam stripper data is not required to
determine the number of theoretical trays. The number of
theoretical trays can be determined using the Kremser equation
knowing only the steam-to-feed ratio, the desired treatment
performance, and compound-specific Henry's law constants.
Scale-up from the number of theoretical trays to actual trays
can be accomplished using an overall tray efficiency which
corrects for the assumption that vapor-liquid equilibrium is
not reached at each individual stage in the column. For the
final rule, the EPA used an overall tray efficiency of
30 percent to scale from theoretical to actual trays.
Comment: One commenter (A-90-19: IV-D-75) provided
results of steam stripper simulations using a PRO-II program.
The results indicate that every separation in the study was
achieved with three or four theoretical stages by using a
steam rate of > 100 Ib/hr, corresponding to a steam-to-feed
ratio of 0.12 to 0.14 kilogram of steam per kilogram of
wastewater.
Response: The EPA agrees that the required treatment
performance levels can be achieved by a steam stripper with
three theoretical trays. However, the commenter assumes a
2-27
-------
different wastewater feed temperature to the steam stripper.
Whereas the commenter assumed a feed temperature of 35 °c, the
EPA .assumed that the wastewater feed would be preheated to
95 °C before entering the steam stripper column. Therefore,
the commenter's steam-to-feed ratio of 0.12 to 0.14 kilograms
of steam per kilogram of wastewater is higher than would be
required for a system in which the wastewater feed to the
steam stripper is preheated. The EPA's analysis shows that a
steam-to-feed ratio of 0.04 kilograms of steam per kilogram of
wastewater is sufficient to achieve the required performance.
Comment; One commenter (A-90-19: IV-D-32) expressed
several concerns with the EPA's current design and stated that
the EPA must correct the inconsistency between the number of
theoretical and physical trays required for the steam stripper
in the proposed rule. Based on an 80 percent tray efficiency,
the commenter (A-90-19: IV-D-32) calculated that 13 physical
trays instead of 10 physical trays, as stated in the proposed
rule, would be required for a steam stripper with ten
theoretical stages.
Response: The EPA agrees with the commenter and has
corrected the inconsistency in the proposed rule. The final
rule requires 10 actual trays, assuming a 30-percent tray
efficiency.
Comment; One commenter (A-90-19: IV-D-32) disagreed
with the EPA's assumption of an 80-percent tray efficiency in
the RCT design stripper7 stating that it overestimates
stripping performance. The commenter (A-90-19: IV-D-32)
stated that Dr. James Fair of the Department of Chemical
Engineering at the University of Texas at Austin was
contracted to calculate tray efficiencies for each HAP listed
in table 9 of subpart G using the Kremser equation. The
commenter (A-90-19: IV-D-32) included, the tray efficiency
estimates in appendix K of the comment letter. The commenter
(A-90-19: IV-D-32) stated that the EPA may have overlooked
the possibility that tray efficiency is a function of the
vapor/liquid ratio. The commenter (A-90-19: IV-D-32)
contended that to achieve the EPA's target strippabilities for
2-28
-------
less-strippable HAP's, a bigger column and more steam would be
required, thus, increasing both capital and operating costs of
the steam stripper.
Response: The EPA agrees with the commenter that a tray
efficiency of 80 percent is too high, and the EPA has revised
the design steam stripper to include a tray efficiency of
30 percent. Three theoretical trays are required to achieve
the target removal efficiencies, and therefore, ten actual
trays are required to achieve the target removal efficiencies.
The revised capital and annual costs were both based on a
steam stripping column with ten actual trays. Therefore, a
bigger column and more steam are not required, and the capital
and annual costs are not underestimated.
Furthermore, the commenter (A-90-19: IV-D-32) did not
provide tray efficiencies for each compound; appendix K only
provides data on the tray efficiency for toluene. The EPA had
to consider the tray efficiency of every table 9 compound in
their analysis.
Comment; One commenter (A-90-19: IV-D-32) suggested
that the EPA specify the number of theoretical "stages" rather
than the number of trays in §63.138(f)(2) because this
approach would be more consistent with design terminology for
steam strippers.
Response: The EPA intended to specify 10 actual trays,
not 10 theoretical trays as is found in §63.138(f)(2) of the
proposed regulation [§63.138(g)(2) of the final provisions].
When specifying theoretical trays, the term "theoretical
stages" is often used. However, in the final regulation,
§63.138(g)(2) specifies 10 actual trays. The EPA has chosen
to use the term "trays" instead of "stages" to clarify that
§63.138(g)(2) requires the design steam stripper to have
10 actual, not theoretical, trays.
2.1.6.2 Condenser
Comment: One commenter (A-90-19: IV-D-32) stated that
the EPA should not specify the type of condenser used for the
RCT steam stripper in §63.138(f)(6) of the proposed rule
2-29
-------
because downstream vapor control requirements ensure that
emissions from the primary condenser will be controlled.
One commenter (A-90-19: IV-D-78) reasoned that the EPA
should specify the vapor temperature at the outlet of the
final steam stripper overheads condenser, since some steam
strippers will have a series of overheads condensers.
Response; The EPA has deleted the specification for a
condenser from §63.138(f) of the proposed rule
(i.e., §63.138(g) of the final rule). If a primary condenser
is used as part of a steam stripper, the non-condensible gas
stream from the primary condenser must be controlled per the
requirements of §63.138(i).
Comment; One commenter (A-90-23: IV-D-20) stated that
in §63.138(f)(6) of the proposed rule, the steam stripper
requirements should not require the use of a water-cooled
condenser because refrigerated condensers can achieve the same
results.
Response: Although the design steam stripper defined by
§63.138(g) did require a specific type of condenser, the EPA
has reviewed public comments and concluded that any air
pollution control device, including but not limited to a
condenser, is allowable under the final rule as long as the
steam stripper achieves all emission control requirements
required in §63.138(i).
Comment: One commenter (A-90-23: IV-D-18) suggested.
that the EPA remove from tables 8 and 9 in proposed §63.131
all chemicals that have boiling points less than 50 °C,
because the condenser of the design steam stripper will not be
able to liquify the vapor emitted from the steam stripper.
The commenter (A-90-23: IV-D-18) stated that the design steam
stripper would evaporate those HAP's with boiling points less
than 50 °C directly into the air without any reduction in air
emissions.
Response; The intention of the design steam stripper
provisions in §63.138(g) of the final rule is not to allow
HAP's removed from wastewater by steam stripping to be emitted
to the atmosphere. The regulation requires control of these
2-30
-------
emissions, as specified in §63.138(i). Therefore, rather than
removing chemicals with boiling points below 50 °C, as
suggested, the condenser requirement in proposed §63.138(f)(6)
has been removed. However, the owner or operator must comply
with §63.138(i) of the final rule, which specifies options for
controlling the emissions from the steam stripper overheads
primary condenser.
2.1.6.3 Steam-to-Feed Ratio
Comment: One commenter (A-90-19: IV-D-32) stated that
the steam-to-feed ratio and liquid loading of steam strippers
will depend on site-specific design and operating conditions.
The commenter (A-90-19: IV-D-32) provided information [in
appendix M of the comment letter (letter from B. Davis to
J. Meyer, April 1, 1993)], which indicates that the steam-to-
feed ratio will vary depending on the chemicals, their
concentrations, and the operating pressure of the tower.
Response; The steam-to-feed ratio and the liquid loading
specified in §63.138(g)(3) and (g)(5) of the final rule will
achieve the required performance levels under the conditions
specified for the design steam stripper in §63.138(g) of the
final rule. However, some facilities may choose to install a
steam stripper design other than that specified in §63.138(g).
For example, an owner or operator may choose to install a
packed column rather than a tray column, or operate the steam
stripper under vacuum. Any steam stripper design is allowed
under the wastewater provisions as long as it meets the
performance criteria of the RCT. However, for any steam
stripper not consistent with the design and operating
requirements in §63.138(g), the owner or operator must
demonstrate compliance with the required performance levels.
A compliance demonstration is not required if the steam
stripper meets the provisions of §63.138(g).
2.1.7 Biological Treatment System Specifications
Comment: One commenter (A-90-19: IV-D-34) recommended
general guidelines for a well-operated biological system
including the use of maximum or minimum limits instead of
ranges, the specification of operating parameters that are
2-31
-------
controllable or can be modified to meet MACT, the elimination
of redundant or conflicting parameters such as food-to-
microorganisms ratio and sludge age, the use of typical
operating conditions for existing sources, and the control of
new sources with more stringent designs.
Response; The commenter (A-90-19: IV-D-34) provided
only a general set of qualitative guidelines, but provided no
numerical data or suggestions regarding how these general
guidelines could be used in the HON. Neither did the
commenter (A-90-19: IV-D-34) provide information regarding
the HAP emission reductions that would result from
implementing such guidelines. In order to assess the
feasibility of implementing the general guidelines suggested
by the commenter (A-90-19: IV-D-34), the EPA requires data,
as requested in the preamble to the proposed regulation.
Without data to support the general guidelines suggested by
the commenter (A-90-19: IV-D-34), it is not possible for the
EPA to implement these guidelines into the final regulation.
2.2 OTHER CONTROL REQUIREMENTS
Comment; One commenter (A-90-19: IV-D-32) recommended
that the EPA develop a definition of an emissions-suppressed
wastewater collection and treatment system and include the
following system components: individual drains fitted with
s-traps and p-traps; junction boxes with water seals; junction
boxes that are flooded to eliminate flow of air from inlets;
covered drop boxes and lift stations where splashing may
occur; and covered treatment and storage tanks. The commenter
(A-90-19: IV-D-32) provided several figures illustrating the
components of an emissions-suppressed collection system and
included component data in appendix P of the comment letter
that were collected by Enviromega for CMA (1993, Measurement
of Hazardous Air Pollutant Emissions from Drop Structures and
Process Drains, Burlington, Ontario), which indicated that a
suppressed collection system consisting of these components
would decrease HAP emissions from wastewater.
Response; The EPA agrees that a suppressed wastewater
collection and treatment system will decrease organic HAP
2-32
-------
emissions from wastewater. Emissions suppression of the
wastewater collection and treatment system is the result of
applying covers and water seal controls on the individual
components of the wastewater collection and treatment system.
The terms "cover" and "water seal controls" are defined in
§63.111 of the regulation. The term "emissions-suppressed
wastewater collection and treatment system" has not been added
to the final regulation because separate requirements for
individual wastewater collection and treatment components are
detailed in the regulation. These requirements apply only to
those individual wastewater collection and treatment
components which receive, manage, or treat Group 1 wastewater
streams or residuals removed from Group 1 streams. The issue
of flooded sewers is addressed in section 2.2.2 of this BID
volume.
Comment; One commenter (A-90-19: IV-D-64) stated that
in §63.136(c)(1) in the proposed rule the second sentence
should be changed to read, "for each drain using a p-trap or
s-trap, the owner or operator shall maintain a water seal in
the p-trap or s-trap" and the remainder of the paragraph
should be deleted. The commenter (A-90-19: IV-D-64) stated
that the purpose for established operating practices is to
maintain a water level in these traps. The commenter
(A-90-19: IV-D-64) contended that the examples in the
proposed HON are unreasonable and that SOCMI sources should be
responsible for failure to operate equipment properly.
Response; The EPA contends that maintaining water in a
p-trap or s-trap will ensure that a water seal will be
maintained in a p-trap or s-trap, thus preventing emissions to
the atmosphere. Furthermore, for monitoring purposes, an
owner or operator must ensure that water is maintained in the
trap either by visual inspection of the trap or by an
alternative means. Maintaining continuous water flow to the
trap is only one example of how an owner or operator would
ensure that there is water in the trap. The owner or operator
may monitor traps in numerous other ways if continuous flow is
unreasonable for the particular situation.
2S33
-------
One commenter (A-90-19: IV-F-1.2 and IV-F-4)
stated that the EPA has presented no valid data for its claim
that additional environmental benefits can be derived by
requiring an enclosed collection system prior to steam
stripping. The commenter (A-90-19: IV-F-1.2 and IV-F-4)
claimed that the EPA could not justify the additional cost of
using an enclosed collection system and a design steam
stripper instead of using a biological treatment system.
Response; Estimates conducted by the EPA based on
several studies indicate that significant volatilization of
organic HAP's occurs from unenclosed collection systems. The
final wastewater provisions are based on cost-effective
control of HAP emissions from wastewater assuming that the
Group 1 wastewater streams are hard-piped to a steam stripper.
Enclosed individual drain systems are allowed as an
alternative control approach that can be combined with any
treatment device, including biological treatment units, that
meets the required treatment level. A biological treatment
system without an enclosed individual drain system is the
baseline level of control that results in significant HAP
emissions.
Comment; One commenter (A-90-19: IV-D-64) stated that
in §63.138(h)(3)(i), the EPA should not require that every
cover on a treatment process or waste management unit have a
vent. The commenter (A-90-19: IV-D-64) stated that the
paragraph should require that each opening from the treatment
process or waste management unit be covered, and any opening,
treatment process, or management unit that is vented should be
covered.
Response; The EPA disagrees with the commenter (A-90-19:
IV-D-64) that the provisions in §63.138(h)(3)(i) of the
proposed rule be changed. Any treatment process or waste
management unit requiring control that is covered shall also
be vented to a control device to control HAP emissions from
the vapors in the treatment process or waste management unit.
Comment: One commenter (A-90-19: IV-D-73) claimed that
the requirements in §63.136(c)(2) to cover junction boxes, and
2-34
-------
if the junction boxes are vented, to have a vent pipe of
certain dimensions, do not serve any emission control purpose
since this subsection deals with water-sealed drain systems.
One commenter (A-90-23: IV-D-9) claimed that the requirements
to cover and treat emissions from wastewater management tanks
would be costly for biological treatment units and would offer
little emissions benefits. The commenter (A-90-23: IV-D-9)
said that adopting biological treatment as an RCT would
eliminate the requirements to cover these units.
Response; The basis for the commenter's claim is
unclear. Provisions for drains are stated in §63.136(e)(1),
not §63.136(c)(2), as suggested by the commenter. The
provisions for junction boxes are stated separately in
§63.136(e)(2). Only junction boxes receiving Group 1 process
wastewater streams must be covered.
The EPA clarifies that, although the proposed rule and
final rule allow biological treatment to be used, equipment
used to receive, manage, or treat Group 1 process wastewater
streams must meet the provisions of §63.133 through §63.137
for covering the units and venting HAP emissions to a control
device.
A properly operated biological treatment unit which meets
the mass removal requirements of §63.138(b)(1)(iii)(C) or
§63.138(c)(1)(iii)(D) or meets the 95-percent HAP mass
reduction requirements of §63.138(e) need not be covered and
vented to a control device.
comment; One commenter (A-90-23: IV-D-2) suggested that
covering surface impoundments and individual drain systems may
cause fire, explosion, and confined-entry danger. Another
commenter (A-90-19: IV-D-45) stated that organic vapors from
wastewater, which are trapped inside tanks with fixed lids,
may pose a major explosion hazard.
Response; The EPA agrees that covering surface
impoundments may cause fire and explosion danger if the cover
and closed-vent system are incorrectly designed. The EPA also
agrees that the incorrect and unsafe use of fixed roof tanks
for storage or treatment of wastewater may pose the risk of
2-35
-------
fire or explosion. However, the EPA anticipates that
facilities will avoid such unsafe conditions in specifying and
designing these systems. Most individual drain systems
currently exist as subsurface structures with the potential
for explosive vapor buildup. The EPA does not anticipate that
covering individual drain systems and venting them to a
closed-vent control system will import additional risk ol: fire
or explosion greater than that which may currently exist.
Confined-entry areas currently exist in most wastewater
systems. The EPA does not anticipate that covering individual
drain systems and surface impoundments will significantly
increase confined-space entry hazards at typical SOCMI
facilities.
Comment; One commenter (A-90-19: IV-D-92) disagreed
with the requirements to cover surface impoundments claiming
that the covers would be very costly and that this requirement
would produce little environmental benefit. The commenter
(A-90-19: IV-D-92) claimed that it is unlikely that
wastewaters with high concentrations of HAP's would be stored
in surface impoundments, because of RCRA prohibitions of using
surface impoundments to store hazardous waste or hazardous
wastewater.
Response; The EPA clarifies that RCRA does not prohibit
the use of surface impoundments for the storage or treatment
of hazardous wastes, but does require that these surface
impoundments be equipped with a double liner. Emissions
estimates made by the EPA indicate that surface impoundments
are significant emission sources. Therefore, the requirement
to cover surface impoundments receiving Group 1 wastewater
streams is retained in the final rule.
Because RCRA requires surface impoundments which receive
hazardous wastes to be double lined, and requires monitoring
to detect leaks, many facilities have replaced their surface
impoundments with tanks. Similarly, the EPA anticipates that
owners or operators will not place Group 1 wastewater streams
into surface impoundments if alternative devices, such as
2-36
-------
tanks or containers, are more cost effective to collect and/or
treat Group 1 wastewater streams.
One commenter (A-90-23: IV-D-9) claimed that
separate treatment equipment should not be required for each
stream because most sources have a centralized treatment
system.
Response ; The EPA clarifies that separate treatment of
each Group 1 wastewater stream is not required by the HON.
Streams may be combined for more efficient and cost effective
treatment .
2.2.1 Clarification of Requirements for Control Devices
Comment ; One commenter (A-90-19: IV-D-32) stated that
§63.139(h) (1) and (2) in the proposed rule, which require flow
monitoring or a locked valve on bypasses, should be modified
to exempt emergency relief valves from these bypass
requirements .
Response ; The EPA clarifies that emergency relief
devices are not subject to requirements for car seals, locked
valves, and flow monitoring.
Comment : Two commenters (A-90-19: IV-D-32; IV-D-54)
stated that §63 . 139 (c) (2) should be expanded to allow
scrubbers, particularly scrubbers controlling non-halogenated
gas streams, to show compliance through the use of design
analysis as an alternative to performance testing.
Response : The provisions in §63.139 apply to control
devices used to control the organic vapors removed from
Group 1 wastewater streams. In response to comments, the EPA
has added specific language into §63.139 of the final rule
which allows the use of scrubbers as a control device.
Scrubbers could also be used in the proposed rule under the
general allowance in §63. 139(c) (4) , which is paragraph (c) (5)
in the final rule. In the final rule, the EPA specifies that
if a scrubber is used it must achieve 95 percent by weight
destruction of HAP's by chemical reaction with the scrubbing
liquid.
The EPA has added a new paragraph (c) (4) to §63.139 which
reads :
2-37
-------
A scrubber shall reduce the total organic compound
emissions, less methane and ethane, or total organic
HAP emissions in such a manner that 95 weight
percent is destroyed by chemical reaction with the
scrubbing liquid.
Additionally, in §63.139(d)(2)(vii), the EPA has added the
following:
For a scrubber, the design evaluation shall consider
the vent stream composition; constituent
concentrations; liquid-to-vapor ratio; scrubbing
liquid flow rate and composition; temperature; and
the reaction kinetics of the constituents with the
scrubbing liquid. The design evaluation shall
establish the design exhaust vent stream organic
compound concentration level and will include the
additional information in paragraph (d)(2)(vii)(A)
of this section for a tray column scrubber or
paragraph (d)(2)(vii)(B) of this section for a
packed column scrubber.
(A) Type and total number of theoretical and actual
trays;
(B) Type and total surface area of packing for entire
column, and for individual packed sections if column
contains more than one packed section.
This language parallels the requirements of §63.139(d)(2)(iv),
(v), and (vi) for the other control devices.
Comment; One commenter (A-90-19: IV-D-54) stated that
emission control devices installed to comply with the
provisions for closed-vent systems and control devices
(§63.139) and that are located upstream of an RCT (e.g.,
flare) device should not be subject to testing and monitoring
requirements. The commenter (A-90-19: IV-D-54) contended
that for controls in series, only testing at the outlet of the
train of the control system should be required, and only
monitoring which is necessary to ensure performance of the
overall train of control systems should be required.
Response; The requirements of §63.139 are not intended
to mean that a separate performance demonstration is required
for each individual control device operated in a series. A
facility must comply with only one of the four paragraphs
under §63.139(c) of the final rule. If compliance is achieved
2-38
-------
with one of the control devices in a series, then compliance
need only be demonstrated for that one device.
Another option would be to demonstrate the reduction of
the total organic compound emissions, less methane and ethane,
or total organic HAP emissions by 95 weight percent or
greater. This could be done across a single control device or
across a series of control devices. However, control devices
in series, up to and including the control devices which
achieves compliance with §63.139(c), are subject to
§63.139(d), (e), and (f). Any control device or series of
control devices located after the control device where the
owner or operator demonstrates compliance is not subject to
either §63.139 or the inspection and monitoring requirements
in §63.143 because such a control device or series of control
devices achieves reductions in excess of the requirements of
§63.139(c).
The owner or operator need only monitor those control
devices in series which are used to comply with §63.139. As
an option to show compliance with §63.139, the owner or
operator may install an organic monitoring device at the
outlet of the control device in accordance with §63.143(e)(2).
The owner or operator may also request approval from the
implementing agency per §63.143(e)(3) to monitor parameters
other than those specified in §63.143(e)(1) or (2).
The EPA clarifies that regardless of how the owner or
operator chooses to comply with the monitoring requirements in
§63.143, the owner or operator must establish, for each
parameter monitored, a range that indicates proper operation
of the closed-vent system.
2.2.2 Water Seal Controls
Comment; One commenter (A-90-23: IV-D-14) claimed that
the HON definition of "water seal controls" is different than
the definition in the Benzene Waste NESHAP. The commenter
(A-90-23: IV-D-14) recommended adding to the HON definition
the specifications from the Benzene Waste NESHAP, which state
that "the water level of the seal must be maintained in the
2-39
-------
vertical leg of a drain in order to be considered a water
seal."
Response: The EPA has changed the proposed definition of
"water seal controls" in §63.111 of subpart G of the final
rule. The EPA agrees w^ith the conunenter that the definition
in the Benzene Waste NESHAP and the HON should be the same,
and has therefore amended definition as follows:
"Water seal controls means a seal pot, p-leg trap,
or other type of trap filled with water
(e.g., flooded sewers that maintain water levels
adequate to prevent air flow through the system)
that creates a water barrier between the water level
of the seal and the atmosphere. The water level of
the seal must be maintained in the vertical leg of a
drain in order to be considered a water seal."
The objective of the controls specified for drains and
junction boxes in an individual drain system is to isolate
them such that the free flow of vapors within the system is
prevented. By including additional examples in the final
rule, the EPA has clarified that other types of water seals
such as flooded sewers also are acceptable.
2.2.3 Definition of "Cover"
Comment; Two commenters (A-90-19: IV-D-90; IV-D-100)
requested that the EPA include a definition of "cover" in the
wastewater provisions.
Response: The EPA has added the following definition of
"cover" to §63.111 of subpart G to clarify the term as it is
used throughout the wastewater provisions:
Cover. as used in the wastewater provisions, means a
device or system which is placed on or over a waste
management unit containing wastewater or residuals
so that the entire surface area is enclosed and
sealed to minimize air emissions. A cover may have
openings necessary for operation, inspection, and
maintenance of the waste management unit such as
access hatches, sampling ports, and gauge wells
provided that each opening is closed and sealed when
not in use. Examples of covers include a fixed roof
installed on a wastewater tank, a lid installed on a
container, and an air-supported enclosure installed
over a waste management unit.
2-40
-------
2.2.4 Submerged Fill Pipes
Commenti Two commenters (A-90-19: IV-D-86; IV-D-32)
stated that the proposed provisions in §§63.I35(c)(1) and (2),
which require the use of submerged fill pipes for filling a
container with residuals should be deleted because wastewater
residuals such as sludge may clog the outlet of the fill pipe.
Response; The EPA agrees that the use of submerged fill
pipes for viscous materials may be difficult because the thick
material may clog the pipe. In developing the final rule, the
EPA reconsidered the use of submerged fill and has changed the
requirements. Submerged filing is required for containers
with a capacity of 0.42 m3 or greater that are filled by
pumping the Group X wastewater or residual. The HAP emissions
generated by filling containers warrants the use of submerged
fill pipes.
Commenti One commenter (A-90-19: IV-D-32) contended
that the proposed emission control requirement for containers
in §§63.135(d)(1) through (3), which requires that treatment
in a container (including aeration, thermal, or other
treatment) be conducted within an enclosure with a closed-vent
system that is routed to a control device, discourages
treatment in containers. The commenter (A-90-19: IV-D-32)
stated that treatment in containers provides environmental and
safety benefits with little potential for emissions. The
commenter (A-90-19: IV-D-32) stated that the proposed control
requirement will complicate such management methods to the
point that some facilities may decide to omit the treatment
step. The commenter (A-90-19: IV-D-32) suggested that if
this requirement is retained in the rule, it should apply only
to treatment in containers that is shown to cause significant
HAP emissions.
Response; The EPA maintains that whenever it is
necessary for a container to be open, treatment in a container
of a Group 1 wastewater stream or residual removed from a
Group 1 wastewater stream, must be conducted within an
enclosure with a closed-vent system that is routed to a
control device. If the container is not open and cannot emit
2-41
-------
HAP's, then the container is not required to be within an
enclosure with a closed-vent system that is routed to a
control device. The commenter provided no data to
substantiate their statement that treatment in containers
provides little potential for emissions.
Commentt One commenter (A-90-23: IV-D-18) opposed the
EPA's proposal to require the filling of tanks and containers
using a submerged fill pipe that must extend within two pipe
diameters of the bottom of the vessel being filled. The
commenter (A-90-23: IV-D-18) has developed corporate-wide
mechanical piping standards that require a distance of
6 inches to prevent undue wear to the vessel during filling
using a 2-inch submerged pipe. In addition, the commenter
(A-90-19: IV-D-18) stated that a submerged fill pipe that is
too close to the bottom of a tank will impart a sideways force
on the pipe, deflecting the fill pipe sideways with a
potential of the pipe breaking off due to this movement. The
commenter (A-90-23: IV-D-18) requested that the EPA allow
three fill pipe diameters instead of two.
Response; The EPA clarifies that the filling of tanks
does not require the use of a submerged fill pipe. The EPA
has re-evaluated the requirement that submerged fill pipes
must be located within two pipe diameters of the bottom of the
vessel being filled. In the final rule, the EPA is allowing
submerged fill pipes to be located no more than 6 inches or
two pipe diameters from the bottom of the container and has
deleted the requirement for submerged fill for containers less
than 0.42 m3. This will not increase HAP emissions but will
provide greater flexibility for industry compliance.
2.2.5 Maintenance Wastewater
Comment; One commenter (A-90-19: IV-D-34) stated that
the wastewaters from routine maintenance do not result in
significant HAP emissions. The commenter (A-90-19: IV-D-34)
stated that one facility roughly estimates that less than
0.05 Mg/yr is lost to the process sewer from pump maintenance.
The commenter (A-90-19: IV-D-34) recommended deleting the!
2-42
-------
requirement to control maintenance-turnaround and routine
maintenance wastewaters.
Response: It is difficult for the EPA to assess the
commenter's data. The commenter who provided an estimate of
maintenance wastewater emissions from pump maintenance for
only one facility did not provide documentation of the
estimate and did not provide an estimate of emissions from
other maintenance activities.
The EPA has made a change from the proposed rule so that
requirements for routine maintenance and maintenance-
turnaround wastewaters are now addressed in the facility's
start-up, shutdown, and malfunction plan as was proposed for
only maintenance-turnaround wastewaters. Given the
variability in maintenance wastewaters and the difficulty in
measuring their flow rates and concentrations, the EPA has
determined that it is more appropriate for individual
facilities to determine site-specific housekeeping procedures
to properly manage maintenance wastewater and control organic
HAP emissions to the atmosphere from maintenance wastewaters.
The requirements to collect and manage routine maintenance
wastewaters in a controlled drain system have been eliminated.
Comment; One commenter (A-90-19: IV-D-77) stated that
the EPA should include provisions allowing maintenance-related
wastewater to bypass the control devices (e.g., design steam
stripper) because variable feed composition and dissolved and
suspended solids create operational and maintenance problems.
The commenter (A-90-19: IV-D-77) suggested that the EPA
incorporate an exclusion for five-percent downtime to allow
for maintenance of the steam stripper.
Response; The HON does not require maintenance
wastewaters to be treated in control devices. The rule only
requires that owners or operators include a description of
procedures in their start-up, shutdown, and malfunction plan
that, when followed, ensure that maintenance wastewaters are
properly managed and HAP emissions are minimized. Process
wastewater will sometimes have variable feed compositions and
dissolved and suspended solids also. Therefore, it is
2-43
-------
reasonable to assume that facilities will have equipment such
as feed tanks and filters already in place to account for
composition variation and solids in process wastewater
streams. Therefore, if the owner or operator chooses to do
so, this same equipment may be used for maintenance wastewater
streams.
Furthermore, the wastewater provisions do not require the
design steam stripper to operate continuously. If a steam
stripper requires repair, wastewaters can be collected in a
hold tank and routed to the steam stripper once the repairs
are complete. The wastewaters cannot bypass the steam
stripper during the repair period.
Comment; Two commenters (A-90-19: IV-D-97) (A-90-23:
IV-D-20) recommended that the EPA specify one type of
maintenance wastewater, just as there is only one type of
process wastewater, and address the management of all
maintenance-generated wastewater in each facility's start-
up/malfunction plan per §63.102(b)(1)(i) of the proposed rule.
Two commenters (A-90-19: IV-D-32; IV-D-75) objected to the
requirement that routine maintenance wastewater be collected
in a closed system. One commenter (A-90-19: IV-D-75) claimed
that the requirement is inconsistent with the Benzene Waste
NESHAP which allows the owner or operator to determine if
control of these wastewaters is required. Two commenters
(A-90-19: IV-D-97) (A-90-23: IV-D-20) stated that
maintenance-related wastewater should not be regulated in the
same manner as process wastewater (i.e., Group I/Group 2
determination).
One commenter (A-90-19: IV-D-34) stated that the EPA
should clarify one of the management options for maintenance
wastewater in §63.102(b)(2)(ii) of the proposed rule, which
specifies that the maintenance wastewater can be collected and
managed in a controlled drain system. The commenter (A-90-19:.
IV-D-34) expressed concern that because the EPA has not
defined a "controlled drain system," requirements for managing
maintenance wastewater could be interpreted to mean that
maintenance wastewater must be collected and managed in a
2-44
-------
drain system that meets the requirements of §63.133 through
§63.140. The commenter (A-90-19: IV-D-34) contended that the
preamble to the proposed rule (57 FR 62677-8), which states
that routine maintenance wastewater will be controlled using
general procedures contained in a start-up, shut-down, and
malfunction plan, does not seem to be consistent with the
possible interpretations of "controlled drain system." The
commenter (A-90-19: IV-D-34) requested that the EPA restate
the rule to be consistent with the concepts discussed in the
preamble.
One commenter (A-90-19: IV-D-73) recommended that
routine maintenance wastewaters and wastewaters generated
during shutdown be subject to the same requirements and also
suggested dealing with maintenance and shutdown wastewaters
using one site-specific plan. The commenter (A-90-19:
IV-D-73) claimed that volume and hydrocarbon content need to
be considered before controlling maintenance wastewater and
that the requirement for routine maintenance wastewater to be
collected and recycled, destroyed, or collected and managed in
a controlled drain is not related to developing a maintenance
wastewater plan. One commenter (A-90-19: IV-D-33) expressed
concern that in §63.102(b)(1)(ii) of the proposed rule, which
requires that routine maintenance wastewaters are either
collected and recycled or are destroyed or are collected and
managed in a controlled drain system, seems to require a
special procedure and system to manage such wastewater. The
commenter (A-90-19: IV-D-33) stated that most wastewaters,
both routinely and non-routinely generated, are handled
through the same systems pursuant to CWA requirements and that
the EPA should not require special systems to handle only
maintenance wastewater subject to the HON.
Response; The requirements for routine maintenance
wastewaters have been revised. Routine maintenance wastewater
and maintenance-turnaround wastewater as defined in §63.101 of
the proposed rule will now both be referred to as "maintenance
wastewater" as defined in §63.101 of subpart F. Both types of
maintenance wastewater are now subject to the requirements
2-45
-------
proposed for maintenance-turnaround wastewater. The
requirements for all maintenance wastewaters are addressed in
the facility's start-up, shutdown, and malfunction plan.
Routine maintenance wastewaters are no longer required to be
collected and recycled, destroyed, or collected and managed in
a controlled drain system as specified in §63.102 (b) (1) (i.L) of
the proposed rule. The owner or operator must only specify
the procedures that will be followed to properly manage
maintenance wastewater and minimize HAP emissions from
maintenance wastewater. All maintenance wastewater can be
handled in the same sewer systems.
The Benzene Waste NESHAP requires control of wastewaters
with a concentration greater than 10 ppmw if the facility
total annual benzene (TAB) is 10 megagrams per year or
greater. The requirements do not allow the owner or operator
to determine if control of maintenance wastewater is required.
It is assumed that the commenter means routine
maintenance wastewater when referring to maintenance-related
wastewater. As stated above, routine maintenance wastewater
is not subject to the same requirements as process wastewater.
Comment; One commenter (A-90-23: IV-D-20) stated that
compliance with an NPDES permit for a SOCMI facility should be
sufficient to treat maintenance-related wastewater.
Therefore, the commenter (A-90-23: IV-D-20) recommended that
requirements for all maintenance-related wastewater streams be
eliminated from the HON.
Response; An NPDES permit only specifies the amount of
organics that may be present in the wastewater before it is
discharged from the facility. These permits do not limit the
air emissions that can be released from wastewater prior to
their discharge from the facility. Therefore, compliance with
s
an NPDES permit is not sufficient to reduce air emissions from
maintenance wastewater. The HON requirements for maintenance
wastewater are now listed in §63.105 of subpart F and ensure
that maintenance wastewater will be properly managed and HAP
emissions to the air from these wastewaters will be
controlled.
2-46
-------
2.2.6 Control of Steam Stripper Overheads
one commenter (A-90-19: IV-D-85) stated that
the control device at the end of the process should be
required to meet at least a 98-percent reduction standard, and
that the EPA should conduct an analysis assuming 98 percent
control of stripped organics.
Response ; The EPA assumes that the commenter means the
overheads from the steam stripper when referring to "the end
of the process.11 The EPA requires that the HAP emissions from
the steam stripper overheads primary condenser to be reduced
by 95 percent. The EPA allows the use of recovery devices
such as secondary condensers and carbon adsorbers to recover
the overheads from the steam stripper, and these devices may
not be able to achieve a 98 percent reduction. Recovery
devices typically achieve removal efficiencies of 95 percent,
and therefore, combustion would be required to achieve removal
efficiencies of 98 percent for many compounds. Requiring a
9 8 -percent control of stripped organics would discourage
resource recovery, because many of the overhead streams would
have to be treated by combustion.
2-47
-------
3.0 IMPACTS ANALYSIS
3.1 COST ANALYSIS
Comment: One commenter (A-90-19: IV-D-85) supported the
EPA's cost analysis for justifying steam stripping as RCT and
stated that the EPA justified the cost of steam stripping in
1987 while establishing effluent limitations for OCPSF
regulations (52 FR 42561). The commenter (A-90-19: IV-D-85)
stated that if further cost comparisons are made between steam
stripping and biological treatment, section 112(d) of the
Act requires the EPA to consider both the air and water
quality benefits that could be achieved by each treatment
technology.
Response; The EPA based effluent limitations and
compliance costs for OCPSF regulations on steam stripping with
product recovery and justified the cost of steam stripping in
the OCPSF regulation. The EPA has determined that the cost
estimate in the RON represents the true cost of installing and
operating a steam stripper. The cost impacts for controlling
wastewater for the HON are based on steam stripping and are
presented in section IV.C of the preamble to the final rule.
If further cost comparisons are made between biological
treatment and steam stripping, the EPA will consider air and
water quality benefits, along with energy impacts, NOX
emissions, CO emissions, and solid waste generation.
Comment; One commenter (A-90-23: IV-D-17) claimed that
the EPA's estimate of TCI for a steam stripper was several
orders of magnitude too low. The commenter (A-90-23:
IV-D-17) provided an attachment indicating that the TCI for a
steam stripper is $3,456,200, and listed several reasons why
the EPA's estimate of TCI is lower.
3-1
-------
Response; The EPA's estimate of the steam stripper TCI
was based on published data and vendor information. The TCI
is composed of the EEC, the PEC, and direct and indirect
installation costs. The BEC that was estimated by the EPA
differs from the estimate provided by the commenter by less
than 20 percent. However, the PEC and the indirect and direct
installation costs estimated by the commenter are 44 to 91
percent larger than those estimated by the EPA. According to
the Office of Air Quality Planning and Standards Control Cost
Manual (OCCM), the PEC and the direct and indirect
installation costs are based on the BEC. In the EPA's
estimate of TCI, the components of the PEC, and the direct and
indirect installation costs are represented as a percentage of
the BEC as published in the OCCM. The estimates from the
OCCM are accurate to within ±30 percent. The commenter's
estimates of PEC and direct and indirect installation costs do
not agree with the guidelines presented in the OCCM. The
commenter listed significantly larger engineering costs and
direct installation costs, and these estimates are not
representative of typical costs. The commenter (A-90-23:
IV-D-17) did not provide support for these higher costs.
Comment; Two commenters (A-90-19: IV-D-77; IV-D-110)
stated that the EPA should account for the costs associated
with adding a steam stripper to an existing facility with
limited sp." 7e or requiring installation in a remote location.
Response: When developing regulatory options, the EPA
must consider impacts on a nationwide basis. Therefore, cost-
effectiveness for control of wastewater for the regulatory
option chosen ($495/ton for new and existing sources in the
fifth year) represents a nationwide estimate. Estimates a.re
based on control equipment arrangements that are most common
in the industry. Most facilities using a steam stripper to
control wastewater streams will have adequate space and will
not have to install the steam stripper in a remote location.
Therefore, these types of facility-specific cost
considerations were not accounted for in the HON impacts
analysis. If the use of a steam stripper is not cost-
3-2
-------
effective for a particular facility, §63.138 includes several
other alternatives the facility can use to comply with the
regulation, including biological treatment.
Comment; One commenter (A-90-19: IV-D-98) stated that
the costs associated with enclosing a wastewater treatment
plant to avoid installing expensive internal piping and
process changes is not a cost-effective option for facilities
with multi-acre wastewater treatment lagoons.
Response: The impacts analysis done by the EPA indicated
that control of Group 1 wastewater streams using steam
stripping is cost-effective. Therefore, facilities have the
option to use steam stripping for control of Group 1
wastewater streams. Covering smaller lagoons or surface
impoundments may also be cost-effective for some facilities.
However, covering larger wastewater lagoons or other surface
impoundments may not be cost-effective for every facility, and
most facilities will not opt to cover multi-acre wastewater
lagoons. Such facilities can select another one of the
compliance options described in §63.138 of the wastewater
provisions in subpart G.
Comment: Two commenters (A-90-19: IV-D-32; IV-D-62)
stated that biological treatment could be a. more cost
effective option than the steam stripper for biodegradable
HAP's. One commenter (A-90-23: IV-D-9) stated that the cost
of biological treatment was less than the installation cost of
steam stripping for a wide variety of waste streams with low
HAP concentrations. The commenter (A-90-23: IV-D-9) claimed
that many facilities have invested in pre-treatment facilities
for biological treatment units that will need to be abandoned
at a great economic loss. The commenter (A-90-23: IV-D-9)
stated that steam strippers would offer no advantage over the
existing systems in emission reduction.
Response: The EPA allows the use of biological treatment
as a compliance option for treating wastewater streams when
the biological treatment unit can achieve removal efficiencies
at least as high as steam stripping. Therefore, a facility
treating a wastewater stream containing highly biodegradable
3-3
-------
HAP's has the option to use either biological treatment or
steam stripping, whichever is more cost-effective. Facilities
treating wastewater streams with low HAP concentrations also
have the option to use biological treatment if the biological
treatment unit can achieve removal efficiencies at least as
high as steam stripping. For those wastewater streams where
biological treatment cannot achieve removal efficiencies
equivalent to steam stripping, the EPA has shown that the
installation of a steam stripper is cost-effective.
Comment: One commenter (A-90-19: IV-D-98) concluded
that the energy costs associated with wastewater controls
appear to be underestimated by at least a fa: ?r of 10. The
commenter (A-90-19: IV-D-98) stated that the 2PA assumes a
national energy impact of 5,300xl09 Btu/yr based on the use of
steam stripping in the proposed rule. The commenter (A-90-19:
IV-D-98) contended that if only 300 facilities are subject to
the HON, the energy estimate would be sufficient to treat only
5 gal/minute/site or less at an expenditure of $50,000 to
$100,000 per year.
Response; Based on the options chosen for new and
existing sources of wastewater, it was estimated in the
proposal preamble that approximately 8,000 liters per minute
of wastewater would be controlled at new sources and
approximately 27,000 £pm of wastewater would be controlled at
existing sources. It was also estimated that 89 wastewater
streams would be controlled at new sources, and 127 wastewater
streams would be controlled at existing sources. It was
estimated, based on the enthalpy f water and steam, that the
energy required to produce steam jr use in the steam stripper
is 1.46X108 Btu/year/£pm of wastewater treated (Memorandum
from Chuck Zukor, Radian, to Penny Lassiter, EPA/CPB,
"Development of Secondary Environmental Impact Factors for
Steam Stripping Wastewater Streams in the HON," January 31,
1992). Therefore, the total energy required for new and
existing sources for the final rule to steam strip wastewater
is approximately 5,100xl09 Btu/year.
3-4
-------
In the cost analysis done by the EPA, it was determined
that steam stripping is cost-effective for the treatment of
SOCMI wastewater streams. The cost of the steam required by
the steam stripper was included in the calculation of annual
cost and was estimated to be $9.26 per megagram (Memorandum
from Chris Bagley, Radian, to Mary Tom Kissell, EPA/SDB,
"Steam Costs," August 23, 1993). Assuming a heat content of
1,206 Btu per pound, approximately 4.23X109 pounds per year of
steam are required to steam strip the wastewater streams
subject to the HON. This equates to a steam cost of
17.8 million dollars for all facilities subject to the
wastewater provisions of the HON.
It is unclear where the commenter (A-90-19: IV-D-98)
obtained the cited data. The commenter's estimate of the
number of facilities affected by the wastewater provisions of
the HON does not agree with the EPA's estimate. However, a
steam cost of $50,000 to $100,000 per facility per year for
300 facilities equates to a cost of 15 to 30 million dollars
per year for all affected facilities which agrees with the
EPA's, estimate.
Comment: One commenter (A-90-19: IV-D-85) claimed that
the EPA did not consider the cost savings of all the
wastewater options when determining cost-effectiveness
exemptions. The commenter (A-90-19: IV-D-85) stated that the
EPA should consider the cost savings from pollution prevention
and from routing multiple wastewater streams to a single steam
stripper.
Response; The EPA did consider.the cost savings from
combining wastewater streams and routing them to a single
steam stripper, because the EPA's cost analysis assumed one
steam stripper for each SOCMI facility. If an owner or
operator chooses to use pollution prevention techniques, a
wastewater stream will not be generated and the facility will
not have a control cost. A cost-effectiveness analysis was
done only for those facilities that would be required to apply
additional controls based on the applicability criteria in the
wastewater provisions.
3-5
-------
3.1.1 Recycling vs. Disposal of Residuals
Comment: Several commenters (A-90-19: IV-D-32; IV-D-77;
IV-D-110) (A-90-23: IV-D-20) stated that the EPA
underestimated the cost for steam stripping by incorrectly
assuming that SOCMI facilities could recycle HAP's that are
collected in the overhead. One commenter (A-90-23: IV-D-17)
disagreed with the recovery credit that the EPA included in
its estimate of TAG. The commenter (A-90-23: IV-D-17)
claimed that the organics in wastewater cannot be reused
without further processing because contaminants can interfere
with the process.
One commenter (A-90-23: IV-D-17) alleged that TAG may be
higher than the EPA has estimated because steam stripper
overheads may not be able to be incinerated onsite and may
have to be handled as a hazardous waste. Two commenters
(A-90-19: IV-D-32; IV-D-77) contended that the disposal
costs for residuals may be a significant fraction of total
annual operating costs. One commenter (A-90-19: IV-D-110)
stated that the EPA should account for the cost of disposing
of residuals offsite. Two commenters (A-90-19: IV-D-32)
(A-90-23: IV-D-20) stated that the EPA needs to include in
its cost analyses the management of the aqueous-phase waste
generated by the decanter.
Response: The cost analysis for the proposed rule did
not assume that HAP's recovered from the steam stripper
overheads were recycled to the process. Rather, the cost
analysis for the proposed rule assumed that the recovered
VOHAP's are incinerated in a boiler, and thus generate a fuel
credit.
The credit for incinerating recovered HAP overheads was
eliminated from the total annual cost of the steam stripper
for the final rule. However, the annual cost of the steam
stripper was not underestimated because the fuel credit only
represented approximately 3 percent of the total annual cost.
Therefore, eliminating the fuel credit did not greatly affect
the total annual cost.
3-6
-------
Information submitted to the EPA by the CMA indicated
that the residuals from the operation of a steam stripper are
managed in one of three ways:
(1) by on-site incineration;
(2) by off-site incineration; or
(3) by recycling to the process.
For simplicity, the CMA suggested that the EPA assume
that these three methods are used in equal proportion. Both
on-site incineration and recycling to the process generate a
fuel or raw material credit for the facility. Off-site
incineration generates a waste disposal cost for the facility.
It is assumed that the waste disposal cost and fuel and raw
material credits cancel each other, so that residuals disposal
results in a net cost of zero.
3.1.2 Carbon Steel vs. Stainless Steel
Comment; Two commenters (A-90-19: IV-D-32; IV-D-110)
stated that the EPA underestimated the cost of the design
steam stripper by using carbon steel as the primary
construction material. One commenter (A-90-19: IV-D-32)
stated that the EPA should have based the cost on stainless
steel.
Response; Based on comments and new information, the EPA
has revised the cost of the steam stripper (Memorandum from
Kristine Pelt, Radian, to Mary Tom Kissell, EPA/SDB, "Steam
Stripper Total Capital Investment and Total Annual Costs,"
December 1, 1993). The final nationwide impacts are based on
the revised costs estimate. The costs of the steam stripping
column and trays, the primary condenser, the overheads
collection decanter, and the pumps are based on stainless
steel construction. The feed preheater cost is based on a
carbon steel shell with copper tubing. The cost of the
wastewater feed storage tanks was based on carbon steel
construction, because these tanks would not contain any
materials that would require stainless steel construction
(e.g., steam or water at elevated temperatures). Therefore,
carbon steel is an adequate material of construction for the
feed tanks.
3-7
-------
3.1.3 Heat Transfer Coefficient and Heat Exchange System
Comment; Two commenters (A-90-19: IV-D-32) (A-90-23:
IV-D-20) stated that the EPA should reevaluate the heat
transfer coefficient of 180 Btu/hr per square foot per
degree F (Btu/hr • ft2 • °F) for the feed preheater, because
the EPA's estimate is too high for use with a shell and tube
heat exchanger in aqueous-to-aqueous service. The commenters
(A-90-19: IV-D-32) (A-90-23: IV-D-20) also recommended the
use of a piate-and-frame heat exchanger, instead of the
proposed shell-and-tube exchanger because the latter is only
appropriate for wastewaters with no suspended solids. One
commenter (A-90-19: IV-D-32) recommended that the EPA re--
evaluate the effect that such changes may have on cost.
Response; One heat exchange system vendor contacted by
the EPA suggested that a heat transfer coefficient of
approximately 180 Btu/hr • ft2 • °F is low unless there is a
large amount of fouling. Furthermore, the heat transfer
coefficient used by the EPA in the preliminary design
calculation for the preheater is slightly less than the range
of values recommended by accepted references (M.S. Peters and
K.D. Timmerhaus, Plant Design and Economics for Chemical
Engineers, 3rd ed., McGraw-Hill Book Co., 1980). Therefore,
the information received by the EPA contradicts the
information provided by the commenter (A-90-19: IV-D-32),
Based on the vendor information and the references cited, the
EPA has concluded that the value the EPA used for the heat
transfer coefficient is not too high.
According to an article in Chemical Engineering Magaj:ine,
shell and tube heat exchangers are less likely to be clogged
by particulate matter than plate and frame heat exchangers; (3.
Boyer and G. Trumpfheller, "Specification Tips to Maximize
Heat Transfer," Chemical Engineering, May 1993). Therefore,
a shell and tube design is a reasonable basis for the
preheater cost.
3.1.4 Use of "Temporary" Tanks
Comment; One commenter (A-90-23: IV-D-20) stated that
the EPA should not require the addition of a control
3-8
-------
technology (e.g., a floating-roof tank) for storage of a
wastewater stream during occasional shutdowns of a wastewater
treatment unit if the wastewater is hardpiped directly to the
treatment unit. The commenter (A-90-23: IV-D-20) provided an
example and stated that the cost for compliance with such a
provision would be excessive, and should not be imposed. The
commenter (A-90-23: IV-D-20) suggested that storage of
streams for 14 days or less in temporary storage vessels or
vessels that are usually not used for the storage of
wastewater should not be subject to control requirements when
a wastewater treatment unit is nonfunctional. The commenter
(A-90-23: IV-D-20) stated that the facility uses an open pond
for storage of wastewater in these situations.
Response; The control cost for wastewater is based on
steam stripping, and the EPA assumed in the cost analysis that
the facility would install holding tanks upstream of the steam
stripper and not hard pipe the wastewater directly to the
steam stripper. These holding tanks would be available for
temporary storage of wastewater if the steam stripper needed
repair, and do not present an excessive cost to the facility.
The wastewater stream cannot bypass control during shutdowns
of the treatment unit. The wastewater must either be stored
until the treatment unit is functional or routed to an
alternate treatment unit. It is inherent in the startup,
shutdown, and malfunction plan required under §63.6(e)(3) of
subpart A of this part that when wastewater is stored in a
"temporary" tank because a wastewater tank or treatment unit
is non-functional, the "temporary" tank is uncontrolled. The
startup, shutdown, and malfunction plan requires repair of
wastewater tanks and control equipment as soon as technically
feasible because "temporary" tanks are uncontrolled.
3.1.5 Cost of RCRA Permitting
Comment; One commenter (A-90-19: IV-D-98) stated that
the EPA should consider the cost of developing and obtaining a
BIF permit under 40 CFR part 266 of RCRA in order for SOCMI
facilities to incinerate residuals in high-temperature
combustion devices.
3-9
-------
Response: Information submitted to the EPA indicated
that residuals from the operation of a steam stripper are
managed in one of three ways: (l) by on-site incineration;
(2) by off-site incineration; (3) or by recycling to the
process. Therefore, if the cost of on-site incineration is
high due to the cost of obtaining a RCRA permit, the facility
has two other options for disposing of residuals.
3.2 EMISSION ESTIMATES
Comment; Several commenters (A-90-19: IV-F-1.2 and
IV-F-4; IV-D-32; IV-D-75; IV-D-77; IV-D-58; IV-D-108)
(A-90-23: IV-D-9) stated that some compounds (e.g., methanol)
in the list of volatile organic HAP's are non-volatile or
semi-volatile and are not likely to be emitted during normal
wastewater handling and treatment. One commenter (A-90-19:
IV-D-97) recommended that the wastewater portion of the HON
be limited to only significant streams of truly volatile
compounds and that monitoring, recordkeeping, and reporting be
minimal considering the small amount of emissions that will be
controlled.
Response: The EPA has reviewed its estimates of the
volatility of the HAP's subject to the wastewater provisions,
which are listed in table 9 of subpart G. Based on this
analysis, the following seven compounds have been dropped from
the list of HAP's on table 9 of subpart G:
• 2-Chloroacetophenone (532274)
• Aniline (62533)
o-Cresol (95487)
• 3,3'-Dimethylbenzidine (119937)
Diethylene glycol diethylether (112367)
• Diethylene glycol dimethylether (111966)
• Ethylene glycol monoethylether acetate (111159)
The EPA's analysis has shown that the remaining 76 compounds,
including methanol, are volatile and can potentially be
emitted during wastewater handling and treatment.
3-10
-------
It is assumed that by "significant streams," the
commenter (A-90-19: IV-D-97) means wastewater streams with
significant flow rates and significant concentrations. There
are flow rate and concentration criteria in the wastewater
provisions to ensure that "insignificant" streams will not be
subject to the control requirements in the wastewater
provisions (i.e., the "insignificant" wastewater streams will
not meet the definition of wastewater, or will be Group 2
wastewater). Furthermore, monitoring, recordkeeping, and
reporting is only required for wastewater streams that contain
volatile organic HAP's and meet the flow rate and
concentration criteria. Monitoring, recordkeeping, and
reporting are not required for wastewater streams not meeting
the definition of wastewater. These are the wastewater
streams with the highest emission potential.
Comment: One commenter (A-90-19: IV-D-85) supported the
wastewater provisions of the HON, but stated that the EPA may
have underestimated the proportion of emissions from
wastewater in the SOCMI.
Response; The commenter did not provide any detail
regarding reasons why the proportion of organic HAP emissions
from wastewater in the SOCMI may have been underestimated.
The wastewater emission estimates are based on information
obtained from the SOCMI via a section 114 survey and from
public comment. Therefore, the EPA maintains that the
wastewater emission estimates are representative of the SOCMI.
Comment; Two commenters (A-90-19: IV-F-1.2 and IV-F-4;
IV-D-97) stated that the EPA has overestimated total emissions
from wastewater operations by including non-volatile and
semi-volatile compounds in baseline emission estimates, and by
estimating emission reductions from control based on these
substances. One commenter (A-90-19: IV-F-1.2 and IV-F-4)
asserted that, by incorrectly estimating the removal
efficiency of certain compounds and including insignificant
wastewater streams in the regulation, the EPA overestimated
total emissions from wastewater.
3-11
-------
Response: The list of HAP's that is subject to the
wastewater provisions of the HON has been revised to include
only those HAP's that volatilize from wastewater. Seven
compounds have been dropped from the list of HAP's shown in
table 9 of subpart G of the proposed rule. The EPA's analysis
indicated that the remaining 76 compounds are volatile and can
potentially be emitted from wastewater. Baseline emissions
and emissions reduction estimates are not greatly affected by
including semi-volatile and non-volatile compounds. The EPA
calculated baseline emissions using the fraction emitted (Fe)
values for each compound and calculated the emission reduction
using the fraction removed (Fr) values for each compound.
Most of the baseline emissions and emission reductions are
generated by the highly volatile compounds (those with the
largest Fe and Fr values) that are readily emitted from
wastewater during handling operations and that are readily
removed from wastewater during -reatment operations.
The EPA has revised the estimates for Fe values for the
HAP's listed in table 9 of subpart G and has included these
values in table 34 of subpart G of the final rule. This
analysis idicated that some of the Fe values increased and
some of cne Fe values decreased. Although compound-specific
emissions may change, the total baseline emissions from
wastewater would not change. Furthermore, "insignificant"
streams will not greatly affect the magnitude of baseline
emissions from wastewater, because these streams have low flow
rates, low concentrations, or contain low volatility
compounds.
Comment; Several commenters (A-90-19: IV-D-32; IV-D-34;
IV-D-53; IV-D-54; IV-D-77; IV-D-97; IV-D-110; IV-D-112)
(A-90-23: IV-D-20) claimed that the requirements for
maintenance wastewater and maintenance-turnaround wastewater
are "resource-intensive" compared to the significance of the
emissions from these sources. The commenters (A-90-19:
IV-D-32; IV-D-34; IV-D-53; IV-D-110; IV-D-112) stated that the
EPA has not done an emissions analysis of maintenance and
maintenance-turnaround wastewater and has not shown if
3-12
-------
emissions from these wastewaters are significant. One
commenter (A-90-19: IV-D-34) stated that the EPA has not
complied with the.requirements of §112(d) of the Act for
maintenance wastewater, which specifies that the EPA must
provide data on emissions, the floor, cost, and environmental
impacts. The commenter (A-90-19: IV-D-53) stated that all
maintenance and maintenance-turnaround wastewaters should be
classified as maintenance wastewaters and exempted from the
HON.
One commenter (A-90-23: IV-D-17) favored having a
de minimis level for maintenance wastewater. The commenter
(A-90-23: IV-D-17) claimed that the de minimis level should
be higher than the 2 Mg/yr level in the Benzene Waste NESHAP
because several HAP's may be present in the wastewater.
Response: In the final RON, the EPA continues to
regulate maintenance wastewater in a facility's start-up,
shutdown, and malfunction plan because the General Provisions
in §63.6(e)(l)(i) require that a source be operated in a manor
consistent with good air pollution control practices. The EPA
has determined that it is appropriate to address the handling
of wastewater generated by maintenance activities in a
facility's start-up, shutdown, and malfunction plan. The EPA
has concluded that the concentration and flow rates of
maintenance wastewater streams are extremely difficult to
determine. Thus, facility determination of a de minimis level
for maintenance wastewater and subsequent enforcement would be
difficult. The EPA decided that it was more appropriate to
require facilities to develop a site-specific plan for
reducing emissions from all maintenance-related wastewater,
rather than to try to distinguish between which maintenance-
related wastewaters should be subject to additional control in
the final rule. Therefore, the requirements for routine
maintenance wastewater have been changed and are now the same
as the requirements proposed for maintenance-turnaround
wastewater. Routine maintenance and maintenance-turnaround
wastewaters are now both being referred to as "maintenance
wastewater." The provisions in proposed §63.102(b)(1)(ii)
3-13
-------
which required routine maintenance wastewater to be collected
and recycled, destroyed, or collected and managed in a closed-
drain system have been eliminated. The control requirements
for maintenance wastewater are to properly manage and control
HAP emissions. The commenters did not define what was meant
by "significant emissions" or "resource-intensive."
The EPA is not required to determine a floor for the
control of maintenance wastewater. The Act requires the EPA
to ensure that control of maintenance wastewater is at least
as stringent as the floor. Because estimating air emissions
from maintenance wastewater is difficult, the EPA reduced the
control requirements for routine maintenance to wastewater
recordkeeping and reporting requirements which are addressed
in the start-up, shutdown, and malfunction plan.
Comment; One commenter (A-90-19: IV-D-32) provided the
results of a study which examined several different conditions
for both drop structure and process drain collection system
components. Two commenters (A-90-19: IV-D-32; IV-D-108)
stated that the study on drains and drop structures indicated
that the EPA overestimated emissions for a number of chemicals
and suggested that these chemicals be removed from the HAP
lists. One commenter (A-90-19: IV-D-108) stated that the
CMA's study on drop/drain systems specifically indicated that
methanol was not emitted. The commenter (A-90-19: IV-D-L08)
stated that methanol volatilized very slowly. The commenter
(A-90-19: IV-D-108) claimed that if methanol does not
volatilize from drop/drain systems, then it is unlikely that
methanol will volatilize in a steam stripper operated at
higher temperatures. The commenter (A-90-19: IV-D-108)
suggested that only HAP's listed in table 8 of subpart G will
be emitted from wastewater collection and treatment.
Response; The cited report presents emissions data on
four compounds: 1,4-dichlorobenzene, tetrachloroethylene,,
trichloroethylene, and toluene. In the study, pilot scale
structures were used to simulate full scale operating
conditions for drains and drop structures. Based on the
results of the study, the EPA revised the emission models for
3-14
-------
junction boxes, sumps, lift stations and drains to include the
assumption that the organic HAP compound vapor phase
concentration above the wastewater corresponds to
approximately one-half of the saturated vapor concentration.
In the proposal analysis, it was assumed that the vapor phase
was at equilibrium with the wastewater. The EPA also revised
the emission model for junction boxes to be based on a
quiescent surface rather than turbulent flow. Emission
measurement for drains presented in the study were within
approximately six percent of EPA's original estimates.
The revised emission models were used to revise estimates
of Fe for junction boxes, open drains, open sumps, and lift
stations. Further review of the CMA drop/drain study
indicates that the EPA's assumption that water seal controls
would be equivalent to hard piping is in error. Based on this
finding, the EPA revised the requirements for water seals. In
§63.136(e) of the final wastewater provisions if a water seal
is used on a drain hub receiving a Group 1 wastewater, the
owner or operator shall either extend the drain pipe
discharging the wastewater below the liquid surface in the
water seal, or install a flexible cap (or other enclosure
which restricts wind motion) that encloses the space between
the drain discharging the wastewater to the drain hub
receiving the wastewater.
Comment; Two commenters (A-90-19: IV-D-75; IV-D-32)
provided data indicating that chemicals with Henry's law
constants less than 10~4 atm/(mole/m3) have little potential
for emissions from wastewater and that this value should be
the cutoff for VOHAP's.
One commenter (A-90-19: IV-D-33) stated that the process
wastewater provisions in §63.131 should apply only to those
chemicals with- significant potential for emissions. The
commenter (A-93—19: IV-D-33) stated that the range of Fe
values for table 8 compounds varies from 0.72 to 0.99, and
agreed that these 24 chemicals have a significant emission
potential. The commenter (A-90-19: IV-D-33) suggested that
all chemicals in table 9 with Fe values less than the lowest
3-15
-------
Fe value for the 24 table 8 chemicals should be deleted from
§63.131(b) table 9 and not be subject to all HAP regulatory
requirements.
Response; The EPA has revised the list of HAP's that are
included in table 9 of subpart G of the wastewater provisions.
In the proposed HON, the EPA identified 83 compounds in
table 9 of subpart G to be regulated in the wastewater
provisions. These HAP's are a subset of the HAP's regulated
by the HON. In selr -ing the HAP's identified in the proposed
table 9, the EPA el- _nated compounds that do not exist in
water and compounds the EPA determined would be unlikely to be
emitted in significant quantities. Another factor that
influenced the EPA se; ction of compounds was the
biodegradability and • a fraction removed by steam stripping.
The lower volatility compounds that were eliminated from the
table 9 list are already biodegraded to a significant extent
and are not removed to a significant extent by steam
stripping. Based on comments received from industry, the EPA
re-evaluated the emission estimates. Changes were made to the
emission models and new scenarios were developed. Based on
these revisions, new emission estimates were calculated. The
EPA reviewed the new values and decided to eliminate seven
additional compounds that were on proposed table 9 of
subpart G based upon the same criteria used to develop the
proposed table 9 list. Therefore, these 76 compounds are
included in table 9 of subpart G and are subject to the
wastewater provisions of' the HON.
Comment: One commenter (A-90-19: IV-D-34) stated that
the EPA has not established a sound technical basis for the
cutoff range of table 9 organic HAP's. The commenter
(A-90-19: IV-D-34) stated that the EPA should use ambient
conditions at 25 °C to determine a compound's volatility for
the purpose of estimating emissions rather than using steam
stripper operating conditions at 100 °C. The commenter
(A-90-19: IV-D-34) provided a list of compounds including
methanol, which are miscible in water or have a Henry's law
3-16
-------
constant less than 2x10-5 atm/(mole/m3) and stated that such
compounds should be removed from tables 9, 11, 13, and 33.
Response; For the HON analysis, the EPA did not use
steam stripper operating conditions at 100 °C to estimate
organic HAP emissions from wastewater. Rather, the wastewater
temperature was assumed to be 30 °C and Henry's law constants
at 30 °C were used for the purpose of estimating organic HAP
emissions from wastewater.
The commenter provided no technical basis, other than
compound volatility, for deleting from the wastewater
provisions of the HON those compounds with a Henry's law
constant less than 2xlO~5 atm/(mole/m3). There are other
factors which must be considered in excluding compounds from
the regulation, including the compound's potential to be
emitted as indicated by the Fe value.
The revised emission estimates completed by the EPA show
that removing compounds with a Henry's law value less than
2xlO~5 atm/(mole/m3) from the list of regulated organic HAP's
would result in regulation of only those organic HAP's with Fe
values greater than 20 percent. To revise the list of organic
HAP's as suggested by the commenter would result in nine
additional organic HAP's being removed from table 9 of
subpart G, all of which have the potential to be emitted from
wastewater. Therefore, the EPA has not revised the list of
regulated HAP's as suggested by the commenters. As discussed
in a previous response, the EPA has removed seven HAP's from
the list of table 9 HAP's.
Comment: One commenter (A-90-19: IV-D-108) claimed that
methanol is not strippable but is, according to WATER?, highly
biodegradable. One commenter (A-90-19: IV-D-92) stated that
water-soluble HAP's cannot be effectively removed by steam
stripping, making it unlikely that such HAP's would volatilize
in wastewater collection and treatment systems. One commenter
(A-90-23: IV-D-18) stated that aqueous methanol solutions do
not readily volatilize because of the hydrogen bonding that
occurs between the -OH radicals of water and methanol and
3-17
-------
consequently cannot be stripped to the level indicated in
table 33 (i.e., 0.829).
Response: Methanol can be removed from wastewater by
steam stripping. According to revised estimates made by the
EPA, the design steam stripper removal efficiency for methanol
is 31 percent (Fr = 0.31) (Memorandum from Clark Allen,
Research Triangle Institute, to Elaine Manning, EPA/CPB,
"Efficiency of Steam Stripper Trays to Treat Wastewater
Streams: Prediction of the Fraction Removed (Fr) for Specific
Compounds," January 7, 1994). Water soluble compounds,
including HAP's, are stripped from wastewater and are
concentrated in the overheads vent stream. Revised estimates
completed by the EPA also indicate that water soluble
compounds, including HAP's, are emitted from wastewater
(Memorandum from Clark Allen, Research Triangle Institute, to
Elaine Manning, EPA/CPB, "Estimation of Air Emissions from
Model Wastewater Collection and Treatment Plants," February 2,
1994) .
Comment; One commenter (A-90-19: IV-D-32) stated that
the Fe value used in the equations in §63.150 to calculate
uncontrolled emissions from wastewater collection and •
treatment devices should be related to the specific type and
design of management units used at a plant, not simply the Fe
values in table 13 of the proposed rule, which are based on
entire treatment systems with uncontrolled components. The
commenter (A-90-19: IV-D-32) suggested that the current
equations could serve as the default format for sources that
do not wish to use more detailed emissions factors.
Response; As discussed previously, the revised Fe values
are based on the average of a range of conditions for the type
and design of the wastewater collection and treatment system,
including controls. The EPA judges these estimates as a
reasonable basis for determining both the emission reduction
benefits of the HON and the credits and debits for emission
averaging. The increased burden on the industry and the
permitting authorities that would occur if site-specific
emission estimates are judged to be unreasonable compared to
3-18
-------
the potential for increased accuracy in the emissions
estimates is negligible.
comment: One commenter (A-90-19: IV-D-110) stated that
the EPA used a flawed methodology to calculate the removal
efficiencies (Fr) of HAP's in wastewater. The commenter
(A-90-19: IV-D-110) stated that the EPA's estimates are
incorrectly based on a model that assumes a linear
relationship between stripping efficiency and the Henry's law
coefficients of specific VOHAP's at 25 °C. The commenter
(A-90-19: IV-D-110) contended that the EPA's methodology is
incorrect because there is a sigmoidal, rather than a linear
relationship between these two variables, and because removal
efficiency (Fr) is not simply a function of the Henry's law
constant of a compound. The commenter (A-90-19: IV-D-110)
stated that the use of a flawed methodology results in an
overestimation of target removal efficiencies in table 9 and
an inaccurate removal efficiency (Fr) estimate for many
compounds in table 33.
Response; After reviewing additional technical
information, the EPA has revised the values for Fe and Fr in
the final rule. The Fe values in the final rule were
estimated for each individual compound using the revised
scenarios and are in table 34 of subpart G. Additionally, the
Fr values estimated for the proposed rule using the linear
relationship between Fe and the Henry's law constant have been
replaced in the final rule with the revised values estimated
using the Kremser equation in table 9 of subpart G.
Comment; Several commenters (A-90-19: IV-D-32; IV-D-75;
IV-D-97) (A-90-23: IV-D-20) argued that the EPA's national
emissions estimates and their estimates of removal
efficiencies are based on outdated information, and that the
EPA should use the data supplied by CMA to re-evaluate the
basis of the regulation. One commenter (A-90-19: IV-D-110)
urged the EPA to review any inaccurate and outdated
information used in selecting the RCT. Several commenters
(A-90-19: IV-D-32; IV-D-75; IV-D-97) (A-90-23: IV-D-20)
recommended using revised physical property data, refined
3-19
-------
emissions models, and SOCMI plant scenarios to update emission
factors (Fe) and estimates of removal efficiency factors (Fr)
and to ensure that the rule meets the proposed cost
effectiveness targets. One commenter (A-90-19: IV-D-32)
stated that data were provided in the comment letter for use
in updating emission factors (Fe) and removal efficiency
factors (Fr). Two commenters (A-90-19: IV-D-32; IV-D-73)
suggested using individual compound Fr values because many of
the group B and group C compound removal efficiencies are
overestimated. Two commenters (A-90-19: IV-D-32; IV-D-73)
suggested using the Kremser equation to estimate removal
efficiency factors (Fr).
Response; The national emissions estimates and removal
efficiency estimates made by the EPA for the proposed HON were
not based on outdated information, but on information
available at the time of the analyses. The CMA did provide
information regarding SOCMI plant wastewater system scenarios
and emissions models after the HON was proposed. Some of this
information has been incorporated into the final national
emissions estimates (Memorandum from Clark Allen, Research
Triangle Institute, to Elaine Manning, EPA/CPB, "Estimation of
Air Emissions from Model Wastewater Collection and Treatment
Plants," February 2, 1994) and (Memorandum from Clark Allcsn,
Research Triangle Institute, to Elaine Manning, EPA/CPB,
"Efficiency of Steam Stripper Trays to Treat Wastewater
Streams: Prediction of the Fraction Removed (Fr) for Specific
Compounds," January 7, 1994).
It is assumed that by "revised physical property data,"
the commenters are referring to revised Henry's law constants.
Henry's law constants were updated as part of a joint effort
between the EPA and the CMA. These revised Henry's law
constants have been used in the final estimates of national
impacts (Memorandum from Randy McDonald, EPA/CPB, to HON
Wastewater Docket, "Henry's law Constants for the 83 HAP's
Regulated in the Proposed HON Wastewater Provisions,» May 15,
1993) .
3-20
-------
The revised Fr values for the final HON regulation are
estimated using revised Henry's law constants at 100 °C. The
EPA clarifies that the Kremser equation was the basis for the
estimated values in the proposed regulation. In the final HON
regulation, table 9 of subpart G lists individual compound Fr
values, rather than grouping compounds by a range of Fr values
into the target removal efficiency groups used in the proposed
regulation. The individual Fr values were estimated using the
Kremser equation and are used to demonstrate compliance with
mass removal or percent mass reduction treatment options.
Comment: One commenter (A-90-23: IV-G-2) submitted a
copy of a memorandum, which summarizes raw process wastewater
concentration and loading data gathered in section 114
questionnaires for the OCPSF Industry. The commenter
(A-90-23: IV-G-2) stated that the data indicate that the EPA
may have substantially underestimated the extent of wastewater
emissions and the extent of the use of steam stripping in the
industry.
Response; The information submitted by the commenter was
based on responses to a section 114 questionnaire submitted by
the EPA to the Organic Chemicals, Plastics, and Synthetic
Fibers (OCPSF) industry in July 1986. This survey did not
specifically target the SOCMI nor did it specifically target
the SOCMI chemicals listed in §63.105 of the proposed HON.
Not all of the processes summarized in the data presented by
the commenter (A-90-23: IV-G-2) are SOCMI processes, and not
all of these processes emit HAP's. Additionally, the results
of the June 1986 survey report total organic concentrations,
but not individual compound concentrations, and, therefore,
cannot be used to estimate HAP emissions. The total organic
concentration includes both HAP and non-HAP compounds.
The data cannot be used to estimate HAP emissions from
wastewater because individual compound concentrations are not
reported. Additionally, processes other than SOCMI processes
are represented. The EPA's estimates of wastewater emissions
for the proposed and final HON are based on responses to a
section 114 survey conducted in March of 1990. The
3-21
-------
section 114 survey specifically targets the SOCMI and organic
HAP's. Therefore, the EPA concludes that HAP emissions from
wastewater, which were estimated using the section 114 data
for the SOCMI, are representative of the source.
Further, the data submitted by the commenter do not
substantiate the claim that the EPA may have substantially
underestimated the extent of the use of steam stripping in the
SOCMI. The OCPSF data indicate that out of a total of
356 streams, only 27 are treated by steam stripping
(7.6 percent). This indicates that, while steam stripping is
employed in the OCPSF industry, it is not used to control a
significant portion of the wastewater streams. It is not
possible to estimate how many of the 27 steam strippers are
actually used to control emissions from HAP-containing
wastewater streams. The EPA determined that the MACT floor
for wastewater was no control.
Comment; One commenter (A-90-23: IV-D-2) stated that
the wastewater provisions are based only on the HAP
concentrations in wastewater and the assumption that SOCMI
processes are significant sources of HAP emissions. The
commenter (A-90-23: IV-D-2) claimed that HAP emissions from
wastewater depend on the true partial pressure of the HAP and
the degree of exposure to the atmosphere. The commenter
(A-90-23: IV-D-2) stated that the true partial pressure for a
compound depends on concentration, temperature, and
interactions with other chemicals. The commenter (A-90-23:
IV-D-2) indicated that all of these factors should be
considered when determining control levels for wastewater
streams and closed-vent systems.
Another commenter (A-90-23: IV-D-17) claimed that the
EPA has ignored the variation in vapor-liquid equilibrium in
HON wastewater streams whxch is caused by interaction between
some volatile organics. The commenter (A-90-23: IV-D-17)
specifically cited the interaction between benzene and acetone
in water.
Response; The wastewater provisions in both the proposed
and final HON are based on several technical^analyses. These
3-22
-------
analyses estimate the impacts of implementing the HON. The
impact analyses include a quantitative review of emissions
reduction, cost effectiveness, energy impacts, secondary
environmental impacts, and economic impacts.
In reviewing the emission reduction impact of the HON,
the EPA agrees that several factors including partial
pressure, degree of exposure to the atmosphere, HAP
concentration, and temperature affect HAP emissions from
wastewater. However, there are other factors which also
affect HAP emissions from wastewater including wind speed,
wastewater depth, wastewater flow rate, and physical and
chemical properties of the compounds (e.g., diffusivity,
molecular weight, Henry's law constant, etc.) in the
wastewater.
For purposes of determining Fe values for HAP compounds,
the wastewater was assumed to have an average temperature of
30 °C, and the partial pressure of the organic HAP's in
wastewater is assumed to be described by Henry's law at 30 °C.
The EPA also assumes that multi-component interactions are
negligible. The commenter, who cited the interaction of
benzene and acetone in water as an example of multi-component
interaction, did not provide any data. None of the wastewater
streams used in the analyses conducted by the EPA contain a
mixture of benzene and acetone.
Comment: One commenter (A-90-23: IV-D-17) claimed that
the EPA has ignored the effects of fouling and surfactants or
detergents on the removal efficiency of the design steam
stripper. The commenter (A-90-23: IV-D-17) asserted that
surfactants, which may be present in wastewater, alter the
surface tension or wetting characteristics of the column and
may also cause foaming.
Response; The commenter did not describe any specific
causes of fouling in the steam stripper or discuss the effects
of fouling and foaming on steam stripper performance. In the
absence of such information, the EPA is unable to further
address the comment. The EPA recognizes that fouling and the
effect of surfactants and detergents on the performance of a
3-23
-------
steam stripper are site-specific considerations for which
information is not available. The EPA notes that a variety of
defoaming agents are available for many applications,
including wastewater treatment.
Comment; One commenter (A-90-19: IV-D-85) stated that
the EPA seems to have underestimated the capabilities of the
steam stripper by not accounting for the added emission
reductions by the condenser.
Response; The emission reductions from wastewater result
from the removal of organic HAP's from the wastewater due to
steam stripping. Once the organic compounds are stripped,
they cannot be vented to the atmosphere, but must be routed to
a control device, as required by §63.138(i). The emission of
HAP'S from a control device used to meet the provisions of
§63.138(i) wil* be negligible. Additionally, once the HAP's
are removed fro.a the wastewater and the treated wastewater
exits the steam stripper, no further reduction of HAP
emissions from wastewater is required if the provisions of the
regulation have been met. The condenser referred to by the
commenter was specified in §63.138(f) of the proposed rule and
was intended to control the emission of HAP's removed by the
steam stripper. Although the condenser does reduce emissions
from residuals (i.e., organics removed from wastewater), the
EPA has determined that this reduction is too difficult to
predict and does not make a large difference when calculating
emission credits and debits for averaging. Therefore, the
capabilities of the steam stripper have not been
overestimated.
It should be noted that the requirement for a condenser
in proposed §63.138(f) has been deleted from the final rule in
§63.138(g). The primary condenser may not be used to
demonstrate compliance with the 95-percent control requirement
for control devices.
3.2.1 Emissions from Biological Treatment Units
Comment: One commenter (A-90-19: IV-D-32) stated that
the EPA's emission factors for wastewater collection and
treatment systems overestimate the air emissions from
3-24
-------
biological treatment. The commenter (A-90-19: IV-D-32)
contended that the design and operating parameters used by the
EPA to define a typical biological treatment unit did not
represent those typically found in the SOCMI.
Response; The commenter used WATER? to estimate the
values for F^io an<* Fe f°*" those table 9 compounds for which
biokinetic data are present in WATER? and which the commenter
believes are biodegradable. Table 3-1 of this section
summarizes the required WATER? input parameters defined by the
EPA and those suggested by the commenter as being typical of a
SOCMI biological treatment unit.
Using the input parameters suggested by the commenter,
the EPA was unable to reproduce the commenter's results. The
EPA then requested a computer disk copy of the WATER? input
files used by the commenter. Examination of the WATER? input
files provided by the commenter revealed that the numerical
value for inlet solids (2,000) was entered as the input for
active biomass concentration. That is, the results presented
by the commenter correspond to a biomass concentration of
2,000 g/£. Typical biomass concentrations range from 1 to
6 g/£. This overestimation of active biomass concentration
results in the overestimation of the biodegradation rate and
underestimation of the air emission rate from biological
treatment units.
Using the input parameters summarized in table 3-1 of
this section, the EPA used WATER? to estimate the Fe for a
biological treatment unit. A summary of the results is shown
in table 3-2 of this section. The results indicate that use
of the commenter's suggested inputs, after correction of the
biomass concentration value, results in even higher estimated
emissions from biological treatment units compared to the
estimated emissions using the EPA input values. For example,
the Fe for benzene is 0.198 using the EPA input parameters,
whereas the Fe for benzene is 0.34 using the commenter's
suggested inputs. Therefore, the EPA concludes that emissions
from biological treatment units were not overestimated in the
EPA's impact analysis for the HON. Refer to a memorandum from
3-25
-------
TABLE 3-1. SUMMARY OF EPA AND COMMENTER
WATER7 INPUT PARAMETERS
Water Flow Rate (m3/s)
Total dissolved organics (mg/£)
Inlet solids (mg/£)
Width of aeration (m)
Length of aeration (m)
Depth of aeration (m)
Active biomass (g/1)
Aeration air flow (m3/s)
Number of units
Number of agitators
Area of agitation (each aerator, m^)
Aerator alpha (default=0.83)
Power of agitation (each aerator, HP)
Impeller diameter (cm)
Impeller rotation (HP)
Enter 1 if plug flow
Wind velocity (cm/s at 10m)
Wastewater temperature (°C)
Enter 1 if covered and vented
EPA
0.0693
0
0
132.9
132.9
1.981
4
0
1
8
530
0.83
75
61
1203
0
447
30
0
COMMENTER
0.0693
1000
2000
39.2
39.2
3.5
4
0
1
2
530
0/83
75
61
1.203
0
447
30
0
TABLE 3-2
COMPARISON OF FE VALUES PREDICTED BY WATER?
FOR SELECTED TABLE 9 HAP'S
Benzene
Methanol
Naphthalene
Nitrobenzene
Toluene
EPA
0.1979
0.0103
1100
u.030
0.1192
COMMENTER
0.3398
0.0173
0.1806
0.076
0.2025
3-26
-------
Clark Allen, Research Triangle Institute, to Elaine Manning,
EPA/CPB, "Estimation of Air Emissions from Model Wastewater
Collection and Treatment Plants," February 2, 1994, for
further information.
3.2.2 Use of Wastewater Models
Comment; Several commenters (A-90-19: IV-F-1.2 and
IV-F-4; IV-D-112; IV-D-77) stated that the EPA's approach for
estimating total emissions from wastewater operations is
inaccurate because the EPA used data solely generated from
models. One commenter (A-90-19: IV-D-75) indicated that the
EPA's model plant was oversimplified and unrealistic and that
wastewater streams are not centrally collected for treatment
at a single steam stripper. One commenter (A-90-19: IV-F-1.2
and IV-F-4) credited the EPA for conducting several field
studies, but concluded that these studies were poorly designed
and resulted in questionable data.
Response: The data used by the EPA for estimating
organic HAP emissions from wastewater are not based solely on
models. In March 1990, a section 114 wastewater questionnaire
was submitted to nine corporations. While the proposal BID
mistakenly cited that 84 model streams were used, actually, a
total of 461 wastewater streams from 110 SOCMI production
processes were reported in responses. An additional 107 model
wastewater streams were developed for 75 SOCMI product
processes that were not characterized by wastewater streams in
the section 114 responses. These 107 model wastewater streams
were developed based on a combination of process knowledge,
engineering judgement, and information provided in the
section 114 responses.
The EPA agrees that some facilities may choose to treat
all or some of their Group 1 streams in multiple locations.
However, the EPA's final impact analysis indicates that only
approximately 8 percent of the total SOCMI industry wastewater
will be affected under this rulemaking. Therefore, the
assumption that a single steam stripper is adequate to treat
the Group 1 streams in a facility is reasonable for developing
cost impacts.
3-27
-------
For the proposed rule, the wastewater emission estimates
were based on three example wastewater collection and
treatment scenarios. Based on public comment, new wastewater
collection and treatment scenarios were developed to more
accurately represent the SOCMI.
It is not clear what field studies are being referenced
by the commenter. The commenter provided no information
regarding which aspects they considered to be poorly designed
and provided no data to substantiate their claim that the data
is questionable. The EPA has thoroughly reviewed the studies
in "Technical Support for the Identification of Collection
Systems at Major Emission Sources," January 4, 1994. These
studies were used to revise the fraction emitted (Fe) values
as described in the memorandum from Clark Allen, Research
Triangle Institute, to Elaine Manning, EPA/CPB, "Estimation of
Air Emissions from Model Wastewater Collection and Treatment
Plants," February 2, 1994.
Comment; Two commenters (A-90-19: IV-D-68; IV-D-71)
claimed that the wastewater emission models and data used in
EPA's HON analysis are outdated and overestimate wastewater
emissions. One commenter (A-90-19: IV-D-108) asserted tha_
the EPA overestimated emissions from wastewater collection
systems by making unrealistic modeling assumptions. Two
commenters (A-90-19: IV-D-68; IV-D-71) claimed that the EPA
has ignored data from a study entitled, "Amoco/USEPA Pollution
Prevention Project, Project Summary, January 1992, Revised
June 1992, page 2-6 and Figure 2-8" which indicates that air
emissions from wastewater have been overestimated by the
models by a factor of 21. One commenter (A-90-19: IV-D-97)
contended that the EPA dismissed actual emission measurement
data from a large facility containing both SOCMI and non-£>OCMI
processes.
One commenter (A-90-19: IV-D-108) cited pilot-scale
studies done by the EPA at a pharmaceutical company which
indicate that methanol removal by steam stripping is typically
less than 50 percent.
3-28
-------
Response; The EPA assumes that the commenter's use of
the term "data" refers to the information obtained from a
section 114 survey of the SOCMI. The data were collected in
March and April of 1990. It is unlikely that significant
changes in SOCMI process design and operation have taken place
and/or been implemented since 1990. Therefore, the EPA
maintains that the data are current.
The commenters (A-90-19: IV-D-68; IV-D-71) did not
explain which of the emission models they consider to be
outdated and/or unrealistic; nor were any data or alternative
approaches submitted. Therefore, it is not possible for the
EPA to act on this comment.
The EPA did not ignore data from the Amoco/USEPA
Pollution Prevention Project. The EPA's viewpoint is
documented in a report titled "EPA Follow-Up to the
Recommendations of the EPA/Amoco Yorktown Project." This
report includes a discussion of basic methodological
limitations which the EPA believes resulted in an
underestimation of air emissions from wastewater. These
methodological limitations include location of emission
measurement points in the wastewater system that, in the EPA's
view, were located after substantial emissions could have
occurred. Ambient monitoring cross-checks performed to
validate emission estimates indicated that benzene emissions
may have underestimated by a factor of 2 or more at several
sampling points.
The EPA revised the estimates for the removal
efficiencies of all compounds regulated under the wastewater
provisions of the RON, including methanol. The new estimates
reflect revisions to the Henry's law constants and steam
stripper removal efficiencies calculated using the Kremser
equation. The revised steam stripper removal efficiency for
methanol is approximately 30 percent, which agrees with the
commenter's statement that methanol removal via steam
stripping is typically less than 50 percent.
Comment: One commenter (A-90-19: IV-F-4) objected to
the EPA using model streams, model collection and treatment
3-29
-------
systems, and emission models to estimate national impacts from
wastewater collection and treatment systems. The commenter
(A-90-19: IV-F-4) acknowledged that the EPA did conduct
several field studies, but said that the studies were poorly
designed and resulted in questionable data. The commenter
(A-90-19: IV-F-4) concluded that the theoretical methods used
by the EPA result in an overprediction of emissions, but did
not suggest an alternative approach. Two commenters (A-90-19:
IV-D-69; IV-D-75) said the assumption that wastewater streams
are centrally collected for steam stripping at a single
stripper, and stripped materials are burned in an auxiliary
incinerator were unrealistic assumptions.
Response: The EPA recognizes that some facilities may
choose to treat all or some of their Group 1 wastewater
streams in multiple locations. However, the EPA's final
impact analysis indicates that only approximately 8 percent of
the total SOCMI industry wastewater draw is affected under
this rulemaking. Therefore, the assumption made in developing
cost that a single steam stripper is adequate to treat the
Group I streams in a facility is reasonable.
In the proposed HON, a fuel credit for stripped materials
sent to a boiler was included in estimating the total national
annual costs of steam stripping. Based on comments to the
proposed rule, the EPA removed this credit for the stripped
materials in estimating total national annual cost of steam
stripping in the final rule. The EPA believes that some
sources may earn recovery credits due to recycling or firing
of recovered organics in boilers to produce steam while other
facilities may incur a debit due to disposal costs. However,
for estimating national impacts, the EPA has assumed that the
credits and debits will cancel with no net impact on costs.
3.3 OTHER ENVIRONMENTAL IMPACTS
Comment; Several commenters (A-90-19: IV-F-1.2 and
IV-F-4; IV-D-77) (A-90-23: IV-D-1) expressed concern that
there would be negative environmental impacts caused by steam
stripping, such as the use of large amounts of energy to
3-30
-------
generate steam, the generation of residuals, and the emission
of additional pollutants to the air and other media.
Response: The EPA's analysis shows that secondary
impacts associated with steam stripping of wastewater are not
significant compared to the reduction of HAP's. The residuals
generated by steam stripping must be handled by either on-site
incineration, off-site incineration, or by recycling to the
process. The additional fuel required to generate steam can
be partially offset by recovering organics and using them as
supplementary fuel. Furthermore, combustion of recovered
organics generates less SC>2 and PM than combustion of fossil
fuels. Recycled organic compounds do not contribute to
secondary impacts. Steam stripping has a positive impact on
the quality of water being discharged to a wastewater
treatment system or a POTW. The issue concerning use of large
amounts of energy to generate steam is addressed in more
detail in section 3.4 of this BID volume.
Comment; One commenter (A-90-19: IV-D-50) alleged that
the EPA underestimated the impacts of NOX emissions from
wastewater control, claiming that the EPA did not consider the
NOX generated from steam stripping.
Response; The EPA did estimate the NOX emissions that
are generated from the combustion of fossil fuels to produce
steam for use in steam strippers. The EPA's estimate of the
NOX emissions generated by steam stripping is 600 Mg/yr as
presented in table 5-4 of the proposal BID volume 1A.
3.4 ENERGY IMPACTS
Comment; One commenter (A-90-19: IV-D-110) stated that
steam stripping requires large amounts of energy to generate
steam, and typically uses fossil fuels. The commenter
(A-90-19: IV-D-110) indicated that cooling stripper bottoms
may require additional energy, which may increase the negative
impact on global warming. The commenter (A-90-19: IV-D-110)
suggested that the EPA review these factors to accurately
determine the costs and benefits of steam stripping.
Response: The EPA has assumed that the latent heat from
the steam stripper bottoms is used to preheat the wastewater
3-31
-------
entering the steam stripper. Therefore, only a small amount
of additional energy may be required to cool the steam
stripper bottoms. The EPA has also reviewed the energy and
secondary impacts generated from the use of steam strippers
and has determined that these impacts are insignificant
compared to the achieved emission reduction from wastewater.
Energy and secondary impacts are presented in proposal BID
volume 1C and the proposal preamble.
3-32
-------
4.0 APPLICABILITY AND GROUP I/GROUP 2 DETERMINATION
4.1 APPLICABILITY
Comment; One commenter (A-90-19: IV-D-98) stated that
the EPA should describe its legal authority under the Act to
establish the applicability of MACT-based standards based on
the VOHAP concentration at the point of wastewater generation
(i.e., before HAP's can be emitted).
Response; In the final rule, the EPA requires that the
owner or operator determine applicability of the regulation at
the point of generation or downstream of the point of
generation. Once applicability is.determined (that is, once
the Group 1 wastewater streams are identified), the owner or
operator must ensure that Group 1 wastewater streams are
controlled for HAP emissions. The EPA clarifies that emission
controls are not required until the owner or operator
identifies a Group 1 wastewater stream. At which time, such a
stream must be controlled from the point of generation in
accordance with all applicable regulations.
Comment; One commenter (A-90-19: IV-D-91) stated that
petroleum refinery wastewater collection and treatment systems
should not be regulated by the HON because these systems will
be regulated by a separate MACT rulemaking.
Response; If wastewater is generated by a SOCMI process
unit and is managed in a combined collection and treatment
system (i.e., the system collects and treats wastewater from
both SOCMI and petroleum refinery units), the HON remains
applicable to wastewater generated by SOCMI units. The owner
or operator of a facility that generates wastewater from SOCMI
process units must first determine whether such wastewater is
a Group 1 wastewater stream and consequently must be
4-1
-------
controlled. If the owner or operator elects to manage Group 1
wastewater streams in a combined collection and treatment
unit, the HON provides several compliance options. If a SOCMI
Group 1 wastewater stream or a residual generated from a
Group 1 stream is generated at a petroleum refinery facility,
the wastewater and any residuals are still subject to the HON.
The HON applies to all SOCMI processes. Therefore, even if
the primary function of a facility is non-SOCMI, any SOCMI
process unit at the facility is regulated by the HON assuming
the facility is a major source.
Comment ; One commenter (A-90-23: IV-D-20) supported
raising the flow rate component of the applicability criteria
in §63.110(e) (1) from 0.02 £/m to 0.2 £/m because the proposed
criteria will include streams with very low flow rates. The
commenter (A-90-23: IV-D-20) stated that a facility's
resources could be used more effectively in controlling
streams with higher flow rates.
Response: The definition of "wastewater" in §63.101 of
subpart F, which includes both process wastewater and
maintenance wastewater, defines the applicability criteria for
wastewater below which wastewater streams are not subject, to
the HON. Wastewater streams with a total VOHAP concentration
less than 5 ppmw or a flow rate less than 0.02 £/m are not
subject to the HON. The owner or operator of a wastewater
stream that meets the definition of "wastewater" in §63.101 of
subpart F must determine whether the wastewater stream is a
Group 1 or Group 2 wastewater stream. Both Group 1 and
Group 2 streams are subject to subparts F and G, but only
Group 1 streams require treatment.
One commenter (A-90-19: IV-D-97) supported the
deletion of groups D and E from the strippability groups and
recommended that the EPA delete all HAP's from table 9 that
have a Henry's law constant value lower than
1.0 x 10~3 atm/(mole/m3) . One commenter (A-90-19: IV-D-32)
claimed that many of the excluded compounds would be
biodegradable and can be effectively treated in biological
treatment units.
4-2
-------
Response; Prior to the issuance of the proposed HON, the
EPA determined that the chemicals in strippability groups D
and E should not be subject to regulation by the HON because
such chemicals were not emitted at levels that required
control. The EPA agrees that the HON should not regulate
chemicals with little or no potential to emit, and therefore
deleted strippability groups D and E from the proposed HON.
The EPA also agrees that many of these chemicals may be
effectively treated using biological treatment and encourages
facilities to do so.
Comment; One commenter (A-90-19: IV-D-92) urged the EPA
to exempt water-soluble HAP's from the steam stripping control
requirements in the HON because such HAP's cannot be
effectively removed by steam stripping.
Response: The EPA has removed seven compounds from the
list of regulated HAP's in table 9 of subpart G based on their
low Fe values. Furthermore, any HAP's that the EPA has
determined to be water-soluble or water-reactive are not
regulated by the HON wastewater provisions. For the
76 remaining regulated HAP's, the EPA continues to allow steam
stripping as one of the options for treatment, but also allows
other compliance options in §§63.138(b)(1), (c)(1), (d), and
(e) including recycling and biological treatment.
Comment: One commenter (A-90-23: IV-D-20) suggested
that the EPA clarify language in §63.138(c)(1)(ii) regarding
whether facilities may treat several individual streams in the
same waste management unit.
Response; Although §63.138(c)(1)(ii) discusses the
treatment of individual wastewater streams, the EPA allows
other options for treatment of wastewater streams. According
to §63.l38(c)(1)(iii), facilities may aggregate several
wastewater streams to facilitate treatment.
Comment; One commenter (A-90-19: IV-D-33) stated that
the requirements concerning maintenance wastewater and heat
exchangers in §63.102(b) and the associated definition in
§63.101 should be modified and removed from subpart F and
placed in subpart G. The commenter (A-90-19: IV-D-33) stated
4-3
-------
that subpart F should be reserved for general applicability
issues. The commenter (A-90-19: IV-D-33) recommended that
the provisions in §63.102(b) should be moved to a new
subparagraph §63.110(f) and the associated definitions should
be moved from §63.101 to §63.111. By creating a new
subparagraph, the commenter (A-90-19: IV-D-33) stated that
the regulation would clearly not require such wastewater
streams to be subject to Group I/Group 2 determination
procedures.
Response: The heat exchange system and maintenance
wastewater provisions were placed in subpart F to distinguish
cooling waters and maintenance wastewaters from process
wastewaters, because they are subject to different
requirements than process wastewaters. For example, cooling
waters do not require a Group I/Group 2 determination.
Furthermore, subpart G requirements address routine emissions
from SOCMI operations, while subpart F addresses applicability
and general requirements, such as leak detection and repair
and the start-up, shutdown, and malfunction plan.
The heat exchange system provisions have been moved from
the general standards provisions in §63.102 of subpart F to a
separate heat exchange system section in §63.104 of subpart F.
The maintenance wastewater provisions have been moved from, the
General Standards provisions in §63.102 of subpart F to a
separate maintenance wastewater section in §63.105 of
subpart F. Therefore, §63.102 of subpart F only contains the
general applicability provisions. The definitions of heat
exchange system and maintenance wastewater remain in §63.101
of subpart F. Changes to the maintenance wastewater
provisions are provided in a previous discussion in this
section.
4.1.1 Definition of "Residuals"
Comment; Several commenters (A-90-19: IV-F-1.2 and
IV-F-4; IV-D-112; IV-D-32) (A-90-23: IV-D-21) requested
clarification from the EPA on the definition of "residuals."
Several commenters (A-90-19: IV-D-32; IV-D-53; IV-D-60;
IV-D-110; IV-D-112) (A-90-23: IV-D-2; IV-D-20) stated that
4-4
-------
the definition of "residuals" in the proposed rule was too
broad and could be interpreted to include settled inorganic
solids, polymers, and similar inert materials which may
contain only trace amount of HAP's.
Response; Based on comments received about the
definition of residuals, the EPA has changed the definition in
§63.111 of subpart G to read:
Residual means any HAP-containing water or organic
that is removed from a wastewater stream by a waste
management unit or treatment process that does not
destroy organics (nondestructive Unit). Examples of
residuals from nondestructive wastewater management
units are: the organic layer and bottom residue
removed by a decanter or organic-water separator;
and the overheads condensate stream from a steam
stripper or air stripper. Examples of materials
which are not residuals are: silt; mud; leaves;
bottoms from a steam stripper or air stripper; and
sludges, ash, or other materials removed from
wastewater being treated by destructive devices such
as biological treatment units and incinerators.
In response to several commenters who expressed concern
about the inclusion of polymers in the definition of residual,
the EPA has concluded, based on input from industry, that
polymers may be recycled to a production process. The EPA
encourages this management option for polymers; however, if
polymers generated from the treatment of a Group 1 wastewater
stream are not recycled to a production process, they must be
managed as residuals.
Comment: Several commenters (A-90-19: IV-D-32; IV-D-53;
IV-D-60; IV-D-79; IV-D-110; IV-D-112); (A-90-23: IV-D-20)
suggested that the definition of "residuals" be limited to
materials derived from treatment of Group 1 wastes and should
include a de minimis VOHAP concentration based on Group 1
wastewater criteria. For example, the commenters (A-90-19:
IV-D-32; IV-D-53; IV-D-60; IV-D-79; IV-D-110; IV-D-112);
(A-90-23: IV-D-20) recommended that a residual would have
greater than 1,000 ppmw based only on those HAP's in table 9
of the rule, and for new units, greater than 10 ppmw for
table 8 HAP's. One commenter (A-90-19: IV-D-60) suggested
that the EPA should clarify that residuals removed from a
4--5
-------
Group 1 wastewater stream should be subject to the residual
treatment requirements only when the residual is generated
during treatment which is required in order to achieve
compliance. For example, the commenter (A-90-19: IV-D-60)
stated that residuals, which are generated from the treatment
of a Group 1 wastewater stream and also comply with the
1 Mg/yr cutoff in §63.138(c)(5), should not be required to be
controlled under HON.
Response: The EPA agrees with the commenters that
residuals that are subject to regulation by the HON are
limited to those residuals that are removed fror a Group 1
wastewater stream, which is also subject to conr-ol
requirements in the HON. Residuals removed from the following
wastewater streams are not required to be controlled by the
HON: (1) Group 2 wastewater streams, if the Group 2
wastewater stream is managed separately from Group I
wastewater streams; and (2) Group 1 wastewater streams that
are not required to be controlled because the facility meets
the criteria for the 1 Mg/yr source-wide exemption in
§63.138(c)(5) or (6).
The EPA specifies in §63.138(h) that only residuals
removed from Group 1 wastewater streams must be controlled.
The EPA considered the incorporation of a de minimis VOHAP
concentration based on Group l wastewater criteria for
residuals; however, the EPA has concluded that all residuals
must be managed by: (1) being recycled to the process unit or
sold for the purpose of recycling; (2) being returned to the
treatment process; or (3) being treated to destroy the total
HAP mass flow rate by 99 percent or greater. For each of
these management options, the EPA clarifies that residuals
must be managed in accordance with the requirements in
§§63.133 through 63.137 until they are actually returned to
the process unit or treatment process; are destroyed; or are
converted to a raw material. The owner or operator must
ensure proper management of residuals even if they are handled
offsite. The EPA clarifies that the purpose of the residuals
provisions is to ensure that HAP emissions are actually
4--6
-------
controlled and not just shifted to another part of the
facility.
One commenter (A-90-19: IV-D-73) recommended
adding a minimum cut-off criteria of greater than 10 tons per
year and/or 1,000 ppm of table 9 substances for defining
wastewater or residual organic HAP levels at which control of
wastewater tanks, surface impoundments, containers, individual
drain systems, and oil water separators is required.
One commenter (A-90-19: IV-D-86) suggested that control
of residuals be required only for concentrations of at least
10,000 ppm. The commenter (A-90-19: IV-D-86) alleged that
only HAP's listed on table 9 and not total HAP's in residuals
should require 99 percent reduction.
Two commenters (A-90-19: IV-D-86), (A-90-23: IV-D-17)
favored having a de minimis level for total annual HAP
quantity in wastewater similar to the 10 Mg/yr total annual
benzene de minimis in the Benzene Waste NESHAP.
Response; The HON specifies the applicability criteria
for wastewater streams in the definition of "wastewater" iri
§63.101 of subpart F. If a wastewater stream does not meet
these applicability criteria, the wastewater stream is not
subject to the HON. The commenter (A-90-19: IV-D-73)
provided no reason why the EPA should raise the
Group I/Group 2 criteria for wastewater to 10 tons per year
and/or 1,000 ppm for table 9 HAP's. The Group I/Group 2
determination criteria for existing facilities specifies that
any process wastewater stream with either (1) a total VOHAP
average concentration of table 9 compounds equal to or greater
than 1,000 ppm and a flow rate equal to or greater than
10 £/m, or (2) with a total VOHAP average concentration equal
to or greater than 10,000 ppmw and any flow rate is a Group 1
stream and must be treated in accordance with the requirements
of §63.138.
The EPA does not specify minimum concentration cutoffs
for residuals because only those residuals that are generated
from the treatment of Group 1 wastewater streams must be
controlled. If such residuals were not controlled, there
4-7
-------
would be no point in requiring separation of the organic
residuals from wastewater.
The HON is a technology-based rule and the Benzene Waste
NESHAP is a risk-based rule. The 10 Mg/yr total annual
benzene threshold in the Benzene Waste NESHAP [40 CFR
subpart FF] is a facility-wide applicability threshold based
on risk and is therefore not relevant to the HON.
Furthermore, the 10 Mg/yr threshold applies to the total
annual benzene quantity from all facility waste with greater
than 10 percent water, and not just process wastewater. The
wastewater provisions of the HON apply to wastewater and
residuals generated by treatment of Group 1 wastewater streams
but not to all emission points at the source. Therefore, the
10 Mg/yr threshold that is specified in the Benzene Waste
NESHAP has not been incorporated.
Comment; Two commenters (A-90-19: IV-D-89; IV-D-92)
claimed that the Benzene Waste NESHAP excludes streams with
concentrations of less than 10 ppm while the HON includes
streams with concentrations of greater than 5 ppm.' One
commenter (A-90-19: IV-D-89) claimed that these
inconsistencies may require piping modifications. One
commenter (A-90-19: IV-D-89) alleged that §63.110(e) of the
proposed regulation includes streams with concentrations
greater than 5 ppm. The commenter (A-90-19: IV-D-92)
indicated that §63.132 excludes streams having a concentration
less than 10 ppm (table 8). The commenter (A-90-19: IV-D-92)
urged the EPA to be consistent with the Benzene Waste NESHAP
by excluding all streams with concentrations less than 10 ppm.
The commenter (A-90-19: IV-D-92) claimed that this would
prevent facilities that are in compliance with the Benzene
Waste NESHAP from having to rework any equipment.
Response; The HON defines "wastewater" in §63.101 of
subpart F as "organic hazardous air pollutant-containing water
or process fluid that is discharged from a chemical processing
unit that meets all applicability criteria specified in
§63.100(b)(1) through (b)(3) of this subpart and that is
discharged into an individual drain system and either
4-8
-------
(1) contains at least 5 ppmw total volatile organic HAP's and
has a flow rate equal to or greater than 0.02 £/m, or
(2) contains a concentration of at least 10,000 ppmw total
volatile organic HAP's and any flow rate." This definition
provides the applicability criteria for whether a wastewater
stream will be designated as a wastewater by the HON. The
definition of "wastewater" does not specify which wastewaters
will be controlled by the HON. Rather, it specifies which
wastewaters that the owner or operator must check for Group I
or Group 2 status. If a waste stream has less than 5 ppmw
total volatile organic HAP's, it is not considered a
wastewater stream under the HON, which means it will not be
subject to Group I/Group 2 determination (i.e., stream cannot
be a Group 1 or a Group 2 stream).
Under the HON, the Group I/Group 2 determination for a
wastewater stream designates whether the stream must be
controlled. The Benzene Waste NESHAP does not have a
Group I/Group 2 determination. The Benzene Waste NESHAP
requires the owner or operator to determine whether the
facility-wide total annual benzene quantity from facility
waste is greater than or equal to 10 Mg/yr. The 10 Mg/yr
threshold was selected because the Benzene Waste NESHAP is
risk-based and 10 Mg/yr exceeded the 1 x 10~4 MIR. If the
total annual quantity of benzene is greater than or equal to
10 Mg/yr, the owner or operator must control all streams with
a flow-weighted annual average benzene concentration of
10 ppmw or greater unless the waste stream is a process
wastewater that has a flow rate less than 0.02 £/m. To
compare the wastewater control requirements for the HON and
the Benzene Waste NESHAP, the EPA reviewed the control
requirements for the Benzene Waste NESHAP with the
requirements for both new and existing SOCMI sources subject
to the HON.
For both new and existing sources, the HON requires the
SOCMI owner or operator to determine whether each process
wastewater stream is a Group 1 or Group 2 wastewater stream
with respect to the compounds listed on table 9 of subpart G.
4.-9
-------
A wastewater stream is a Group 1 wastewater stream and must be
controlled in accordance with the HON if the total VOHAP
average concentration for a process wastewater stream at a new
or existing facility is (1) greater than or equal to
10,000 ppmw of the compounds on table 9; or (2) has an average
flow rate greater than or equal to 10 £/m and a total VOHAP
average concentration greater than or equal to 1,000 ppmw.
When the EPA compared these criteria for controlling air
emissions at new and existing facilities subject to the HON
with the control criteria for the Benzene Waste NESHAP, the
EPA concludes that the Benzene Waste NESHAP is more stringent.
In the case of new SOCMI sources that are subject to the
HON, the EPA has developed control criteria based on compounds
listed on table 8 of subpart G. The compounds on table 8,
which the EPA has determined are very volatile compounds, are
a subset of those on table 9. For these more volatile
compounds, the EPA has developed more stringent control
criteria than those required for the table 9 compounds at new
and existing SOCMI sources. For new sources, the HON requires
the SOCMI owner or operator to determine whether each process
wastewater stream is a Group 1 or Group 2 wastewater stream
with respect to the compounds listed on table 8. A wastewater
stream is a Group l wastewater stream and must be controlled
in accordance with the HON if the average flow rate is greater
than or equal to 0.02 £/m and the wastewater stream has an
average VOHAP concentration of 10 ppmw or greater of any one
of the compounds listed in table 8. For new sources, the
control criteria for Group I/Group 2 determinations for
compounds that are listed on table 8 are the same as the
control criteria for the Benzene Waste NESHAP.
While for new sources, the HON is consistent with the
Benzene Waste NESHAPJs control criteria (i.e., 0.02 £/m and
10 ppmw), the Benzene Waste NESHAP remains more stringent than
the HON for control of compounds listed on table 9 for both
new and existing sources. The EPA disagrees with the
commenters1 statement that the 5 ppmw VOHAP concentration in
the definition of "wastewater" in the HON is inconsistent with
4-10
-------
the 10 ppmw concentration in the Benzene Waste NESHAP. As
previously discussed, the 5 ppmw concentration that is
specified in both the definition of "wastewater" in §63.101
and the applicability criteria for wastewater in §63.110(e)
does not require the control of wastewater streams with
organic HAP concentrations greater than 5 ppmw and less than
10 ppmw. In fact, wastewater streams in this concentration
range would meet the definition of a Group 2 stream for both
new and existing facilities for compounds listed on both
tables 8 and 9. The EPA continues to include the 5 ppmw
applicability threshold to indicate that waste streams
containing below 5 ppmw total volatile organic HAP's are not
defined as wastewater streams by the HON. The commenter
(A-90-19: IV-D-92) did not provide any details about why the
equipment may require "rework."
Comment; Several commenters (A-90-19: IV-D-32; IV-D-60;
IV-D-79; IV-D-112) stated that the VOHAP concentration in some
residual materials such as inorganic grits and settleable
solids will typically be low, and in such cases a 99-percent
HAP removal for residuals will be unachievable. The
commenters (A-90-19: IV-D-32; IV-D-60; IV-D-79; IV-D-112)
suggested that the EPA establish a de minimis level for HAP's
in residuals.
Response: Although the EPA has not incorporated an
additional de minimis level for VOHAP concentrations in
residuals, materials such as leaves, silt, mud, and sludge
removed from a treatment device such as a biological treatment
unit have been specifically excluded from the definition of
"residual." Such materials will not contain significant
HAP's. The EPA continues to limit residuals to those streams
that are generated from Group 1 wastewater streams. Also,
because the HON provides other residual management options,
which include recycling the residual to a production process
and returning the residual to the treatment process, the
option to treat all residuals generated by a Group 1
wastewater stream by 99 percent or greater will remain in the
final rule.
4-11
-------
Commenti One commenter (A-90-19: IV-D-34) stated that
the EPA has not considered the safety aspects associated with
treatment of residuals.
Response; The EPA has considered the safety issue
associated with treatment of residuals and has not identified
any safety hazards. The commenter also did not explain any
specific concerns.
Comment: One commenter (A-90-19: IV-D-34) stated that
the EPA should develop rules for residual treatment under the
upcoming rule for TSDF facilities or defer the rulemaking
until a cost analysis is completed.
Response; The EPA clarifies that control of emissions
from residuals is an integral part of the HON and cannot be
separated into another rulemaking. Also, because all
residuals generated by SOCMI facilities are not sent to a
TSDF, the upcoming TSDF rulemaking would not necessarily apply
to residuals generated by SOCMI sources.
4.1.2 Definition of "Wastewater"
Comment; Several commenters (A-90-19: IV-D-32; IV-D-33;
IV-D-53; IV-D-54; IV-D-60; IV-D-102; IV-D-113; IV-D-110;
IV-F-1.2 and IV-F-4) (A-90-23: IV-D-9; IV-D-17; IV-D-20)
expressed concern that the current definition of wastewater in
§63.101 of subpart F seems to include, and thus regulate,
process fluids, products, and intermediate streams, which the
EPA did not intend to regulate. Three commenters (A-90-19:
IV-D-53; IV-D-86) (A-90-23: IV-D-9) recommended that the
definition of "wastewater" should not include raw materials,
intermediate products, finished products, or byproducts. One
commenter (A-90-23: IV-D-9) claimed that the transfer of
process fluids is central to production for batch processes
and that such process fluids are not wastewaters.
Response: The EPA agrees with the commenters that
process fluids, products, and intermediate streams that are in
use in a production or manufacturing process are not subject
to the HON. However, the EPA intends to regulate any such
stream if it is discharged to an individual drain system and
either (1) has a total VOHAP concentration that is equal to or
4-12
-------
greater than 5 ppmw and has a flow rate equal to or greater
than 0.02 £pm; or (2) has a total VOHAP concentration of
10,000 ppmw or greater at any flow rate. The EPA has revised
the definition of wastewater in §63.101 of subpart F as
follows:
Wastewater means organic hazardous air pollutant-
containing water, raw material, intermediate,
product, by-product, co-product, or waste material
that exits a chemical manufacturing process unit
equipment that meets all of the criteria specified
in §63.100(b)(1) through (b)(3) of this subpart and
either (1) contains a total volatile organic
hazardous air pollutant concentration of at least
5 ppmw and has a flow rate of 0.02 Ipm or greater;
or (2) contains a total volatile organic hazardous
air pollutant concentration of at least 10,000 ppmw
at any flow rate. Wastewater includes both process
wastewater and maintenance wastewater.
The EPA has removed the term "process fluid" from the
definition of wastewater in response to commenter confusion
over its use in the proposed rule. «
Comment; One commenter (A-90-19: IV-D-98) stated that
the EPA should provide a technical and legal rationale for the
broad scope of the definition of "wastewater." The commenter
(A-90-19: IV-D-98) indicated that the Act authorized the EPA
to regulate the emissions of HAP's by setting emission
limitations, but questioned the EPA's authority to broadly
define wastewater as "HAP-containing water or process fluid."
Response; The EPA has modified the definition of
"wastewater" in the final rule. The EPA intends to regulate
any HAP-containing water, raw material, intermediate product,
by-product, co-product or waste material that is managed in an
open wastewater collection and treatment system and has the
potential to emit a significant level of HAP's. The
definition of "wastewater" emphasizes that such streams are
not regulated unless they enter an individual drain system.
The EPA has developed the wastewater requirements in the
HON in accordance with the Act. Because wastewater is a
component of the SOCMI source category, the Act provides the
EPA with the authority to control emissions from wastewater.
The EPA has reviewed the emission data submitted by the SOCMI
4-13
-------
on the 114 questionnaires and has determined that the
wastewater provisions in the RON are sufficient to control air
emissions.
Comment: One commenter (A-90-19: IV-D-86) stated that
the definition of process wastewater, which is found within
the definition of wastewater in §63.101 should not include
non-contact cooling water, utility wastewaters, general site
surface runoff, groundwater, and other non-process wastewaters
generated on-site.
Response: In order to further clarify the definition of
"wastewater" in §63.101 of subpart F, the EPA has separated
the definitions of "process wastewater" and "maintenance
wastewater" from the definition of "wastewater," and deleted
the definition of "maintenance-turnaround wastewater." Each
definition remains in §63.101 of subpart F, but is listed as a
separate entry in the definition list. The EPA agrees that
any waste stream that does not meet the definition of
"wastewater" in §63.101 of subpart F is not subject to the
rule. 'The EPA has not specifically excluded non-contact
cooling water, utility wastewater, and other non-process
wastewater generated onsite because such waste streams will
likely not meet the definition of wastewater in §63.101 of
subpart F. However, if these waste streams exceed the
criteria for flow rate and VOHAP concentration, and are
discharged into an individual drain system, such streams would
be considered wastewater.
Comment; Several commenters (A-90-19: IV-D-53; IV-D-86;
IV-D-102; IV-D-110) (A-90-23: IV-D-20) recommended that the
EPA specify a percentage of water that must be present in a
waste stream in order for it to be considered a wastewater in
§63.101. One commenter (A-90-23: IV-D-20) suggested that a
waste stream must have at least 10 percent water to be
considered wastewater under the HON. One commenter (A-90--19:
IV-D-73) favored changing the definition of a wastewater
stream by incorporating a minimum 90 weight percent water
content. One commenter (A-90-19: IV-D-53) suggested that the
definition of maintenance wastewater in subpart F should
4-14
-------
include "aqueous process fluids" or "draining water used to
wash process fluids." Two commenters (A-90-19: IV-D-33;
IV-D-110) suggested that the EPA clarify the definition of
"wastewater" so that the regulated liquid must be water or
have an aqueous fraction, have contact with process fluids or
organic HAP's, and be destined for disposal.
Response: The EPA does not specify a percentage of water
that must be present in wastewater in order for it to be
considered a wastewater as defined in §63.101 of subpart F.
The EPA clarifies that the water content in a wastewater
stream is not a critical issue; but rather, when any
wastewater is discharged to an individual drain system, it is
essential that HAP emissions be controlled. The EPA. maintains
that regulating wastewater streams based on VOHAP
concentration and flow rate is sufficient to determine whether
a wastewater stream has the potential to emit HAP's. The EPA
intends to regulate both water and process fluid waste streams
that are discharged from SOCMI chemical manufacturing process
units and into an individual drain system and either (1) have
a VOHAP concentration equal to or greater than 5 ppmw and a
flow rate equal to or greater than 0.02 £/m; or (2) have a
VOHAP concentration of at least 10,000 ppmw at any flow rate.
Comment; One commenter (A-90-19: IV-D-86) urged EPA to
be consistent with the Benzene Waste NESHAP and OCPSF
guidelines in the HON definition of wastewater.
Response: The EPA recognizes the importance of
consistency with other regulations and has written the
language in the definitions of the final'rule to be
consistent, where possible, with other regulations. The
commenter did not provide specific information about which
portions of the proposed HON definitions were inconsistent or
how any inconsistency would have a negative impact.
Comment: Several commenters (A-90-19: IV-D-32; IV-D-33;
IV-D-77; IV-D-102) provided background data on the ethylene
oxide production process to illustrate the importance to the
industry that the EPA clarify the definitions of "wastewater,"
"wastewater stream," "individual drain system," and "point of
4^15
-------
generation." The commenters (A-90-19: IV-D-32; IV-D-33;
IV-D-77; IV-D-102) expressed concern that the definition of
"wastewater" could be interpreted to include process water
used as a reactant or a carrier which has not yet left the
process units. For example, one commenter (A-90-19: IV-D-77)
stated that one of the processes necessary for ethylene oxide
production generates water, which is recirculated in the
production process. The commenter (A-90-19: IV-D-77)
expressed concern that the Agency may not classify this part
of the ethylene oxide production as "integral to the process,"
which would result in all water from the process being
classified as a wastewater rather than a recirculated process
fluid.
Response; The EPA clarifies that a waste stream is not
subject to the HON unless it is generated from a chemical
processing unit that meets all applicability criteria
specified in §63.100(b)(1) through (b)(3) of subpart F and
until the waste stream exits the process unit and enters an
individual drain system. In addition, such a waste stream is
not regulated by the HON unless the waste stream meets the
applicability criteria in the definition of wastewater in
§63.101, which specifies that an organic-HAP containing water
or process fluid shall contain either (1) a total VOHAP
concentration of 5 ppmw or greater and have a flow rate equal
to or greater than 0.02 £/m; or (2) contain a total VOHAP
concentration of at least 10,000 ppmw at any flow rate.
The EPA maintains that process water used as a reactant
or a carrier which has not yet left the process unit cannot be
a regulated wastewater under the HON because it has not
entered an individual drain system. The EPA has further
addressed the commenters1 concerns in the responses to each of
the comments on the definitions of "wastewater," "wastewater
stream," and "individual drain system."
Comment; One commenter (A-90-19: IV-D-53) claimed that
the concentration cutoff specified in the definition of
wastewater should refer to total VOHAP concentration and not
total organic HAP's.
4-16
-------
Response; The EPA agrees with the commenter, and the
definition of wastewater has been changed in the final rule to
refer to total VOHAP concentration.
comment: One commenter (A-90-19: IV-D-33) stated that
definitions of "wastewater" in §63.101 and "wastewater stream"
in §63.111 contain several confusing differences and should be
clarified. Several commenters (A-90-19: IV-D-33; IV-D-53;
IV-D-77) (A-90-19: IV-G-5; IV-G-10) provided suggestions to
the EPA about how to clarify the confusion between the two
definitions including: (1) combining the definitions into one
definition in §63.101; (2) consistently using the same terms
and examples; (3) adding the phrase "Group 1 or Group 2"
before the term "wastewater" in the definition of "wastewater
stream" in subpart G to clarify which wastewaters are subject
to the control requirements under the RON; and
(4) consistently using the terms "concentration of total
organic HAP's" and "VOHAP concentration."
Two commenters (A-90-19: IV-G-10) (A-90-23: IV-G-5)
claimed that the definitions of wastewater in §63.111,
§63.132(f)(l), and §63.132(f)(2) of subpart G are
inconsistent. The commenters (A-90-19: IV-G-10) (A-90-23:
IV-G-5) claimed that the flow and concentration cutoffs that
define Group 1 and Group 2 wastewater streams are inconsistent
in these three sections.
Response; In the final rule, the EPA has clarified the
definitions of "wastewater" in §63.101 of subpart F and
"wastewater stream" in §63.111 of subpart G by including all
relevant information about wastewater identification in the
definition of "wastewater" in subpart F, §63.101. The EPA
continues to include the definition of "wastewater stream" in
subpart G, §63.111 because the term is used throughout
subpart G. However, the EPA has simplified the definition of
"wastewater stream" in subpart G, §63.111 by referencing the
definition of "wastewater" in subpart F, §63.101.
The EPA further clarifies the definition of "wastewater"
in subpart F, §63.101 by creating separate definitions for
"process wastewater" and "maintenance wastewater." These
4-17
-------
definitions remain in subpart F, §63.101, but are no longer
located within the definition of "wastewater.11
The definitions for "Group l wastewater stream" and
"Group 2 wastewater stream" remain unchanged in subpart G,
§63.111. As a further clarification, the EPA has added a
definition for "process wastewater stream" in subpart G,
§63.111, which references the definition of "process
wastewater" in subpart F, §63.101.
The EPA also clarifies that the parameters for
determining whether a waste stream is a wastewater and
therefore subject to the RON are intended to be different than
the Group I/Group 2 criteria, which must be checked for each
wastewater stream to determine applicable control
requirements. However, the concentration criteria used in the
definition of wastewater in subpart F is listed in terms of
VOHAP concentration and not total organic HAP concentration in
the final rule.
4.1.3 Definition of "Wastewater Stream"
Comment; "Several commenters (A-90-19: IV-D-1; IV-D-53?
IV-D-73; IV-D-97; IV-D-102) (A-90-23: IV-D-20) requested the
following changes in the definition of "wastewater stream" in
§63.111 of subpart G: (1) the term "indirect contact" should
be deleted because it seemed to include stormwater and
non-contact cooling water; (2) the term "reflux" should be
deleted because it is confusing and usually refers to
materials that will never be discharged to an individual drain
system. One commenter (A-90-19: IV-D-53) claimed that the
current definition of "wastewater stream" in subpart G,
§63.111 can include any process stream that has been in
contact with wastewater. The commenter (A-90-19: IV-D-53)
recommended adding the phrase "destined for disposal" to 'the
definition of wastewater stream. Two commenters (A-90-19:
IV-D-53) (A-90-23: IV-D-20) stated that the definition of
"wastewater stream" should be limited to HAP-containing
aqueous (at least 10 percent water) liquid or aqueous material
separated from the liquid. Several commenters (A-90-19:
IV-D-53; IV-D-73; IV-D-97) (A-90-23: IV-D-20) suggested that
4-18
-------
the definition exclude cooling water blowdown, residuals,
safety showers, eye washes, water from fighting fires, spills,
maintenance wastewater, maintenance-turnaround wastewater,
steam trap condensate, once-through cooling water, and
landfill leachate. One commenter (A-90-19: IV-D-53) claimed
that boiler water is carefully treated to remove impurities
which would cause scaling, and therefore, the EPA did not need
to include steam trap condensate as an example of a wastewater
stream. One commenter (A-90-19: IV-D-73) indicated that it
was unclear whether further control was necessary once the
wastewater stream has been treated according to §63.138(b)
or (c).
Response; The EPA agrees that several of the examples
that were included in the proposed definition of "wastewater
stream" in subpart G, §63.111 were confusing and could have
been misinterpreted to regulate materials that would not
normally contain HAP's or would not be discharged to an
individual drain system. In response to comments on such
waste streams, the EPA has removed "cooling tower blowdown,"
"steam trap condensate," and "reflux" from the definition of
"wastewater stream." Cooling tower blowdown was deleted from
the list of wastewater examples because it is regulated by
§63.104 in subpart F. Steam trap condensate was deleted as an
example of a wastewater stream because the boiler water is
already treated to remove any chemical impurities including
HAP's that could cause scaling. Numerous industry comments
were received that stated "reflux" was a commonly used term,
which refers to a stream that is still within a process unit
and has not been discharged. Such streams do not have a
potential for HAP emissions.
The EPA has clarified the definition of "wastewater
stream" in subpart G, §63.111 by stating that wastewater
stream means a stream that contains only wastewater as defined
in subpart F, §63.101.
4.1.4 Definition of "Individual Drain System"
Comment; Several commenters (A-90-19: IV-D-32; IV-D-32;
IV-D-33; IV-D-53; IV-F-1.2 and IV-F-4) (A-90-23: IV-D-20)
4^-19
-------
requested clarification on the definition of individual drain
system. Two commenters (A-90-19: IV-D-32; IV-D-53) stated
that the proposed definition of "individual drain system"
should be clarified to allow the combination of stormwater,
Group 2 wastewaters, and non-SOCMI wastewaters in collection
systems. Three commenters (A-90-19: IV-D-33; IV-D-53)
(A-90-23: IV-D-20) disagreed with the requirements to
segregate the vapors within the individual drain system
because it would be impractical. One commenter (A-90-19:
IV-D-53) recommended that the requirements to segregate the
vapors within the individual drain system be deleted from the
definition because it may be difficult, because at many SOCMI
facilities, storm water from process areas will enter the
individual drain system. Three commenters (A-90-19: IV-D-33;
IV-D-53) (A-90-23: IV-D-20) suggested adding a sentence to
the definition of individual drain system that exempts drains
and sewers that feed an individual drain system if the system
is designed to isolate the vapor connection between the two.
One commenter (A-90-19: IV-D-32) expressed concern that the
proposed definition of individual drain system would require
the segregation of vapor spaces of sewers carrying non-SOCMI
wastewaters and stormwaters from vapor spaces of sewers in
SOCMI service. Two commenters (A-90-19: IV-D-32) (A-90-23:
IV-D-20) recommended a definition of "individual drain system"
that they determined was consistent with the definition in the
NSPS for petroleum refinery wastewater systems at 40 CFR
part 60 subpart QQQ §60.691.
Response; In response to commenter concerns that the
definition of "individual drain system" is too broad and
inclusive, the EPA restates that segregated stormwater sewers
are not subject to the HON. However, if stormwater is mixed
with HAP-containing wastewater streams in the individual drain
system, then all of the streams must be treated because the
stormwater will be in direct contact with the HAP-containing
wastewater that is subject to the HON.
The EPA continues to require vapors which are generated
in an individual drain system that is subject to the HON to be
4-20
-------
segregated from other drain systems. The EPA requires this
provision in order to eliminate fugitive emissions that would
escape through connecting drain systems. The EPA recognizes
that the definitions of "individual drain system" in the HON
and in the Petroleum Refinery NSPS (40 CFR subpart QQQ) have
different wording. The definition of "individual drain
system" in the Petroleum Refinery NSPS includes drains,
junction boxes, and associated sewer lines, and extends down
to the point where the wastewater enters the oil-water
separator. The HON, however, requires control from the point
of generation through treatment that meets specified levels.
Both regulations are the same conceptually. That is, both
regulations require emission suppression from the point of
first control (i.e., drain hub for the Petroleum Refinery NSPS
and the exit of the process unit equipment for the HON) to the
treatment unit (i.e., oil-water separator for the Petroleum
Refinery NSPS and options in §63.138 for the HON). Therefore,
the EPA maintains that the definition of "individual drain
system" in the HON should be and is different from the
definition in the Petroleum Refinery NSPS.
Comment; One commenter (A-90-19: IV-D-33) stated that
the definition of "individual drain system" should be modified
to mean a system used to convey wastewater streams from a
process unit, product or feed storage tank, or emission
control unit to a waste management unit. The commenter
(A-90-19: IV-D-33) stated that an individual drain system
should not be a system that conveyed wastewater from one waste
management unit to another waste management unit. Another
commenter (A-90-19: IV-D-32) stated that §63.138(b) and (c),
which provide the treatment options for Group 1 wastewater
streams, should clearly state that after Group 1 wastewater
streams are treated to target levels they are no longer
regulated.
Response; The EPA continues to regulate individual drain
systems that convey wastewater from one waste management unit
to another waste management unit, because HAP's can be emitted
between the units if the wastewater stream is uncontrolled.
4-21
-------
For example, a wastewater stream that first passes through an
oil-water separator and is then conveyed to a steam stripper
must be conveyed in a controlled drain system when it leaves
the oil-water separator until it enters the steam stripper in
order to prevent HAP emissions between the waste management
units. The EPA also notes that if a wastewater stream first
enters a steam stripper, which treats the wastewater to comply
with the HON, the wastewater may be conveyed in an
uncontrolled drain system to any other treatment system such
as a biological treatment unit.
The requirements for Group 1 wastewater streams in
§63.138(b)(1) and (c)(1) state that a Group 1 wastewater
stream must be either recycled to the process or treated to a
target level. Section 63.138 also lists the requirements for
any residuals that are removed from the Group 1 wastewater
stream during the treatment process. After a Group 1
wastewater stream is treated in accordance with §63.138 it is
no longer subject to the HON. However, treated Group 1
wastewaters may be subject to other regulations (e.g., they
may require NPDES discharge permits).
4.1.5 Clarification of "Point of Generation"
Comment; Several commenters (A-90-19: IV-D-32; IV-D-53;
IV-D-110) stated that the definition of "point of generation,"
should specifically include provisions to allow worker health
and safety, and other applicable State and Federal
regulations, to be considered (e.g., where OSHA regulations
may preclude flow monitoring and sampling of wastewater
because of the presence of adjacent equipment or wastewater
characteristics that could endanger worker health and safety).
Three commenters (A-90-19: IV-F-1.2 and IV-F-4; IV-D-32;
IV-D-34) stated that the proposed definition for "point of
generation" was confusing because of the inclusion of the
phrase "integral to the process unit". Three commenters
(A-90-19: IV-D-34; IV-D-77; IV-D-102) requested clarification
of "integral to the process unit." Some equipment may be
essential to a unit's normal mode of operation, but may be
removed for short periods during maintenance without shutting
4r22
-------
down the entire process. One commenter (A-90-19: IV-D-102)
stated that such equipment should be considered an integral
part of the process. One commenter (A-90-19: IV-D-53)
claimed that some control devices are integral to the process
unit, because they cannot be shut down without violating a
permit.
Several commenters (A-90-19: IV-D-32; IV-D-33; IV-D-46;
IV-D-53; IV-D-62; IV-D-73; IV-D-77; IV-D-79; IV-D-92;
IV-D-110; IV-D-112; IV-F-1.2 and IV-F-4) (A-90-23: IV-D-17)
suggested that the point of generation should be designated as
the first point downstream of a process unit where emissions
can enter the atmosphere. One commenter (A-90-19: IV-D-53)
claimed that there is no potential for emissions before the
wastewater enters the process sewer because many facilities
have emissions-suppressed piping systems. The commenter
(A-90-19: IV-D-53) claimed that this definition of "point of
generation" would.allow direct sampling and flow monitoring.
Several commenters (A-90-19: IV-D-32; IV-D-62; IV-D-77;
IV-D-102; IV-D-110) stated that this approach would allow
facilities where waste is hardpiped to a sewer to maintain the
current configuration without equipment modification. One
commenter (A-90-23: IV-D-9) claimed that, for some processes,
it will be impossible to determine the flow rate and
concentration at the point of generation because sampling will
be too difficult. Three commenters (A-90-19: IV-D-53;
IV-D-73; IV-D-110) claimed that the proposed "point of
generation" may be in closed piping or closed piping routed to
controls, and these piping systems may have to be disconnected
or a process unit shut down to determine whether a stream is a
Group 1 or Group 2.
Three commenters (A-90-19: IV-D-32; IV-D-53; IV-D-75)
claimed that the EPA should not be concerned with dilution of
Group 1 streams because non-contact cooling waters and
wastewaters are required to be separated and processes will
not generate large enough quantities of noh-HAP-containing
wastewaters to dilute Group 1 streams. Two commenters
(A-90-19: IV-D-32; IV-D-75) reasoned that the incompatibility
4-23
-------
of the streams and the costs associated with this method of
wastewater management would discourage mixing. One commenter
(A-90-19: IV-D-53) cited a report entitled "EPA, Contractors
Engineering Report, Analysis of Organic Chemicals and
Plastics/Synthetic Fibers Industries, Appendix S," Contract
No. 68-01-6024, Effluent Guidelines Division, November 16,
1981, which presents process flow diagrams of SOCMI wastewater
systems. The commenter (A-90-19: IV-D-53) claimed that these
diagrams should be used to determine which wastewater streams
are subject to the HON. Two commenters (A-90-19: IV-D-53;
IV-D-112) stated that a decrease in the level of HAP's in
wastewater due to mixing with other wastewater streams usually
results in a decrease in overall HAP emissions. One commenter
(A-90-19: IV-D-46) claimed that reduction of pollutants may
occur in the hard-piped systems because chemicals may continue
to react due to mixing.
Response; Although the final rule does not change the
conceptual basis of the point of generation, the definition
has been simplified and the phrase "integral to the process
unit" has been deleted from the definition of point of
generation. The EPA has determined that the point of
generation means the location where process wastewater exits
the chemical manufacturing process unit equipment. The
primary function of chemical manufacturing process unit
equipment is to produce chemical products. Wastewater
management units may, in the process of treating wastewater,
produce small amounts of product that can be recycled to the
process. For example, steam strippers would generally be
wastewater treatment units because they would not produce an
appreciable amount of product.
The final rule allows the owner or operator to determine
the characteristics of a wastewater stream (1) at the point of
generation, or (2) downstream of the point of generation if
corrections are made f r changes in flow rate and VOHAP
concentration. Such changes include losses by air emissions,
reduction of VOHAP concentration or changes in flow rate by
mixing with other wastewater streams, and reduction in flow
4-24
-------
rate or VOHAP concentration by treating or otherwise handling
the wastewater streams to remove or destroy HAP's. The EPA
has concluded that by including two options for how to
determine the characteristics of a wastewater stream, the need
for specifying whether a piece of equipment is integral to the
process unit is irrelevant because HAP emissions will be
accounted for if the Group I/Group 2 determination is made
downstream of the point of generation and an accurate flow
rate and VOHAP concentration can be determined.
In response to comments about sampling within closed
piping, the EPA agrees that options must be available to
ensure worker safety, and clarifies that the owner or operator
has several options under the HON when determining flow rate
and concentration at the point of generation. Besides
sampling, the owner or operator has the option to determine
VOHAP concentration using process knowledge and bench-scale or
pilot-scale test data, instead of sampling at the point of
generation as summarized in §63.144(b) of the final rule. In
§63.144(c) of the final rule, the EPA also allows other
options for determining flow rate, including use of process
knowledge based on production capacity and historical records.
In addition, the EPA has added a provision in §63.144(d) of
the final rule to allow an owner or operator to designate as a
Group 1 wastewater stream a single wastewater stream or a
mixture of wastewater streams. By choosing this option, an
owner or operator is not required to make a Group I/Group 2
determination. The owner or operator who elects to use this
option must suppress emissions from the point(s) of generation
by complying with all requirements in §§63.133 through 63.137
and must treat the stream in accordance with the requirements
for Group 1 wastewater streams in §63.138. The EPA has added
the option of designating a single wastewater stream or
mixture of wastewater streams as a Group 1 wastewater stream
because several commenters, who have facilities where HAP
emissions are already suppressed from the point of generation
to a downstream location, will not be required to determine
wastewater stream characteristics at each point of generation.
4-25
-------
The owner or operator will still need to determine stream
characteristics for the point of generation where stream(s)
are designated as Group 1 wastewater streams in order to
ensure that the stream is treated in accordance with §63.138.
The primary difference between the final rule and the proposed
rule is the addition of the option to designate Group 1
wastewater streams.
The EPA continues to prohibit dilution of Group 1
wastewater streams to meet compliance. The owner or operator
who elects to determine flow rate and concentration for a
mixture of wastewater streams at a location downstream of the
point(s) of generation, and determines that the mixture of
wastewater streams is a Group 2 wastewater stream, must verify
whether each wastewater stream in the mixture is Group 1 or
Group 2. All Group 1 streams in the mixture are subject to
the control requirements of the wastewater provisions in
§63.133 through §63.139. Commenters provided no data that
dilution reduces the fraction of individual HAP's emitted from
a given wastewater stream. The EPA maintains that the
emission estimates for the HON represent reasonable estimates
of the concentration of HAP's in the wastewater system and
that the benefits of wastewater controls are not overstated.
Comment; One commenter (A-90-19: IV-D-102) stressed
that the wastewater definition should clarify that a. material
is subject to the HON only at the point that it exits a
process unit and enters an individual drain system. One
commenter (A-90-19: IV-D-77) stated that the point of
generation should be established after the last product
recovery device and before the discharge to a wastewater
treatment unit or disposal system.
Response: The EPA clarifies that a wastewater stream is
subject to the wastewater provisions in the HON (1) where it
exits the process unit equipment, and (2) if it meets the
criteria in the definition of "wastewater" in §63.101 of
subpart F. The EPA continues to allow the owner or operator
to recycle wastewater and recover HAP's as a compliance
option; however, the EPA emphasizes that when a Group 1
4-26
-------
process wastewater exits a piece of process unit equipment,
HAP emissions must be.suppressed until the wastewater stream
meets the treatment requirements.
Comment: One commenter (A-90-23: IV-D-9) suqgested that
the EPA define the point of generation as the point at which
waste is combined with other waste and no longer has the
potential for reuse or recycling. Two commenters (A-90-19:
IV-D-32; IV-D-75) claimed that this definition would be
consistent with RCRA and would encourage pollution prevention
and recycling. Several commenters (A-90-19: IV-D-32;
IV-D-34; IV-D-53; IV-D-62; IV-D-75; IV-D-77; IV-D-110)
(A-90-23: IV-D-9; IV-D-20) stated that the definition of
"point of generation" could be simplified by using the
approach that is used to define a solid waste under RCRA,
which would be the first air-water interface after the stream
reaches the point where it is "destined for disposal." One
commenter (A-90-19: IV-D-89) claimed that the definition of
point of generation is inconsistent with RCRA and the
Pollution Prevention Act and leads to a definition of waste
which is not consistent with the Act, RCRA, and other air
regulations such as NSPS subpart QQQ. One commenter (A-90-23:
IV-D-9) claimed that the definition of point of generation
will discourage waste recovery operations. One commenter
(A-90-19: IV-D-89) claimed that the EPA defines the point of
generation as the first point where a stream must be
controlled, regardless of its potential to emit HAP's. Two
commenters (A-90-19: IV-D-89; IV-D-92) claimed that the
controls required at the point of generation and the
definition of point of generation discourage, inhibit, and may
disallow the reuse, reprocessing, or recycling of materials.
One commenter (A-90-19: IV-D-89) argued that the material
which is recycled to a process unit does not have the
potential to emit HAP's and claimed that a resource recovery
unit should be considered a process unit and material streams
exiting the unit should only be subject to controls if they
have the potential to emit HAP's. One commenter (A-90-19:
IV-D-53) claimed that units used for recycling of wastewater
4-27
-------
will be regulated under the current definition of point of
generation, even if the entire recycle system is suppressed.
Response; The EPA disagrees with the commenters who
claim that the HON does not promote recycling and reuse of
materials: Owners or operators are encouraged to recycle
wastewater and residuals. Although control of HAP emissions
is required from the point of generation, only Group 1 process
wastewater streams, which the EPA has determined are a source
of HAP emissions, require control. If an owner or operator
generates Group 2 wastewater streams, the owner or operator is
not obligated to control such streams.
The EPA maintains that the definition of "point of
generation" in the HON is consistent with the concept of point
of generation in RCRA. Because the point of generation is not
explicitly defined in the RCRA regulations, but is commonly
known to be the point at which a waste is destined for
disposal, the EPA considers the HON and RCRA to be consistent.
When a Group 1 wastewater stream exits any process unit
equipment and enters an individual drain system, the stream
must be controlled, treated, and disposed to suppress and
destroy HAP's contained in the wastewater stream. The
commenters did not provide examples where the definition of
"point of generation" in the HON would be inconsistent with
the Petroleum Refinery NSPS in 40 CFR part 60, subpart QQQ.
Comment; Two commenters (A-90-19: IV-D-32; IV-D-53)
contended that the EPA's concern for dilution and combination
of waste streams is based on the erroneous assumption that
equilibrium between the liquid phase and the vapor phase of
VOHAP's occurs in the collection system.
One commenter (A-90-19: IV-D-53) indicated that the
concentration of the VOHAP in the wastewater, and not the
total mass of HAP in the wastewater, affects the emissions*
when vapor-liquid equilibrium is not reached. The commenter
(A-90-19: IV-D-53) asserted that a decrease in concentration
leads to a decrease in the driving force for volatilization,
and therefore, the commenter (A-90-19: IV-D-53) claimed that
dilution decreases HAP emissions.
4--2S
-------
Response; Based on information provided by commenters,
the EPA has modified the collection system emission models.
In the final rule, the EPA bases the emission estimates on
50 percent of equilibrium being achieved rather than
equilibrium.
The EPA maintains, however, that mixing Group 1 and
Group 2 wastewater streams in an individual drain system does
not reduce the fraction of HAP's emitted from the system. The
emission estimates for SOCHI facilities are based on
characterization of the wastewater collection and treatment
systems and the mixture of Group 1 and Group 2 process
wastewaters that are managed in these systems. The EPA did
not estimate emissions based on Group 1 streams alone so the
reason for the comment is not clear. The rule is based on
managing and treating Group 1 wastewater streams to remove or
destroy HAP's which will reduce not only the VOHAP
concentration but also the mass of HAP's in the wastewater
stream. The emission reduction that will occur as a result of
this treatment is due only to the reduction in the mass of
HAP's, not the VOHAP concentration.
Comment; One commenter (A-90-23: IV-D-17) claimed that
the current definition of "point of generation" will result in
a significant increase in sampling, analysis, and
recordkeeping. Although the HON allows the use of process
knowledge to determine VOHAP concentration at the point of
generation, the commenter (A-90-23: IV-D-17) claimed that
regulatory agencies rarely accept process knowledge without
analyses to demonstrate compliance.
Response: The EPA has tried to minimize sampling
requirements by not requiring sampling at each point of
generation. The HON includes additional sampling options
downstream of the point of generation for a single wastewater
stream or after mixing different wastewater streams.
Additionally, the HON allows the owner or operator to
designate that a wastewater stream or combination of
wastewater streams is a Group 1 wastewater stream without
sampling. The implementing agency may require additional data
4-29
-------
if an owner or operator uses process knowledge to determine
whether the HON applies to a particular waste stream; however,
the EPA continues to allow owners or operators to use process
knowledge.
Comment; One commenter (A-90-19: IV-D-98) contended
that by determining the applicability of the emissions
standard at the "point of generation", the EPA is regulating
HAP's before they can be emitted and should explain its
authority to regulate emissions prior to the point of the
first air/water interface.
Response; Emissions enter the atmosphere at the first
point where an air/water interface exists. However, the EPA
did not define the point of generation as the point where
emissions can first enter the atmosphere, because a stream
with a high VOHAP concentration may be mixed with more dilute
streams prior to reaching the first air/water interface.
Mixing a Group 1 stream with a Group 2 stream may result in a
single Group 2 stream because of dilution. The total mass of
HAP's however, is unaffected by dilution. Therefore, the HAP
emissions from the combined streams will be the same or
greater than the original Group 1 stream, depending on the HAP
mass contribution of the Group 2 stream. Consequently, the
EPA defined the point of generation at a point before dilution
can.occur in order to control emissions from all Group 1
streams.
Furthermore, HAP emissions are not "regulated" prior to
the point of the first air/water interface. Rather,
determination of whether or not a wastewater stream requires
control may be done prior to the point of the first air/water
interface. This determination must be performed before the
wastewater stream is diluted and before any of the HAP's in
the wastewater have a chance to volatilize. If a wastewater
stream is determined at the point of generation to be a
Group 1 wastewater stream, control is not required until the
first air/water interface. If a Group 1 wastewater stream is
never exposed to the atmosphere, control is not required at
all.
4-30
-------
4.1.6 Definition of "Waste Management Unit"
romment: One commenter (A-90-19: IV-D-54) contended
that since §63.138 specifies requirements for wastewater
treatment processes, it should not also impose requirements on
waste management units, which are regulated by §63.133 through
§63.137. The commenter (A-90-19: IV-D-54) is confused
because the use of the term "waste management unit" in §63.138
seems to include wastewater tanks, surface impoundments, etc.,
which each have individual provisions specified in §63.133
through §63.137. The commenter (A-90-19: IV-D-54) stated
that if §63.138 is meant to impose additional requirements on
units regulated under §63.133 through §63.137, then the
additional requirements should be specified in the individual
sections and the term "waste management unit" should be
deleted.
Response: The EPA clarifies that the definition of
"treatment process" is a subset of "waste management unit" and
both terms are defined in §63.111 of subpart G. The EPA
continues to use both terms in §63.138 because the term "waste
management unit" defined in §63.111 of subpart G is not
limited to those collection and conveyance units that are
specified in §§63.133 through 63.137. In fact, a waste
management unit could be used to comply with the provisions of
§63.138. Therefore, the EPA continues to use both terms
throughout the regulation.
4.1.7 Solvent Use as a Feedstock
Comment; One commenter (A-90-19: IV-D-60) expressed
concern that used solvents that are routinely collected in
containers and either sent to offsite locations or used as a
feedstock onsite, and which never enter an individual drain
system, may be construed to be wastewater streams under the
proposed definition of wastewater. The commenter (A-90-19:
IV-D-60) provided an example scenario and requested
clarification.
Response: The definition of "wastewater" in §63.101 of
subpart F clearly states that a wastewater stream must enter
an individual drain system in order to be considered a
4-31
-------
wastewater. If the solvents are used as feedstocks onsite or
are sent offsite, then such streams would not be wastewaters.
4.1.8 Wastewater Generated from Fire Fighting
foment; Two commenters (A-90-19: IV-D-33; IV-G-4)
stated that §63.100(b)(3)(vi), which lists materials that are
not subject to control under the wastewater provisions, should
also exclude water generated from both fire fighting and
deluge systems. Another commenter (A-90-19: IV-D-34) stated
that covering drain systems may result in safety hazards
during non-routine conditions (e.g., deluge water during fire
or spill events). The commenter (A-90-19: IV-D-34)
recommended that a provision be added for sources to obtain a
waiver for process-specific safety reasons.
Response; The EPA agrees with the commenter and has
added as §63.100(f)(3) of subpart F, an exclusion from the HON
wastewater provisions for water that is generated by fire
fighting and deluge systems and is discharged to a segregated
sewer. It is unclear why the commenter states that covering
drain systems may result in unsafe conditions during non-
routine conditions. The EPA anticipates that properly
designed wastewater collection and treatment systems will be
equipped to handle non-routine conditions and that the
installation of covers on drain systems will present no
additional hazards.
4.1.9 Relationship Between Wastewater Tank and Storage Vessel
Provisions
Comment; Several commenters (A-90-19: IV-D-17; IV-D-32;
IV-D-33; IV-D-54; IV-D-64; IV-D-73; IV-D-75; IV-D-112)
(A-90-23: IV-D-2; IV-D-20) argued that control of wastewater
tanks should be managed under the storage vessel provisions
instead of the wastewater provisions. The commenters
(A-90-19: IV-D-17; IV-D-32; IV-D-33; IV-D-54; IV-D-64;
IV-D-73; IV-D-75; IV-D-112) (A-90-23: IV-D-2; IV-D-20) stated
that having two different requirements for tanks does not make
sense in terms of the relative potential for the two types of
tanks to emit HAP's. Three commenters (A-90-19: IV-D-17;
IV-D-32) (A-90-23: IV-D-20) stated that the EPA should change
4-32
-------
the definition of "storage vessel" in §63.101 to include
product storage tanks and wastewater storage tanks. One
conunenter (A-90-19: IV-D-73) suggested including
subparagraphs (1) , (2), and (3) of the storage vessel
definition from subpart F in the definition of wastewater
tank.
Several commenters (A-90-19: IV-D-32; IV-D-75; IV-D-112)
(A-90-23: IV-D-2) stated that the requirements should be
based on the partial pressure of the HAP's in the tank. One
commenter (A-90-19: IV-D-31) claimed that there is a
potential for wastewater vessels to be classified as Group 1
although they would be classified as Group 2 storage vessels
based on partial pressure. The commenter (A-90-19: IV-D-31)
provided a hypothetical example of such a case. The commenter
(A-90-19: IV-D-31) also claimed that not considering partial
pressure for wastewater tanks will result in considerable
expense to achieve marginal reductions in HAP emissions. The
commenter (A-90-19: IV-D-31) stated that by using the storage
vessel definition, the cost of controlling wastewater tanks
would be reduced. The commenter (A-90-19: IV-D-31) supported
the use of proposed wastewater tank definitions in cases where
determining the total HAP partial pressure is difficult
because of a highly mixed matrix or highly variable
concentrations. Two commenters (A-90-23: IV-D-2; IV-D-77)
claimed that the control of wastewater tanks should also be
based on the size of the tank. One commenter (A-90-19:
IV-D-32) provided partial pressure data for chemicals in
strippability groups A, B, and c.
One commenter (A-90-19: IV-D-64) requested that the EPA
set de minimis cutoffs based on size and vapor pressure for
wastewater tanks in §63.133 and surface impoundments in
§63.134, and stated that surface impoundments and controlled
oil water separators should be regulated as a Group 1 or
Group 2 storage vessels, based on their capacity and maximum
total HAP vapor pressure.
Response; The EPA agrees that it is appropriate to
regulate wastewater tanks based on their potential for HAP
4-33
-------
emissions. Thus, the EPA has added language to the wastewater
tank provisions in §63.133 that reflects the tank capacity and
vapor pressure criteria used in the HON storage vessel
provisions. The EPA also felt that it would be appropriate
for the final HON wastewater provisions to be consistent with
the proposed RCRA tank and container requirements, which will
be in 40 CFR part 264 subpart CC.
In the final rule, the owner or operator must determine
whether their wastewater tanks meet the criteria in table 4-1
of this section (in the final rule as table 10 of subpart G),
which specifies both tank capacity and vapor pressure
criteria.
TABLE 4-1. WASTEWATER TANK CAPACITY AND
VAPOR PRESSURE CRITERIA
Tank capacity (m3) Vapor pressure (kPa)
75< and <151 <13.1
<5.2
The owner or operator must make this determination for
any wastewater tank that manages Group 1 wastewater streams or
residuals removed from such streams at both new and existing
sources. If a wastewater meets the criteria specified in
table 10 of subpart G, then the owner or operator must operate
and maintain a fixed roof. If a wastewater tank exceeds the
criteria specified in table 10 of subpart G, then the owner or
operator must comply with paragraphs (b) through (h) of
§63.133 and shall operate and maintain one of the emission
control techniques specified in §63.133(a)(2)(i) through
(a)(2)(iv).
4.1.10 Previously Installed Steam Strippers
Comment; Three commenters (A-90-19: IV-D-18; IV-D-32;
IV-D-110) stated that steam strippers installed for other
purposes than compliance with the wastewater provisions in the
HON, including meeting the requirements of other regulations,
4-34
-------
should be grandfathered and limited to treatment of
wastewaters for which they were designed (e.g., OCPSF effluent
limitation guidelines, Benzene Waste NESHAP, pretreatment
standards, and corporate waste minimization targets).
Response; The EPA is directed by the Act to control HAP
emissions from wastewater. Although the rules mentioned by
the commenters were not originally intended to control HAP
emissions, in some cases, the rules may result in a reduction
in HAP emissions from wastewater. The EPA has reviewed the
overlap issues associated with other regulations including
OCPSF effluent limitations, the Benzene Waste NESHAP, and
NPDES pretreatment standards. However, the EPA has concluded
that in most cases the EPA is unable to provide an overall
exemption for steam strippers that were installed to comply
with other regulations.
In §63.110 of subpart G of the final rule, the EPA
provides specific guidance about several regulatory overlap
issues by (1) specifying a combination of different
requirements from the overlapping rule; (2) deferring to the
requirements of one rule; or (3) allowing a case-by-case
determination. Through these approaches, the EPA can ensure
compliance with the HON and minimize duplicative effort.
Comment; One commenter (A-90-19: IV-D-85) disagreed
with the provisions in §63.110(b)(2) of the proposed rule
which exempt vents in wastewater treatment processes from the
process vents requirements. The commenter (A-90-19: IV-D-85)
stated that the EPA offers no justification for exempting
vents associated with wastewater treatment from the
requirements for process vents and that the maximum achievable
emissions reduction standard precludes this exemption.
Response; Proposed §63.110(b)(2) stated that vents from
recovery devices installed to control emissions from treatment
operations that are in compliance with the requirements in
§63.133 through §63.147 are not regulated as process vents.
Rather, such vents have separate regulatory requirements and
must achieve a 95 percent HAP removal. Therefore, the EPA has
neither exempted these vents from control nor dually regulated
4^35
-------
such vents under the process vent provisions. The EPA
maintains that the flow and concentration of HAP's that will
be removed from the wastewater and therefore vented to an air
emissions control device will be low compared to that in a
vent stream from a reactor, air oxidation reactor, or
distillation unit. Thus, if such a stream were regulated
under the process vent provisions, it might not meet the
process vent control criteria because of having a high TEE
index value. In this case, the process vent would not require
control. The EPA has determined, however, that it is
appropriate to require 95 percent control of such streams
since low concentration streams cannot typically be controlled
to levels of 98 .percent.
4.1.11 Control of Maintenance-Related Wastewater
Comment; Several commenters (A-90-19: IV-D-32; IV-D-34;
IV-D-36; IV-D-62; IV-D-79; IV-D-86; IV-D-89; IV-D-92)
(A-90-23: IV-D-20) asserted that cooling towers should not be
subject to the HON because a MACT Standard for Industrial
Process Cooling Towers will be developed by November 15, 1994.
Response; The MACT Standard for Industrial Process
Cooling Towers will regulate only hexavalent chromium
emissions from cooling towers. The HON regulates organic HAP
emissions. Additionally, the EPA notes that emissions from
cooling towers are caused by leaks, which may be occurring
throughout the cooling process, and not just in the cooling
tower. Leaks of this nature would not be addressed by the
MACT Standard for Industrial Process Cooling Towers.
Comment; Two commenters (A-90-19: IV-G-10) (A-90-23:
IV-G-5) expressed confusion regarding which maintenance
wastewaters were subject to the HON. The commenters (A-90-19:
IV-G-10) (A-90-23: IV-G-5) claimed that if process wastewater
includes maintenance and turnaround wastewater as established
by §63.110(e), then most hydrocarbon drains and water drains
where water contacts process fluids would be subject to
subpart G. The commenters (A-90-19: IV-G-10) (A-90-23:
IV-G-5) expressed particular concerned with a phenolic sewer
system, claiming that the sewer system would not be subject to
4-36
-------
the RON during normal operation because it only contains
phenol, but would be subject to the HON during maintenance and
turnaround because it is flushed with cumene. The commenters
(A-90-19: IV-G-10) (A-90-23: IV-G-5) claimed that
controlling the phenolic sewer would be the greatest expense,
and that the impact of the HON on the phenolic sewer system
was not evaluated. The commenters (A-90-19: IV-G-10)
(A-90-23: IV-G-5) provided details on the phenol unit
process.
Response; The proposed regulation did not include
routine maintenance wastewater and maintenance-turnaround
wastewater in the definition of process wastewater in §63.101.
Routine maintenance wastewater, maintenance-turnaround
wastewater, and process wastewater were listed as three
separate types of wastewater in the definition of "wastewater"
in §63.101 of the proposed rule. In the final rule, these
terms have been clarified. The maintenance wastewater
requirements have been moved to §63.105 of subpart F, and
maintenance wastewater is now defined separately from
wastewater in §63.101 of subpart F.
The phenolic sewer system is not subject to the HON
wastewater provisions during normal operation, because phenol
is not a regulated HAP for wastewater. Cumene and
acetophenone are on the list of HAP's regulated for
maintenance wastewater, so the maintenance operations are
subject to the HON. However, the requirements for routine
maintenance wastewater are now the same requirements as those
proposed for maintenance-turnaround wastewater. There are no
longer any specific control requirements for routine
maintenance wastewater. The requirements of both types of
maintenance wastewaters are addressed in the facility's start-
up, shutdown, and malfunction plan.
Comment; One commenter (A-90-19: IV-D-33) stated that
the EPA should clarify that §63.102(b)(1) and (b)(2) refer to
those HAP's listed in §63.104.
Response; In the final rule, the maintenance wastewater
provisions have been moved from §63.102(b)(1) of subpart F to
4-37
-------
a separate section, §63.105, entitled, maintenance wastewater
requirements. In the final rule, the heat exchange system
requirements have also been moved from §63.102(b) (2) to a
separate section, §63.104, entitled, heat exchange system
requirements. The provisions in §63.104 and §63.105 of
subpart F clarify which HAP's are regulated for heat exchange
systems and maintenance activities.
Comment; Two commenters (A-90-19: IV-F-1.2 and IV-F-4;
IV-D-112) stated that the proposed HON included several
insignificant wastewater streams including infrequently
generated sources such as maintenance-related streams, which
one commenter (A-90-19: IV-F-1.2 and IV-F-4) declared should
not be included in the regulation. The commenter (A-90-19:
IV-F-1.2 and IV-F-4) pointed out that the Benzene Waste NESHAP
excludes routine maintenance streams, and that the HON should
be consistent on this point.
Response; Although several commenters contended that the
Benzene Waste NESHAP does not regulate maintenance wastewater
streams, the EPA notes that the Benzene Waste NESHAP controls
all waste and wastewater streams if the facility's total
annual benzene exceeds 10 Mg/yr. Furthermore, the Benzene
Waste NESHAP does not contain a specific exclusion of
maintenance wastewater streams. In the final rule, the EPA
continues to require good air pollution control pract^jes for
maintenance-related wastewater streams, but is not requiring
owners or operators to achieve -.? ^acific removal efficiencies.
Comment: One commenter (A-^0-19: IV-G-4) stated that
wastewater that is generated as part of an unplanned shutdown
should be exempt from control requirements. The commenter
(A-90-19: IV-G-4) suggested that the EPA should complete the
studies required by section 112(d) of the Act to determine
whether such control is appropriate. If such a provision was
deemed necessary, the commenter ^A-90-19: IV-G-4) suggested
that the provision be added to "63.102(b) or in the start-up,
shutdown, and malfunction prov: ,ons of the General Provisions
(subpart A) for part 63.
4-38
-------
Response: Wastewater that is generated during an
unplanned shutdown is maintenance-turnaround wastewater as
described in the definition of "wastewater" in §63.101 of
subpart F in the proposed rule. Maintenance-turnaround
wastewater includes maintenance wastewater generated during
planned and unplanned shutdowns. There were not any control
requirements for maintenance-turnaround wastewater in the
proposed rule. The requirements for maintenance-turnaround
wastewater in the proposed and final rules are the same and
are addressed in the facility's start-up, shutdown, and
malfunction plan.
4.1.12 indirect Discharges
Comment; One commenter (A-90-19: IV-D-86) stressed that
indirect discharges should not be subject to the HON for two
reasons. The commenter (A-90-19: IV-D-86) claimed that the
EPA had not correctly estimated emissions from systems which
discharge to POTW systems and that the EPA had not considered
the effect of flow rate on Fe. The commenter (A-90-19:
IV-D-86) claimed that these systems experience vapor
suppression by dilution with sanitary wastewater. The
commenter (A-90-19: IV-D-86) provided a derivation which the
commenter (A-90-19: IV-D-86) claims relates Fe to flow rate.
The commenter (A-90-19: IV-D-86) suggested that the EPA do a
sensitivity analysis to determine the effect of the assumed
design size criteria on transfer surface area and transfer
coefficients which are used to develop Fe.
Response; It is assumed that the commenter is referring
to the claims regarding the effect of flow rate and dilution
on Fe when stating that the EPA incorrectly estimated
emissions from systems discharging to POTW's. The derivation
presented by the commenter, which presents Fe as a function of
flow rate, ignores the effect of increased flow rates on
transfer surface area. As transfer surface area increases, Fe
increases. The derivation presented by the commenter assumes
that the waste management unit remains the same size
regardless of flow, which requires a proportional decrease in
the residence time of the waste in the waste management unit.
4-39
-------
The residence time is the volume of the waste management unit
divided by the volumetric flow rate of the waste. In actual
practice, a waste management unit of any given size has
limited flexibility with regard to the flow rate of waste
which it can accommodate. Therefore, larger waste flows
require larger waste management units. Alternatively,
multiple smaller waste management units may be employed. Tha
end result is an increase in surface area which may result in
an increase in Fe depending on the change in residence time.
An increase in residence time will augment the increase in Fe
resulting from any increase in transfer surface s.sa. A
decrease in residence time will offset the increase in Fe
resulting from any increase in transfer surface area.
Furthermore, the EPA does not recognize dilution as a viable
treatment option.
4.1.13 Clarification of Cooling Tower System
Comment; One commenter (A-90-19: IV-D-34) stated that
the EPA has not completed the analysis required under
section 112(d) of the Act to include water from heat exchange
systems (i.e., water from cooling towers and once-through
cooling water systems) in the HON. Three commenters (A-SO-19:
IV-D-34; IV-D-50; IV-D-54) also stated that the EPA does not
provide any information in the preamble or BID regarding
emissions, the floor, or alternate control strategies from
cooling water. The commenter (A-90-19: IV-D-34) stated that
the CMA's study of leaks in chemical industry heat exchange
systems (Cooling Tower Project Report, June 1992) indicates
that heat exchanger leaks are "a rare occurrence" ranging from
3.4 to 12.9 years. The commenter (A-90-19: IV-D-34)
suggested that the EPA delete all proposed controls on
recirculating cooling water systems.
Response; Although leaks may not occur every year in a
heat exchange system, the EPA has shown that leaks as small as
1 ppm can cause considerable emissions if left undetected.
For example, an average size cooling tower (15,000 gpm) will
emit almost 3 tons of organics in one month if a leak of 1 ppm
is not detected. Table 4-2 of this chapter summarizes the
4-40
-------
possible emissions from heat exchange systems with a leak of
I ppm. Larger leaks will produce proportionately larger air
emissions. For example, a leak of 3 ppm will produce
emissions three times as great as those presented in the
table.
TABLE 4-2. EMISSIONS FROM HEAT EXCHANGE SYSTEMS (TONS)
Flow Rate (gpm)
20,000
10,000
5,000
2,000
Time Period (Months)
1
3.7
1.8
0.92
0.37
3
11
5.5
2.75
1.1
6
22
11
5.5
2.2
12
44
22
11
4.4
The EPA met all statutory criteria in its analysis of
whether or not to regulate HAP emissions from heat exchange
systems. The heat exchange system provisions, which control
leaks from cooling towers and once-through cooling water
systems, require the owner or operator to comply with
monitoring, recordkeeping, and reporting requirements. The
heat exchange system requirements are a specific example of an
emission control program necessary for the source to be
operated in a manner consistent with good air pollution
control practices as specified in the General Provisions
§63.6(e)(1)(i). These provisions were specified in the rule
based on the potential for high HAP emissions. The cost of
monitoring the system for leaks was considered as part of the
monitoring, recordkeeping, and reporting requirements in the
rule; and thus, met the statutory criteria.
With regard to the floor determination, the EPA is not
required to determine a floor for heat exchange systems. The
EPA is required to ensure that the standard for heat exchange
systems is at least as stringent as the floor. The EPA has
reviewed currently available information and has determined
that leaks in heat exchange systems are more common than the
commenter suggested.
4-41
-------
One commenter (A-90-19: IV-D-73) supported the
proposed definition of "heat exchange system" stating that it
implies that the scope of a heat exchange system can be
defined by the source to be the entire cooling tower system
rather than a single heat exchanger. However, the commenter
(A-90-19: IV-D-73) stated that this definition might be
interpreted to mean each individual heat exchanger. The
commenter (A-90-19: IV-D-73) stated that the EPA should
clarify the definition in §63.102(b) to state that a heat
exchange system can include an entire recirculation system.
Response; The EPA intended for the definition of heat
exchange system to mean the entire cooling tower system or the
entire once-through cooling system and not a single heat
exchanger. The EPA agrees with the commenter that the
definition of heat exchange system may be misinterpreted.
Therefore, the EPA has modified the definition in §63.101 of
subpart F to clarify that a heat exchange system can include
an entire recirculating system or once-through cooling system.
Comment; One commenter (A-90-19: IV-D-33J stated that
the definition of "heat exchange system" should be clarified
by excluding the parenthetical phrase "(river or pond water)",
which actually limits the definition of heat exchange system.
The commenter (A-90-19: IV-D-33) stated that the definition
should include other sources of water as well.
Response: The definition of heat exchange system has
been modified to clarify that river or pond water are only two
examples of the type of water that is used in once-through
cooling systems.
4.1.14 Alternative Methods for Determining Applicability
Comment; One commenter (A-90-19: IV-D-75) stated that
direct injection gas chromatography methods should be allowed
for determining applicability and for determining design
criteria for equipment intended to treat single-phase streams.
The commenter (A-90-19: IV-D-75) also stated that TOC methods
should be allowed for determining applicability. The
commenter (A-90-19: IV-D-75) claimed that direct injection
4-42
-------
gas chromatography and TOC methods are more readily available
and more cost effective than Method 25D or Method 305.
Response : The commenter did not specify which direct-
injection gas chromatography methods should be allowed. If
the test method measures organic HAP concentrations in the
wastewater and has been validated according to section 5.1 or
5.3 of Method 301, then the method meets the requirements of
§63.144 and is therefore allowed for determining applicability
and compliance. The EPA does not agree that TOC methods can
be allowed for determining applicability. Currently available
TOC methods measure organically bound carbon, not HAP
concentration. The commenter provided no details on how TOC
test results would be used as a surrogate parameter for VOHAP
concentration. Without additional information, the EPA cannot
further address the suggestions made by the commenter.
4.1.15 Exclusion for Laboratory Waste
Comment : One commenter (A-90-19: IV-D-54) suggested
that laboratory waste should be specifically excluded from the
definition of wastewater because such streams should not be
subject to the Group I/Group 2 determination requirements.
Response; Laboratory waste is exempt from- subparts F, G,
and H. Section 63.100(j)(l) of subpart F in the final rule
exempts all research and development facilities, regardless of
whether the facilities are located at the same plant sites as
a chemical manufacturing process unit that is subject to
subparts F, G, and H.
4.1.16 One Mg/yr Source-Wide Determination
one commenter (A-90-19: IV-D-77) suggested
that the EPA should consolidate §§63. 144 (a) and (e) or explain
why the paragraphs should remain separate.
Response; The EPA agrees that the relationship between
paragraphs §63. 144 (a) and (e) in the proposed rule is
confusing and has deleted the need to calculate an "annual
wastewater quantity" as required by proposed paragraph (a) .
The EPA has reorganized §63.144 in the final rule to clarify
the requirements an owner or operator to demonstrate whether
the HON is applicable to a wastewater stream and to determine
4-43
-------
whether a wastewater stream is a Group 1 or Group 2 wastewater
stream. In §63.144, the EPA continues to include the
compliance demonstration for the 1 Mg/yr source-wide
compliance option and also provides an additional option in
paragraph (d), which allows an owner or operator to designate
a single wastewater stream or a mixture of wastewater streams
as a Group 1 wastewater stream.
In reorganizing §63.144, the EPA has changed proposed
paragraph (e) to paragraph (c) in the final rule.
^Q^np^ent• Two commenters (A-90-23: IV-D-1) (A-90-19::
IV-D-86) agreed with having a mass flow rate de minimis value
to minimize cost and secondary impacts from control of minor
sources. Several commenters (A-90-19: IV-D-53; IV-D-62;
IV-D-63; IV-D-73; IV-D-79; IV-D-86; IV-D-92; IV-D-110)
(A-90-23: IV-D-1) expressed concern that the mass flow rate
de minimis values for wastewater streams between the HON and
Benzene Waste NESHAP are inconsistent. The commenters
(A-90-19: IV-D-62; IV-D-63; IV-D-79; IV-D-86; IV-D-110)
(A-90-23: IV-D-1) also expressed concern that the mass flow
rate de minimis value of 2 Mg/yr in the Benzene Waste NESHAP
is higher than the source-wide exemption from the control and
treatment of Group 1 wastewater streams of 1 Mg/yr in the HON,
even though many of the HAP's covered by the HON are less
toxic than benzene. Several commenters (A-90-19: IV-D-32;
IV-D-53; IV-D-63; IV-D-112) (A-90-23: IV-D-17) suggested that
adopting the 2 Mg/yr mass flow rate cutoff used in the Benzene
Waste NESHAP would minimize testing, collection, and treating
of de minimis sources of HAP's while still allowing control of
major emission sources. Another commenter (A-90-19: IV-D-85)
disagreed with the facility-wide cutoff of 1 Mg/yr. The
commenter (A-90-19: IV-D-85) claimed that a facility-wide
exemption was allowed in the Benzene Waste NESHAP because the
risk-based targets for benzene were exceeded by more than
1 megagram. The commenter (A-90-19: IV-D-85) stated that
this justification does not apply to the HON.
Response; Although the HON is not a risk-based standard,
the EPA continues to allow the 1 Mg/yr source-wide option in
4-44
-------
the final rule in §§63.138(c)(5) and (6) because compliance
with the wastewater provisions for a SOCMI facility that
generates only a low total HAP mass flow rate of table 9
compounds is very expensive. In addition, the EPA has
determined that most facilities will elect to use this option
to show compliance for wastewater streams with a low flow rate
and high concentration. The EPA recognizes that having a
source-wide de minimis value minimizes the impact of the HON
on those facilities that have HAP-containing wastewater with
low total loading. A source-wide compliance option was
originally included in both the Benzene Waste NESHAP and the
HON to address maintenance wastewater which often has a high
concentration and a low flow rate. Because the requirements
for managing maintenance wastewater in the HON have changed
from the proposed rule, the EPA considered removing the
1 Mg/yr compliance option. However, the EPA continues to
allow this option as it was proposed so that process
wastewater streams which may have a high concentration and low
flow rate are not subject to the control requirements in the
HON.
4.1.17 clarification of Requirements for Containers
Comment; Two commenters (A-90-19: IV-D-32; IV-D-73)
stated that the EPA has failed to quantify emissions from
containers and has failed to evaluate the environmental impact
and cost of the proposed container regulations. Three
commenters (A-90-19: IV-D-32; IV-D-73) (A-90-23: IV-D-20)
argued that marine vessels, tank cars, and tank trucks should
be excluded from the definition of container, because these
vessels are best regulated in another section of the
regulation (e.g., the transfer provisions) or another
regulation. One commenter (A-90-19: IV-D-92) suggested
making the requirements for containers similar to those found
in RCRA [40 CFR 262.34(c)(l) and 261.4(d), (e), and (f)].
One commenter (A-90-19: IV-D-73) suggested excluding
containers that are used less than 15 days. Several commenters
(A-90-19: IV-D-32; IV-D-34; IV-D-54; IV-D-64; IV-D-73;
IV-D-93) suggested that the EPA establish a de minimis
4^45
-------
capacity for containers, below which the containers would not
be subject to the container requirements in §63.135. The
commenters (A-90-19: IV-D-32; IV-D-34; IV-D-54; IV-D-64;
IV-D-73; IV-D-93) recommended different capacities ranging
from the size of laboratory sample bottles and shovels to
1 m3. The commenters provided several reasons for this
suggestion including: (1) small containers have little
potential to emit HAP's; (2) a de minimis capacity would
clarify the definition of "container" in §63.111 of subpart G;
and (3) a cutoff level would narrow the definition of
"container".
Response; The EPA maintains that containers holding HAP-
containing water or process fluids at SOCMI facilities are a
potentially significant source of HAP emissions, which are not
adequately regulated by existing regulations. During the
baseline analysis to estimate emissions from wastewater
operations at SOCMI sources, the EPA estimated emissions from
SOCMI sources using model plant scenarios. From this
analysis, the EPA determined that SOCMI sources as a whole
warranted emission control. The final rule specifies the
management practices that must be followed to achieve HAP-
emission reduction. The EPA's cost estimates assume that
wastewater streams are routed to a feed tank for a steam
stripper. Costs associated with containers are not relevant
to this scenario. An owner or operator may elect to manage
wastewater using containers, however, the EPA does not include
this type of management as part of the emission control
scenarios.
The EPA continues to include barges, ships, rail cars,
and tank trucks as examples of containers in the definition of
"container" in §63.111 of subpart G. This definition of
container is consistent with the definition of "container"1 in
both RCRA and the Benzene Waste NESHAP.
In response to several commenters1 request that the EPA
exempt smaller size containers from the control requirements
for containers in §63.135, the EPA reviewed the types of
smaller containers commonly used to manage Group 1 wastewaters
4-46
-------
or residuals generated from the treatment of such wastewaters,
and concluded that very small containers with a capacity less
than 0.1 m3 (26.4 gallons) should not be subject to the
container requirements in §63.135. The EPA has included this
capacity threshold in the definition of "container" in §63.111
of subpart G. The EPA has decided not to regulate very small
containers because: (1) such containers have little potential
for air emissions; (2) the monitoring, recordkeeping, and
reporting burden outweighs the environmental gain; and (3) lab
bottles and small sampling containers were not intended to be
regulated.
The EPA based this change on a review of container sizes
commercially available from vendors which indicates that the
capacities of safety cans, lab cans, disposal cans, and lab
packs range from less than 0.004 m3 (1 gallon) to 0.08 m3.
These types of small containers are used to collect small
quantities of hazardous waste in laboratories and other
ancillary operations at a SOCMI facility. The EPA
incorporated the container size limitation into the definition
of "container" in §63.111 of subpart G. In addition, the EPA
has revised the control requirements for containers with a
capacity less than 0.42 m3. As discussed in section 2.2.4 of
this BID volume, these containers are exempt from the
submerged fill requirements. These containers are also not
required to be inspected for leaks with Method 21 if DOT-
approved containers are used.
The EPA has decided not to specifically exclude
containers that are on-site for only a certain number of days.
By providing a de minimis container size and allowing less
burdensome compliance and monitoring requirements, the EPA
decided that sufficient flexibility for complying with the HON
is available without adding a specific exclusion for
containers that are on-site for only a short time. A
discussion about reduced monitoring requirements is provided
in section 6.12 of this BID volume.
Comment; One commenter (A-90-19: IV-D-64) stated that
if the EPA established de minimis cutoffs for wastewater tanks
4-47
-------
and surface impoundments, there would be no need for the
regulation of containers.
Response; The EPA clarifies that both the proposed and
final rules contain regulatory requirements for each waste
management unit including wastewater tanks, surface
impoundments, and containers. The EPA has incorporated
several changes to the wastewater tank provisions in §63.i:i3
which include the addition of tank capacity and vapor pressure
thresholds. For additional discussion of these changes, refer
to section 4.1.9 of this BID volume. The EPA maintains that
regardless of any changes made either to the wastewater tank
provisions in §63.133 or the surface impoundment provisions! in
§63.134, the container requirements in §63.135 continue to be
necessary to control HAP emissions from containers.
Containers, which by definition are portable, are not a subset
of either wastewater tanks or surface impoundments, which are
both defined as stationary waste management units. Therefore,
the container requirements in the HON are not directly
affected by any changes to either the wastewater tank or
surface impoundment requirements.
4.2 DETERMINATION OF MOST STRINGENT STANDARDS
Comment: One commenter (A-90-19: IV-D-110) stated that
the EPA's attempt at resolving conflicrs and overlaps between
the HON and other regulations in §63.103(d)(2), which requires
the owner or operator to comply with the most stringent
standards applicable to the emissions point, does not
sufficiently clarify all compliance issues. As an example,
the commenter (A-90-19: IV-D-110) suggested that facilities
that have steam strippers that meet Benzene Waste NESHAP
requirements may need to be reconfigured to meet the HON
requirements. The commenter (A-90-19: IV-D-110) recommended
that the EPA modify proposed §63.103(d)(2) to require
facilities to meet the most stringent standards applicable to
"sources" rather than "emission points." The commenter
(A-90-19: IV-D-110) stated that this approach will reduce the
burden of making a stringency determination for each emission
4-48
-------
point and will make comparisons between the HON and other
rules simpler.
Response: The EPA agrees with the commenter that several
regulatory overlap issues were unclear in the proposed rule.
The EPA has clarified many of these issues in §63.110 of
subpart G. For the final rule, the EPA continues to address
most regulatory overlap issues based on specific emission
points because comparing different regulations on a broader
scale may be misleading and could cause air emissions that are
subject to the HON to be uncontrolled. Refer to chapter 6 of
BID Volume 2D for additional discussion about regulatory
overlap and stringency decisions.
4.2.1 Overlap with the Benzene Waste NESHAP
Comment ; One commenter (A-90-19: IV-D-102) cited a
portion of the supplemental final Benzene Waste NESHAP rule
that clarifies the distinction between product and waste
(58 FR 3072, 3076-7) and suggested that the EPA use the
language to clarify the scope of the definitions in the HON.
Response ; The distinction between product and waste in
the Benzene Waste NESHAP is analogous to the definition of
wastewater in the HON. The intent of the HON is the same as
the Benzene Waste NESHAP; materials are subject to the
standards at the point they exit the production process
equipment. To the extent language in the Benzene Waste NESHAP
rule clarifies the EPA's intent, the language is equally
relevant to the HON. The EPA has not, however, added the
specific language for the Benzene Waste NESHAP.
One commenter (A-90-19: IV-D-75) recommended
that benzene-containing wastes which are subject to the
Benzene Waste NESHAP be exempt from the HON. The commenter
(A-90-19: IV-D-75) expressed concern that because some
facilities have recently installed equipment to comply with
the Benzene Waste NESHAP, conflicting requirements between the
two NESHAP 's may result in expensive rework with no
environmental benefit.
Response ; The EPA disagrees with the commenter because
the HON regulates 75 additional chemicals other than benzene.
4-49
-------
Without a compliance demonstration, the EPA cannot determine
whether a piece of equipment that was installed to comply with
the Benzene Waste NESHAP also will be in compliance with the
HON. The EPA does encourage facilities to continue using
equipment that was installed to comply with other regulations
and nothing in the HON precludes the owner or operator from
using such equipment. However, to comply with the HON, the
equipment must reduce air emissions of all organic HAP's,
including benzene, that are present in the wastewater stream
and are listed on table 9 of subpart G of the final rule.
Comment: One commenter (A-90-23: IV-D-14) suggested
deleting manholes, sumps, and lift stations from the HON
definition of individual drain system to be consistent with
the Benzene Waste NESHAP.
Response: The EPA maintains that air emissions from
manholes, sumps, and lift stations should be controlled under
the HON. Allowing such parts of a drain system to remain
uncontrolled could allow emissions to escape to the atmosphere
before the wastewater stream reaches a treatment process.
Therefore, the EPA continues to include these components in
the definition of individual drain system.
Comment; One commenter (A-90-23: IV-D-14) supported the
exemption of POTW's from the Benzene Waste NESHAP and
suggested adding this exemption to the HON. The commenter
(A-90-23: IV-D-14) claimed that POTW's lack the funds to
install the required controls. One commenter (A-90-19:
IV-D-110) opposed the requirement for POTW's to comply with
HON provisions, reasoning that the pretreatment requirements
under the CWA are adequate to control HAP emissions. One
commenter (A-90-19: IV-D-58) expressed concern that not
allowing biological treatment as RCT may increase the
potential for POTW's to decline to accept treated wastewater
due to applicability and compliance uncertainty with the HON.
Response; Neither the Benzene Waste NESHAP nor the HON
allow owners or operators to avoid control of HAP emissions by
sending wastewater offsite for treatment. Under the HON, the
POTW is not subject to the HON requirements, but the owner or
4-50
-------
operator of a SOCMI facility must ensure that Group 1
wastewater that is sent offsite to a POTW or other facility
for treatment or recycling is handled in compliance with the
HON.
4.2.2 Overlap with the Resource Conservation and Recovery Act
Comment; One commenter (A-90-19: IV-D-92) claimed that
v . ,
the definition of "waste management unit" is not consistent
with RCRA, CWA, and other air quality rules such as NSPS
subpart QQQ. The commenter (A-90-19: IV-D-92) also indicated
that "waste" was not defined in the HON. The commenter
(A-90-19: IV-D-92) urged the EPA to define "waste" and "waste
management unit" consistent with RCRA (§261.3 and 260.10,
respectively).
Response: The EPA points out that the HON does not
define the term "waste" because the HON does not apply to
waste. In §63.111 of subpart G, the HON provides definitions
for both "wastewater stream" and "waste management unit."
Although the definition of waste management unit in the HON
may differ from the definition, in other rules, the definition
in the HON explains the scope, use, and meaning of the term as
it is used in the HON.
Comment; Three commenters (A-90-19: IV-D-32; IV-D-54;
IV-D-113) stated that the EPA has correctly exempted RCRA-
permitted treatment units from the HON. However, the
commenters (A-90-19: IV-D-32; IV-D-54) claimed that some of
the provisions are contradictory and erroneously referenced.
The commenters (A-90-19: IV-D-32; IV-D-54; IV-D-113) stated
that §63.138(1) should declare that RCRA units, which are
exempt under §63.138(1), are considered to be in compliance
with §63.138(d), (b), (c), and (g). The commenters (A-90-19:
IV-D-32; IV-D-54; IV-D-113) stated that these RCRA units
should not be subject to §63.138(f) and (i). Furthermore, the
commenter (A-90-19: IV-D-32) stated that since §63.138(c)
references §63.131(d) and §63.138(f), these units should not
be subject to either §63.131(d) or §63.138(f). One commenter
(A-90-19: IV-D-54) stated that RCRA-regulated sources should
comply with (e), (h), (j), and (k).
4-51
-------
Response; The EPA agrees with the commenter that in the
proposed rule §63.138(1) of subpart G contained several
contradictory and erroneous references. In the final rule in
§63.138(m) of subpart G, the EPA has corrected these errors so
that a treatment process, wastewater stream, or residual is
considered in compliance with the requirements of §§63.138(b),
(c), and (h), as applicable and is exempt from the
requirements of §63.138(j), which requires a design analysis
or performance test, provided that the owner or operator is in
compliance with §§63.138(f), (i), (k), and (1) and documents
that the treatment process, wastewater stream, or residual is
in compliance with §63.138(m)(1) through (3). Emissions from
wastewater must be controlled until the point that the HAP's
are destroyed. Prior to this point, an owner or operator must
ensure compliance with §§63.133 through 63.137. The EPA notes
that the placement in the final rule of several of the
citations differs from the proposed rule.
Comment; One commenter (A-90-19: IV-D-92) requested
that a definition of ttempty container" similar to the
definition under RCRA in 40 CFR part 261.7 be included in the
HON.
Response: Because the HON does not apply to the disposal
of hazardous waste, the issue of whether a container is
"empty" under RCRA is not directly relevant to the HON.
Regulatory overlap with RCRA may occur when an owner or
operator of a SOCMI facility elects to send residuals placed
in containers to an off-site treatment or recycling facility.
In such cases, the owner or operator must ensure that the
residuals are managed in compliance with the HON. In cases
where the residuals also are hazardous waste, neither the HON
nor RCRA apply to any material that may be remaining in a
container that meets the "empty" criteria in 40 CFR
part 261.7. Ai~.iough the commenter does not specify why a
definition of " ~.apty container" should be added to the HON,
the EPA clarifie- that any container that has been emptied
using practices mat are commonly employed to remove materials
from that type of container (e.g., pouring, pumping) are no
4-52
-------
longer required to meet the container requirements specified
in §63.135 of subpart G of the HON.
Comment; One comraenter (A-90-19: IV-D-92) suggested
that the exemption for conditionally exempt small quantity
generators found under RCRA in 40 CFR part 261.5 be included
in the HON.
Response; The RCRA provision in 40 CFR part 261.5, which
allows hazardous waste generators who generate small
quantities of hazardous waste to be exempt from most of the
hazardous waste management provisions, was established to
relieve generators of small quantities of waste from the
financial burden associated with RCRA compliance. These
conditionally exempt small quantity generators (CESQG) are
required to manage their waste using methods that protect
human health and the environment.
Under the Act, provisions already exist which exempt
small quantity generators of HAP emissions from the
requirements in the HON. In section 112(a) of the Act,
Congress defines "major source" as a stationary source or
group of stationary sources that have the potential to emit in
aggregate, 10 tons per year or more of any HAP or 25 tons per
year or more of any combination of HAP's. Because the HON
applies only to major sources, any SOCMI plant that is not a
major source is not subject to the HON. Therefore, the EPA is
not adding a provision similar to the CESQG exemption in RCRA.
Comment; One commenter (A-90-19: IV-D-92) suggested
that listed and characteristic hazardous wastes should not be
covered by the HON, because they are regulated under RCRA
(40 CFR part 264 subparts AA and BB and 40 CFR part 265
subparts AA, BB, and CC).
Response: The HON wastewater provisions are applicable
to all HAP's listed on table 9 of subpart G in the final rule
regardless of whether some of the HAP's may also be classified
as listed or characteristic wastes under RCRA. The primary
purpose of the RCRA regulations is to require safe management
of hazardous waste from "cradle to grave." Although the RCRA
regulations do contain several provisions pertaining to the
4-53
-------
control of air emissions, the Act specifies that the EPA
promulgate regulations that control the emission of HAP's to
the air. The RCRA requirements in 40 CFR parts 264 and 265
subparts AA and BB focus on the control of emissions from
process vents and equipment leaks. The EPA also has proposed
subpart CC provisions which will require control of air
emissions from tanks and containers, but these requirements
have not been finalized. These RCRA requirements are not
sufficient to control HAP emissions from SOCMI facilities.
The EPA is trying to minimize the burden of overlapping
regulations and has provided the option for a case-by-case
determination for regulatory overlap between the HON and RCRA
in §63.110 of subpart G of the final rule.
Comment: One commenter (A-90-19: IV-D-45 and IV-F-7.7)
expressed concern that emissions from RCRA corrective action
hazardous waste surface impoundments were excluded from
control under the HON.
Response; The EPA clarifies that the SON does not apply
to corrective actions under RCRA. The RCRA regulations
designate the procedures for implementing corrective actions.
The HON applies to chemical manufacturing process units at
major sources that manufacture as the primary product one or
more of the chemicals listed in table 1 of subpart F of the
final rule and use as a reactant or manufacture as a product,
by-product, or co-product one of the organic HAP's listed in
table 2 of subpart F of the final rule. For additional
discussion on the applicability of the HON, the commenter
should refer to BID volume 2D.
4.2.3 Overlap with the Clean Water Act
rommtan-K; one commenter (A-90-19: IV-D-34) disagreed
with the provisions in §63.132(i)(2), which require that an
owner or operator be responsible for the treatment of
wastewater once it has been sent offsite to a facility that is
r.rit under the control of the owner or operator. The commenter
(A-90-19: IV-D-34) stated that this provision is virtually
impossible to comply with and does not consider the
significant investment in place to control these wastewater
4-54
-------
discharges and comply with NPDES pretreatment requirements.
Two commenters (A-90-19: IV-D-33; IV-D-110) stated that the .
HON should not impose redundant or conflicting requirements
for wastewater treatment at plants which are already subject
to CWA requirements. One commenter (A-90-19: IV-D-73)
suggested deleting §63.132(i) (1) and (2) from the wastewater
provisions, because small plants currently using POTW's cannot
ensure that POTW's will comply with §63.138(c). The commenter
(A-90-19: IV-D-73) claimed that these small plants would have
to treat their own Group 1 wastewater streams which may not be
technically feasible or cost effective.
Two commenters (A-90-19: IV-D-33; IV-D-34) suggested
that wastewater transfers to an off-site POTW as defined by
the 40 CFR 403 regulations should be exempt from §63.132(i),
since POTW's will be subject to future MACT regulations. One
commenter (A-90-19: IV-D-33) provided background information
about the development and stringency of the pretreatment
standards that must be met before wastewater is sent to a
POTW. The commenter (A-90-19: IV-D-33) also provided a copy
of an affiliated plant's permit with a local POTW to
illustrate that the permit designates specific levels of
pollutants that can be sent to the POTW.
The commenter (A-90-19: IV-D-33) agreed with the EPA
that it is a facility's responsibility to manage wastewater
onsite up to the point where it is discharged through a
connection to the POTW system. The commenter (A-90-19:
IV-D-33) contended, however, that the POTW is responsible for
"transport" of the wastewater from the plant site to the POTW
and for treatment of the industrial wastewater at the POTW.
The commenter (A-90-19: IV-D-33) suggested that wastewater
should no longer be regulated under the HON once it is
discharged to a POTW.
Response; To ensure control of HAP emissions from
wastewater, the EPA continues to require owners or operators
to certify that any Group 1 wastewater stream that is sent
offsite for treatment is controlled for air emissions in
accordance with the HON. Without this requirement, nothing
4--5S
-------
would prevent owners or operators from sending untreated
wastewater to an offsite location where HAP's could be
emitted. Even if the offsite location was a permitted POTW,
the CWA may not require a reduction in HAP emissions that is
equivalent to the HON, therefore owners or operators of SOCMI
plants shall either comply with the requirements of the HON
onsite or ensure that equivalent emission suppression and
treatment techniques are used. Refer to section 6.0 in BID
volume 2D for additional information about the overlap of the
HON with other regulations.
However, the EPA has clarified §63.132(i)(2) to allow
wastewater treatment offsite by facilities that meet the
provisions of today's regulation, or a federally-approved
alternative standard. The proposed rule would have required
that a source treating wastewater covered by this rule meet
only the applicable treatment requirements contained therein,
unintentionally excluding sources where alternative standards
in lieu of this regulation have been issued to the wastewater-
treating source. Specifically, acceptable alternative
standards include those granted under §63.102(c), where
equivalent emission reductions have been demonstrated; and
subpart D, the Early Reductions provisions, where a source has
been granted a 6-year extension from meeting the provisions of
this rule, in return for achieving reductions several years
earlier than otherwise required and accepting a mass emissions
"cap" limiting HAP emissions to 10 percent or less of what
they were prior to reductions. While the latter alternative
may or may not.be as stringent as the provisions of this rule,
the achievement of emission reductions earlier than otherwise
required more than makes up for a tighter section 112(d)
standard, and the emissions cannot exceed the alternative
standard, including emissions from additional wastewater
treated by the source. At the end of the Early Reductions
compliance extension, the source must meet today's standards.
Both types of alternative standards are subject to public
review and comment and will become title V permit conditions.
4-56
-------
comment: One commenter (A-90-19: IV-D-32) argued that
SOCHI plants that discharge wastewaters to POTW's should not
be required to notify the POTW of such discharges, and should
not have to demonstrate compliance with §§63.133 through
63.138 of the HON. Several commenters (A-90-19: IV-D-32;
IV-D-86; IV-D-73) advised that indirect discharges be
controlled under a future MACT standard for POTW's. One
commenter (A-90-19: IV-D-32) suggested that generators of
Group 1 wastewaters manage them as required by the HON up to
the point of discharge to the POTW collection system, at which
point the existing CWA regulatory programs should take
precedence.
Response; The EPA disagrees with the commenters'
suggestion to allow generators of Group 1 wastewater streams
to manage such streams only up to the point of discharge to a
POTW. The existing CWA regulatory programs require POTW's to
comply with pollutant effluent limitations, which do not
control air emissions. For this reason, the EPA continues to
require generators of Group 1 wastewater streams to ensure
that the receiving POTW is in compliance with all applicable
requirements in §63.133 through §63.139 of the HON.
Comment: One commenter (A-90-23: IV-D-2) claimed that
for materials that are easily biodegraded, the requirement to
treat prior to discharge to a biological treatment system or
to a POTW should be eliminated.
Response: The HON does not require treatment of
wastewater prior to discharge to a biological treatment unit
or a POTW. In §63.132(i) of the final rule, the HON does
require suppression of emissions in accordance with §63.133
through §63.137 during transport from the point of generation
to the waste management unit.
Comment; One commenter (A-90-19: IV-D-33) stated that
the requirements in proposed §63.138(1) should be expanded to
allow an exemption if the wastewater is treated pursuant to
OCPSF 40 CFR Part 414 regulations, which are detailed in a CWA
permit. The commenter (A-90-19: IV-D-33) stated that
proposed §63.138(1) allows wastewater that is managed in
4-57
-------
compliance with a final permit under 40 CFR Part 270
(i.e., RCRA) to be exempt from certain HON wastewater
provisions. To illustrate the similarity between RCRA permit
requirements and CWA permit requirements, the commenter
(A-90-19: IV-D-33) provided a portion of an NPDES permit,
which requires extensive monitoring recordkeeping and
reporting requirements to ensure compliance.
Response; The EPA clarifies that proposed §63.138(1)
does not allow reduced requirements under the HON when
treating wastewater using any treatment unit that is permitted
under RCRA 40 CFR part 270, but rather allows reduced
requirements only for those permitted RCRA treatment units
that already meet the requirements of the HON. The EPA has
not expanded proposed §63.138(1) [which is §63.138(m) in the
final rule] to specifically include treatment that is
performed in accordance with OCPSF requirements in 40 CFR
part 414; however, nothing in the HON precludes an owner or
operator from using such treatment to achieve compliance with
the HON. In addition, owners or operators may demonstrate
compliance with the HON through the use of the records and
reports that are required by the OCPSF rules.
4.2.4 Underground Injection Wells
Comment: Three commenters (A-90-19: IV-D-32; IV-D-S1;
IV-D-112) stated that the requirements in proposed §63.133(1)
should clearly state that all wastewater streams destined for
disposal via an underground injection well are exempt from all
requirements of proposed §63.138(e),(h),(j), and (k). The
commenters (A-90-19: IV-D-32; IV-D-61; IV-D-112) also stated
that the EPA should extend the exemption which is currently
limited to RCRA permitted wells, to include any underground
injection well permitted under 40 CFR 144-147. One commenter
(A-90-19: IV-D-61) asserted that Class I nonhazardous wells
have construction, operating, testing, monitoring, and
reporting requirements identical to RCRA wells, with the
exception that a "no migration" petition is not required for
the permitting of a nonhazardous well.
4-58
-------
Response; The EPA agrees with the commenters that the
exemption in §63.138(m) of subpart G of the final rule should
be expanded to include not only Class I hazardous waste wells
that are permitted under RCRA in 40 CFR part 270, but also
Class I nonhazardous wells, Class II, III, IV, and V wells
permitted under 40 CFR 144. The EPA has expanded this
provision to include additional categories of injection wells
reasoning that once a wastewater is pumped into the ground, no
air emissions will result. The owner or operator of a SOCMI
facility who sends wastewater for disposal via an underground
injection well must ensure that air emissions are suppressed
in the collection and conveyance system for all Group l
wastewater streams and that such a system is in compliance
with all applicable HON requirements prior to the point where
the wastewater is pumped into the ground.
4.3 GROUP I/GROUP 2 DETERMINATION
Comment; One commenter (A-90-19: IV-D-33) stated that,
.since §63.144(b) of the proposed rule allows the use of
engineering calculations to define the wastewater
characteristics at the point of generation and such
calculations may be based on samples taken at the first
air/water interface, these calculations should also be
available to determine Group I/Group 2 applicability at the
first air/water interface.
Response: The provisions in §63.144 of subpart G in the
final rule include the test methods and procedures allowed for
determining applicability of the HON wastewater provisions
(i.e., whether a wastewater stream meets the flow and
concentration criteria in the definition of "wastewater
stream") and for determining Group I/Group 2 status. In the
final rule, the title of §63.144 of subpart G has been revised
to clarify that the procedures in that section are for both
applicability and Group I/Group 2 determination.
Comment; One commenter (A-90-19: IV-D-77) stated that
the EPA should change the minimum average flow rate for
Group 1 wastewater streams at new sources from 0.02 £pm to
12 £pm for continuous flow streams, because equipment is not
4-59
-------
commercially available to treat streams with a flow less than
12 £pm.
Response; The EPA clarifies that process wastewater
streams may be treated on an individual or a combined basis.
An owner or operator may combine Group 1 process wastewater
streams for treatment. The EPA also assumed that the
wastewater streams that are subject to the wastewater
provisions would be collected in a holding tank before being
sent to the control equipment. Holding tanks can be used: to
equalize the flow of the wastewater to the control device.
Therefore, wastewater streams with low flow rates can be
collected in a holding tank and then pumped to the control
device at the appropriate flow rate when the holding tank is
full.
Comment; One commenter (A-90-19: IV-D-85) claimed that
the EPA should lower the total VOHAP average concentration for
Group I/Group 2 criteria for existing sources to 10 ppmw to
prevent emissions from wastewater streams which contain more
hazardous or more volatile pollutants than benzene. The
commenter (A-90-19: IV-D-85) also suggested that the 10 ppmw
concentration cutoff apply to all HAP's regulated by the HON,
and not just table 9 compounds. The commenter (A-90-19:
IV-D-85) stated that the Act requires regulation of all
compounds listed in section 112(b). The commenter (A-90-19:
IV-D-85) expressed concern about the total emissions from
compounds on table 1 that may be generated and not controlled
because the Group I/Group 2 criteria for existing sources is
based on the concentration of only table 9 compounds.
The commenter (A-90-19: IV-D-85) recommended that the
EPA eliminate cutoffs for existing sources, so that facilities
will not be encouraged to pipe new source wastewater streams
into existing systems. The commenter (A-90-19: IV-D-85)
stated that if the EPA continues to require a concentration
cutoff of 10 ppmw for new sources, the cutoff should apply to
aggregated VOHAP"s from all chemicals regulated under the rule
or at least apply to aggregated table 8 compounds. The
commenter (A-90-19: IV-D-85) claimed that a wastewater stream
4-60
-------
containing a total of 10 ppraw of several table 8 compounds is
as significant as a waste stream containing 10 ppmw of a
single table 8 compound.
Response: After reviewing the impact that the HON will
have on existing SOCMI sources, the EPA is not lowering the
Group I/Group 2 determination criteria for compounds listed on
table 9. The wastewater provisions in the HON focus on
controlling air emissions from wastewater collection and
treatment systems. The compounds listed on table 9 represent
those volatile organic HAP's which will be emitted from
wastewater if they are not controlled. The inorganic HAP's in
section 112 of the Act will not be emitted into the air from
wastewater handling and treatment operations. Thus, it is not
necessary or appropriate to include them as regulated
compounds. Many of the HAP's cannot exist in water because
they react to form other compounds. Other organic compounds
have inherent characteristics such that they are not emitted
to the air, and the EPA finds no purpose for requiring
emission suppression and treatment for chemicals with no
potential to emit. In the case of cooling towers, the EPA
does require the owner or operator to monitor for all HAP's
listed on table 1 of subpart F because the cooling tower acts
as an air stripper, which could generate air emissions from
HAP's that are not on table 9 of subpart G.
In response to the commenter's concern that the control
requirements for both new and existing sources should be based
on a concentration of > 10 ppmw, the EPA states that the
wastewater provisions for new sources were established based
on the best-controlled similar source at the floor. The best-
controlled similar sources were subject to the Benzene Waste
NESHAP (40 CFR part 61, subpart FF) and the Vinyl Chloride
NESHAP (40 CFR part 61, subpart F), which require control of
streams containing greater than or equal to 10 ppmw benzene or
vinyl chloride. The EPA does not know of. any source using
steam stripping to treat wastewater streams that contain a
total loading of 10 ppmw organic. Thus, requiring such a
level of control would constitute control beyond the floor.
4-61
-------
At the floor, the cost effectiveness of control for the
proposed option was $495 per megagram. The EPA estimates that
the cost effectiveness of implementing the commenter's
preferred option would be $1,690 per megagram. The EPA has
determined that the cost of implementing the commenter's
option, which is more stringent than the floor, is burdensome
and prohibitive.
rnimnent; one commenter (A-90-19: IV-D-85) recommended
that the EPA regulate all wastewater streams from new and
existing sources with a flow rate greater than 0.02 £pm
regardless of their concentration. The commenter (A-90-19:
IV-D-85) suggested that only streams falling below stringent
concentration and flow rate limits should qualify for
exemption. The commenter (A-90-19: IV-D-85) expressed
concern about emissions from wastewater streams with low flow
rates and high concentrations. The commenter (A-90-19:
IV-D-85) claimed that the EPA has provided no justification
for the Group I/Group 2 determination flow rate and
concentration criteria for existing sources other than cost-
effectiveness and that the cost-effectiveness justification
does not make sense in light of the low cost of treating
wastewater streams.
Response: The concern expressed by the commenter is
addressed by §63.132(g)(1) which states chat process
wastewater streams with either (1) a total VOHAP average
concentration of table 9 compounds greater than 10,000 ppmw
and any flow rate, or (2) a total VOHAP average concentration
is greater than or equal to 1,000 ppmw and the average flow
rate is greater than or equal to 10 £/m, are Group 1 streams.
The wastewater provisions of the HON require treatment of
Group 1 wastewater streams. The commenter did not provide
documentation supporting the statement that the cost-
effectiveness approach used by the EPA in determining the
applicability criteria for the wastewater provisions was
unfounded. The commenter also did not provide any information
on the cost of wastewater treatment. The details of the EPA's
cost analysis were described in the proposed BID and revisions
4^-62
-------
to the analysis are documented in memoranda in the docket.
The EPA is required by §112(d)(2) of the Act to consider cost
in establishing MACT standards.
4.3.1 Testing at Peak Levels for Applicability Determination
Comment; Several commenters (A-90-19: IV-D-32; IV-D-77)
(A-90-23: IV-D-20) stated that the EPA should change the
language defining "annual average flow rate" in §63.144(e)(1)
to clarify that the maximum annual average production capacity
should be used to calculate the annual average flow rate. The
commenter (A-90-19: IV-D-32) stated that the EPA should
clarify in §63.144(e)(2) that the selection of the "highest
average flow rate" as referred to in §63.144(e)(2) is the same
as the annual average flow rate. One commenter (A-90-23:
IV-D-20) stated that §63.144(e)(3) should allow a source to
use process knowledge to estimate the flow rate at the point
of generation, which would be consistent with the option
allowed in §63.144(e)(1).
Response; The language defining "annual average flow
rate" in proposed §63.144(e)(1) has been revised in §63.144(c)
of the final rule to clarify that the maximum annual average
production capacity should be used in estimating the annual
average wastewater flow rate or the total annual wastewater
volume. The term "average flow rate" has been revised to read
"annual average flow rate" to further clarify the intent of
the procedures in §63.144(c) in the final rule. Owners or
operators who desire to use process knowledge to estimate the
annual average flow rate can use the provisions of §63.144(c).
4.3.2 Determining VOHAP Concentration
gQTi^ngnt; one commenter (A-90-19: IV-D-64) stated that
in §63.144(b)(1) engineering judgment could be a satisfactory
basis for concluding that the VOHAP concentration in a stream
will be minimal. For example, the commenter (A-90-19:
IV-D-64) stated that if the pressure of steam in a heat
exchanger will always be higher than the pressure of the
process fluid being heated or cooled, the probability of a HAP
leak into the steam and eventualIJr into the steam condensate
system will be extremely low. The commenter (A-90-19:
4^63
-------
IV-D-64) requested that such an example be added to the final
rule.
Response: The EPA agrees with the commenter (A-90-19:
IV-D-64) that engineering judgement is an allowed method for
determining the VOHAP concentration in a wastewater stream for
applicability and Group I/Group 2 determination. The
provisions in §63.144(b)(3) of the final rule specifically
allow knowledge of the wastewater for determining the VOHAP
concentration, provided the owner or operator has proper
information to document the engineering judgement.
Section 63.104 of subpart F exempts heat exchange systems
operating at a pressure at least 35 kilopascals greater than
the maximum pressure on the process side from the heat
exchange system requirements. Heat exchangers using steam as
the heating fluid are regulated by subpart G. However,
§63.144(b) does not cite specific examples of how process
knowledge can be used to determine the VOHAP concentration in
a wastewater stream. The owner or operator would only need to
provide documentation of the pressure of the steam and the
pressure of the process fluid to demonstrate that the VOHAP
concentration in the steam condensate is negligible.
Comment; One commenter (A-90-23: IV-D-20) suggested
that §63.144(b) should be modified to allow an owner or
operator to determine the average VOHAP concentration based on
"process wastewater," rather than "wastewater stream," since
both Group 1 and Group 2 wastewater streams include only
process wastewater.
Response; Group I/Group 2 determinations are made for
wastewater that is discharged from a chemical manufacturing
process unit to an individual drain system. The EPA has added
a separate definition for "process wastewater" to the final
rule to further clarify which wastewaters are subject to
Group I/Group 2 determinations.
Comment; -Two commenters (A-90-19: IV-G-10) (A-90-23:
IV-G-5) were unsure if the total volatile portion of the
organic HAP meant the total VOHAP average concentration of
only table 9 compounds. The commenters (A-90-19: IV-G-10)
4-64
-------
(A-90-23: IV-G-5) claimed that there are compounds that do
not appear on table 9 of §63.131 or table 1 of §63.104 that
are volatile and that are included in the list of SOCMI
production process chemicals regulated under subparts F and G.
The commenter (A-90-23: IV-G-5) gave acetone as an example.
Response; The EPA clarifies that only those HAP's listed
on table 8 and table 9 of subpart G are included in the
determination of total VOHAP average concentration for the
HON. Table 9 contains only those volatile organic HAP's for
which the EPA has identified a potential to be emitted from
wastewater. Table 8 is a subset of table 9 and contains
compounds which have a volatility equal to or greater than the
volatility of benzene, as defined by the Henry's law constant.
Acetone is not a HAP and is therefore not regulated under
the HON. However, acetone is listed as a SOCMI production
process subject to the HON in table 1 of subpart F because the
acetone production process uses HAP compounds as raw materials
which may be released to the environment.
Comment; One commenter (A-90-19: IV-D-73) supported the
provisions which allow alternatives to direct measurement by
EPA methods when determining VOHAP concentration at the point
of generation.
Response: Section 63.144(b) of the final rule presents
the different ways that an owner or operator can determine the
VOHAP concentration for the point of generation. The
provisions allow knowledge of the wastewater, bench-scale or
pilot-scale test data, and direct measurement by EPA methods.
The EPA has allowed alternatives to direct measurement to
provide flexibility to owners or operators of a facility where
other methods are sufficient.
4.3.3 Sampling at Point of Generation
Comment; One commenter (A-90-19: IV-D-46) advised that
point of generation sampling may not be possible for tanks
that are inaccessible, pressurized, constructed of special
materials, or designed to flow to a common drain header.
Response: The EPA clarifies that sampling at the point
of generation is not required. Owners or operators can use
4-65
-------
knowledge of the wastewater or bench-scale or pilot-scale test
data; or may designate a single wastewater stream or a mixture
of wastewater streams as a Group 1 wastewater stream without
sampling at the point of generation.
Comment: One commenter (A-90-19: IV-D-46) suggested
that point of generation sampling decisions be made by
facility operators based on the product and system
characteristics.
Response; In both the proposed and final rule, the EPA
allows an owner or operator to apply knowledge to determine
wastewater characteristics at the point of generation. In the
final rule, both the total VOHAP average concentration or the
average VOHAP concentration of each individual organic HAP and
the annual average flow rate may be determined downstream of
the point of generation at a location when two or more
wastewater streams have been mixed and prior to treatment.,
However, the owner or operator must make corrections for any
changes in VOHAP concentration and flow rate due to the
mixture of wastewater streams. The EPA clarifies that the
point of generation is a fixed point, as defined in the final
rule.
Comment: One commenter (A-90-19: IV-D-89) reported that
there were no sampling techniques specified for wastewater.
The commenter (A-90-19: IV-D-89) stated that only the water
phase of a sample should be analyzed because the hydrocarbon
phase is usually recovered and recycled to the process. The
commenter (A-90-19: IV-D-89) claimed that the hydrocarbon
phase should only be analyzed if it enters an uncontrolled
waste treatment unit.
Response; The EPA clarifies that Method 25D provides
sampling methods and procedures for wastewater. The EPA
disagrees that only the water phase of a sample should be
analyzed because the total VOHAP concentration defines whether
a wastewater stream is Group 1 or Group 2. Without analyzing
the entire sample, HAP emissions would occur without control
even if the oil was eventually recycled. Total VOHAP is
determined at the point of generation or downstream of the
4-66
-------
point of generation if corrections are made for any HAP losses
that occur after the point of generation.
4-67
-------
5.0 COMPLIANCE OPTIONS
5.1 TARGET REMOVAL EFFICIENCIES
Comment; One conunenter (A-90-19: IV-D-56) suggested
that removal efficiency variances be allowed because removal
efficiencies are dependent on the matrix of compounds in the
wastewater and many of the strippability group A compounds
cannot achieve the 99 percent target removal efficiency. The
commenter (A-90-19: IV-D-56) suggested that the variance
procedure be similar to the Clean Water Act Fundamentally
Different Factor (FDF) variance. The commenter (A-90-19:
IV-D-56) attached two reports entitled, "Development of Unit-
Specific Predictive Emissions Equations for Chlorinated
Hydrocarbons", and "Using Unit-Specific Correlations to
Improve Equipment Emissions Inventory Estimates" which the
commenter stated include data on variable removal
efficiencies. The commenter also attached the executive
summary of a report entitled, "Supplemental FDF Information."
Response; The two reports attached by the commenter
("Development of Unit-Specific Predictive Emissions Equations
for Chlorinated Hydrocarbons," and "Using Unit-Specific
Correlations to Improve Equipment Emissions Inventory
Estimates") discuss the development of unit-specific equipment
leak correlations and contain no data on wastewater or steam
stripper removal efficiencies. Therefore, these reports do
not support the allowance of a removal efficiency variance.
The executive summary .submitted by the commenter contains
a brief discussion regarding three steam strippers operated by
the facility. The focus of the executive summary is that one
of the steam strippers is unable to meet the chloroform
5-1
-------
discharge limits required by 40 CFR part 414 subpart J, which
regulates the direct discharge of wastewaters from the OCPSF
industry. The report does not present any data that
demonstrate that the wastewater treatment requirements under
§63.138 of the HON cannot be achieved by any of the three
steam strippers. Specifically, the executive summary does not
discuss the actual chloroform removal efficiencies of the
steam strippers. Therefore, the executive summary does not
support the allowance of a removal efficiency variance.
Comment; Two commenters (A-90-19: IV-F-1.2 and IV-F-4)
(A-90-23: IV-D-4) stated that the proposed HON provided no
evidence that the design steam stripper can achieve the target
removal efficiency for strippability groups A, B, and C in
table 9 of the proposed HON. Two commenters (A-90-19:
IV-F-1.2 and IV-F-4; IV-D-58) stated that the EPA had selected
a target removal efficiency for strippability groups B and C
based on the most volatile compounds in each group, thereby
overestimating strippability for most compounds. One
commenter (A-90-19: IV-D-32) stated that the target removal
efficiencies for group B and group C compounds should be
consistent with the true strippability for each compound using
steam stripping. The commenter (A-90-19: IV-D-32) stated
that the chemical-specific strippability data found in
table 33 of the proposed rule was inconsistent with the target
removal efficiencies in table 9. The commenter (A-90-19:
IV-D-32) contended that a full-scale steam stripper cannot
achieve the strippabilities required in the proposed
regulation for some group B and group C compounds. Several
commenters (A-90-19: IV-D-73; IV-D-79; IV-F-1.2 and IV-F-4)
(A-90-23: IV-D-4) concluded that compounds at the lower end
of the range of volatility in these two groups cannot be
removed at the efficiency required even using the design steam
stripper.
As an alternative, several commenters (A-90-19: IV-F-1.2
and IV-F-4; IV-D-32) (A-90-23: IV-D-20) recommended
determining target removal efficiencies for each compound
individually or removing the low volatility compounds from the
5-2
-------
list. Several commenters (A-90-19: IV-D-55; IV-D-58;
IV-D-77; IV-D-79) (A-90-23: IV-D-4; IV-D-18) expressed
concern that the Fr factors for several HAP's (e.g., methanol,
2,4-dinitrotoluene, and MTBE) in table 33 are unachievable
using the RCT and requested that the EPA provide public
documentation that the given removal efficiencies are
achievable. One commenter (A-90-23: IV-D-4) stated that the
use of a simple Henry's law model is not adequate estimating
target removal efficiencies for all HAP's.
One commenter (A-90-19: IV-D-32) suggested that the EPA
use ASPEN simulations for at least 25 HAP's that fully
represent the range of volatilities in strippability groups A,
B, and C in order to establish the points for a more accurate
regression analysis. The commenter (A-90-19: IV-D-32)
suggested that a logit transformation of the strippability and
the logarithm of the Henry's law coefficients for each of the
simulated chemicals could be used to develop a regression
equation for estimating Fr values as a function of Henry's law
constants. The commenter (A-90-19: IV-D-32) provided data in
appendix K of the comment letter recommending that the EPA use
the Kremser equation to individually simulate the
strippability for each regulated HAP.
One commenter (A-90-19: IV-D-73) stated that based on
simulation models, many steam strippers that are different
from the design steam stripper can achieve the same or greater
removal efficiencies than the design steam stripper. The
commenter (A-90-19: IV-D-73) claimed that the owners of these
steam strippers are penalized because they would be subject to
a performance test showing HAP removal efficiency.
Response: The EPA clarifies that the basis for the HAP
target removal efficiencies achieved by the design steam
stripper was documented at proposal, although this basis was
not discussed in the proposal BID. For the proposed rule,
steam stripper performance was documented in a memorandum
titled "Approach for Estimating Emission Reductions of
Hazardous Air Pollutants from Wastewater Streams in the HON,"
5-3
-------
(Docket No. A-90-23, Item ll-B-5). For the final rule, steam
stripper performance is documented in a memorandum titled
"Estimating steam stripper Performance and Size." The EPA
further clarifies that the Kremser equation was used to
estimate both group target removal efficiencies for the
proposed HON and individual HAP compound target removal
efficiencies for the final regulation. At proposal, ASPEN was
used to predict the removal efficiency (Fr) for five example
organic compounds with Henry's law constants that spanned the
possible range of Henry's law constants. The Fr values of the
five compounds were plotted versus their Henry's law
constants, and algorithms were used to develop the Fr values
for the remaining HAP's ("Approach for Estimating Emission
Reductions of Hazardous Air Pollutants from Wastewater Streams
in the HON," Docket No. A-90-19: Item II-B-5). For the final
rule, each Fr value was individually calculated using the
Kremser equation. The EPA is not familiar with the term
"logit" transformation used by the commenter, and therefore,
cannot respond to the suggestion for its use.
The EPA agrees that the target removal efficiencies for
the HAP's regulated under the wastewater provisions of the HON
rule should be consistent with the Fr value for each
individual compound. Therefore, the final rule has been
revised such that the treatment provisions, which were based
on target removal efficiency groups in the proposed rule, are
based on the individual compound target removal efficiencies
in the final rule.
The basis is not clear for the statement made by one
commenter that the use of a model based on Henry's law is not
adequate for estimating HAP target removal efficiencies. The
commenter cites a comparison between the methanol removal
efficiency in the proposed HON (70 percent), the methanol
removal efficiency predicted in an ASPEN simulation
(13 percent) and the methanol removal efficiency measured in a
1992 EPA contract report (47 percent) [Treatment of
Pharmaceutical Waste by Steam Stripping and Air Stripping, EPA
5-4
-------
Contract No. 68-CO-OOO, Risk Reduction Engineering Laboratory,
Cincinnati, September 1992]. The revised steam stripper
performance calculations performed by the EPA for the final
HON indicate that the methanol target removal efficiency of
the design steam stripper is 31 percent. This is less than
the average 47 percent measured in actual stripping, as cited
by the commenter. The ASPEN simulation cited by the commenter
assumed the wastewater feed enters the steam stripper at
35 °C. The EPA analysis assumes the wastewater feed enters
the steam stripper at 95 °C. This accounts for the difference
between the EPA estimate of 31 percent and the ASPEN estimate
cited by the commenter (13 percent). The EPA concludes that
the Kremser equation, which uses the Henry's law constants, is
adequate for estimating target removal efficiencies for
individual compounds and that the target removal efficiencies
in the final HON are achievable by the design steam stripper.
The EPA clarifies that owners or operators using steam
strippers which differ from the design steam stripper as
presented in §63.138(g) to comply with §63.138(b)(1), (c)(1),
or (d) can either demonstrate compliance based on design
evaluation that meets the requirements of §63.138(j)(1) or
performance tests that meet the requirements specified in
§63.145.
Comment; Several commenters (A-90-19: IV-D-75; IV-D-58)
(A-90-23: IV-D-4) claimed that selecting a single target
removal efficiency for a group of compounds makes it
impossible to demonstrate equivalency using an alternate
control option. One commenter (A-90-19: IV-D-110) stated
that if the EPA relies on incorrect strippability estimates
and target removal efficiencies, facilities that are
attempting to demonstrate equivalency of alternate control
technologies will be comparing their performance to levels of
performance that cannot be reached by the proposed RCT. Two
commenters (A-90-19: IV-D-32; IV-D-77) expressed concern that
because facilities will be required to use the strippabilities
in table 33 of the proposed rule to demonstrate equivalency of
alternate control technologies to the designated RCT, the
5-5
-------
facilities will actually be comparing the performance of
alternate systems to a level of performance that cannot be
achieved by the EPA's design steam stripper. The commenters
(A-90-19: IV-D-32; IV-D-77) contended that if the design
steam stripper cannot achieve the strippability values in
table 33 of the proposed rule, then alternate control
technologies should not be compared to these values.
Responset The EPA has revised the Fr values based on
revised Henry's law constants at 100 °C and a steam-to-feed
ratio of 0.04 kg of steam per liter of wastewater. The
revised Fr values were estimated using the Kremser equation
for each of the HAP's regulated under the wastewater
provisions of the final rule. The analyses conducted by the
EPA demonstrated that the design steam stripper can achieve
the target removal efficiencies. For the final rule, steam
stripper performance is documented in a memorandum entitled
"Estimating Steam Stripper Performance and Size."
Comment: Several commenters (A-90-19: IV-D-104;
IV-D-108) (A-90-23: IV-D-1) stated that the target removal
efficiencies in table 9 are inappropriately defined, because
the strippability requirements do not account for multi-
component streams or variable inlet concentration of HAP's,
One commenter (A-90-23: IV-D-1) contended that owners will
experience difficulty in achieving the current percent
reductions for wastewater streams with low concentrations of
HAP's. As an example, two commenters (A-90-23: IV-D-1;
IV-D-4) cited data on the strippability of methanol from a
study entitled, "Treatment of Pharmaceutical Wastewater by
Steam Stripping and Air Stripping," published in a
September 1992 report by Radian Corporation subcontracted to
Battelle Memorial Laboratories under U.S. EPA Contract
No. 68-CO-0003. Another commenter (A-90-19: IV-D-32) stated
that the results of this study demonstrate that oxygenated
organic compounds are poorly steam stripped. One commenter
(A-90-23: IV-D-1) stated that the study also presents
variable strippabilities for other organic HAP's. Several
5-6
-------
conunenters (A-90-19: IV-D-104; IV-D-108) (A-90-23: IV-D-1)
claimed that the strippability requirements in proposed
table 9 of §63.131 are overestimated and unobtainable in an
operating facility with multi-component streams, because such
streams exhibit variable strippabilities.
Response; The report referenced by the commenter does
not substantiate the commenter's claim that the strippability
requirements are overestimated, and that multicomponent
streams will exhibit variable strippabilities.
The purpose of the report was to obtain sufficient data
to establish numerical effluent limitations for the
pharmaceutical industry for specific volatile organic
compounds based on steam stripping, and for ammonia based on
air stripping. In the report, three different wastewaters
were studied. For each wastewater, the organic compound
removal efficiencies of the steam stripper were determined at
different steam-to-feed ratios. The report presents steam
stripper removal efficiencies for five table 9 HAP's:
chloroform, methanol, MIBK, methylene chloride, and toluene.
Other table 9 HAP's, if present in the wastewater, were
reported as below the detection limit in the steam stripper
influent and effluent, making it impossible to estimate the
removal efficiency. The study results generally agree with
EPA's revised estimates for the removal efficiencies for
chloroform, MIBK, methylene chloride, and toluene.
The report indicates that the removal efficiency of
methanol does show some variation. The report states that
"the difficulty in accurately measuring the methanol
concentrations in the feed and effluent streams is a likely
contributing factor to the poor comparison between the
computer simulation modeling results and experimental data."
Difficulties with accurately measuring the methanol
concentrations in the wastewater is indicated by data which
show decreases in the methanol removal efficiency of the steam
stripper as the steam-to-feed ratio increases. In reality,
the methanol removal efficiency will increase as the steam-to-
feed ratio increases. The report itself does not consider
5-7
-------
multicomponent interactions as a possible contributing factor
to the variation in methanol removal variation. The EPA also
notes that it is difficult to evaluate matrix effects and an
infinite number of combinations exist which would require
evaluation.
Comment; A commenter (A-90-23: IV-D-1) from a
pharmaceutical company suggested that the outlet concentration
resulting from treatment of wastewater should be used as a
compliance option in lieu of target removal efficiencies. The
commenter (A-90-23: IV-D-1) suggested that the outlet
concentration of organic HAP's be established based on the
inlet concentration instead of set at a single level with no
regard to inlet concentration. The commenter
(A-90-23: IV-D-1) claimed that facilities with wastewater
streams having low concentrations would not be able to obtain
the required percent reduction and proposed that the percent
reduction requirement be replaced with a maximum steam
stripper outlet concentration for each HAP.
Response: The commenter (A-90-19: IV-D-1) did not
define what was meant by "low" concentration and did not
provide any information showing that "low" concentration
streams cannot achieve the required percent reduction. The
definition of wastewater includes a threshold VOHAP
concentration so that "low" concentration streams are not
subject to the wastewater provisions of the HON. In order to
be subject to the wastewater provisions of the HON, a
wastewater stream must have a VOHAP concentration of at least
5 ppmw. Furthermore, there are VOHAP concentration thresholds
associated with the control requirements for wastewater. For
existing sources, the threshold VOHAP concentration for
control is 1,000 ppmw and for new sources, the threshold VOHAP
concentration for control is 10 ppmw. The EPA has calculated
new compound-specific strippabilities which represent the
target removal efficiencies for each compound. The EPA has
determined, using the Kremser equation, that these removal
efficiencies can be achieved in the design steam stripper for
all Group 1 wastewater streams, regardless of their inlet
5-8
-------
concentration to the steam stripper. There are also other
options that can be used to comply with the wastewater
provisions in lieu of meeting the required percent reduction.
For example, new sources can reduce the average VOHAP
concentration of each HAP listed in table 8 of subpart G in
the wastewater stream to below 10 ppmw.
Comment; One commenter (A-90-19: IV-D-73) agreed with
the provisions in §63.138(c)(6)(i)(B) which allow
determination of the VOHAP mass flow rate at the outlet of a
treatment device that treats to less than reference control
levels for purposes of determining the source-wide VOHAP mass
flow rate for table 9 compounds. The commenter (A-90-19:
IV-D-73) also supports the provision in §63.138(c)(6)(i)(C)
which excludes Group 1 wastewater streams that are treated to
reference control levels from the total source VOHAP mass flow
loading determination claiming that both of these provisions
provide incentive for pollution prevention.
Response: The EPA appreciates this support and agrees
with the commenter that these provisions provide incentives
for pollution prevention. However, the EPA clarifies that any
Group 1 wastewater that is treated to comply with
§63.138(c)(6) must comply with all applicable requirements in
§63.133 through §63.139 until the treated wastewater is
discharged.
qommen,^; Several commenters (A-90-19: IV-D-32; IV-D-78)
(A-90-23: IV-D-20) contended that the EPA should establish in
§63.139(b)(4) a concentration-based cutoff of 20 ppmw for
noncombustion control devices as allowed in §63.139(b)(1)(ii)
for combustion devices.
Response; The EPA continues to require non-combustion
control devices (i.e., recovery devices), except flares, to
reduce emissions by 95 percent. An outlet concentration of
20 ppmv is allowed for combustion devices as an alternative to
achieving a 95-percent removal efficiency, because 20 ppmv is
the lower concentration limit for Which combustion devices can
achieve their removal efficiencies. Recovery devices do not
have limits on their removal efficiencies at concentrations of
5-9
-------
20 ppmv. Flares do not have percent reduction or
concentration requirements, but must meet the requirements in
40 CFR 63.11(b).
5.2 MAINTENANCE WASTEWATER
eoTmnept; one commenter (A-90-19: IV-D-32) stated that
the proposed start-up, shutdown, and malfunction plan
requirements for routine maintenance wastevaters will be part
of a plant's air permit, which will ensure proper management
of the wastewater. Two commenters (A-90-19: IV-D-32)
(A-90-23: IV-D-20) requested that the EPA omit the last
sentence in §63.102(b)(1)(ii) and therefore allow each plant
more flexibility in selecting a site-specific wastewater
management option for control of routine maintenance
wastewaters.
Response; The proposed requirements for maintenance-
turnaround wastewater are addressed in the facility's start-
up, shutdown, and malfunction plan. This plan must ensure
that maintenance-turnaround wastewaters are properly managed
and that organic HAP emissions released from these wastewaters
are controlled. If a facility's air permit specifies the
proper management of maintenance-turnaround wastewater, then
the permit can be submitted as part of the facility's start-
up, shutdown, and malfunction plan.
The EPA is changing the requirements for routine
maintenance wastewater as proposed in §63.102(b). Routine
maintenance wastewater will not be subject to the proposed
requirements in §63.102(b)(1)(ii) but will be subject to the
same requirements as listed for maintenance-turnaround
wastewater in §63.102(b)(1)(i) of the proposed rule. In the
final rule, provisions for maintenance wastewater are in
§63.105.
rnmpgn't•- One commenter (A-90-23: IV-D-17) was unsure
whether or not the control of routine maintenance emissions
requires that all process equipment be drained and purged of
all process fluids before opening. The commenter (A-90-23:
IV-D-17) claimed that the fugitive emissions caused by purging
a vessel before maintenance are greater than the emissions
5-10
-------
from the vessel during maintenance if it was not purged. The
commenter (A-90-23: IV-D-17) claimed that the added equipment
used to purge a vessel is in continuous service, whereas
maintenance procedures are periodic. The commenter (A-90-23:
IV-D-17) asserted that the EPA has not accurately assessed the
emissions from purging and provided data which compares
emissions from maintenance activities and equipment leaks from
purging, one commenter (A-90-23: IV-D-17) stated that data
provided in their comment letter indicate that the cost of
controlling emissions from maintenance wastewater under the
Benzene Waste NESHAP is $140,000/ton. The commenter (A-90-23:
IV-D-17) contended that the fugitive emissions from the purge
and block valves, which were added as part of the emission
control equipment, actually exceeded the emissions that
required control under the Benzene Waste NESHAP. Therefore,
the commenter (A-90-23: IV-D-17) concluded that no net
emission reductions were achieved.
Response: It is assumed that the commenter (A-90-23:
IV-D-17) may have misinterpreted the provisions in
§63.102(b)(1)(ii) of the proposed rule which require that an
owner or operator provide a description of the procedures used
when emptying and purging equipment during periods not
associated with a process unit shutdown. The provisions do
not require that all process equipment be drained and purged
of process fluids before opening. Rather, the provisions
require that wastewater generated during emptying and purging
be properly managed if a piece of process equipment needs to
be drained and purged in order for maintenance activities to
be performed. The provisions do not require the installation
of any additional equipment (i.e., purge and block valves) for
purging equipment during routine maintenance procedures.
The requirements in §63.102(b)(l)(ii) in the proposed
regulation for routine maintenance wastewaters have been
revised. Routine maintenance wastevaters no longer have to be
collected and recycled, destroyed, or collected and managed in
a controlled drain system. They are now subject to the
proposed requirements for maintenance-turnaround wastewaters.
5-11
-------
These requirements are not control requirements but are
general "good housekeeping" requirements.
The commenter's (A-90-23: IV-D-17) estimate of emissions
from purge and block valves assumes that these valves are in
operation continuously. The EPA has determined that the
commenter (A-90-23: IV-D-17) overestimated the emissions from
purge and block valves, because these valves are only in
service during maintenance activities. Purge and block valves
are used to drain all process fluids before opening equipment
in order to perform maintenance activities.
5.3 MANAGEMENT OF RESIDUALS
Comment: One commenter (A-90-23: IV-D-21) suggested
that wastewater residuals should not be regulated by the HON,
but rather in a separate MACT standard. The commenter
(A-90-23: IV-D-21) questioned why wastewater residuals are
regulated by the HON while process residuals are not.
Response; The EPA is not regulating wastewater residuals
under a separate MACT standard as suggested by the commenter
because no benefit would be gained by separating the organi.cs
from the wastewater if the organic residuals are not treated.
The Act requires that MACT standards be developed for source
categories. The HON is the MACT standard for the SOCMI source
category. The SOCMI source comprises several emission points
including wastewater collection and treatment systems.
Because wastewater residuals may be generated as a result of
compliance with the wastewater provisions of the HON, such
residuals are regulated by §63.138(h) as part of HON.
The EPA assumes that the term "process residuals" means
wastes that are generated at a SOCMI facility but not as a
result of complying with the HON. Residuals that are not
generated as a result of implementing the HON are not
regulated by the HON. Other wastes generated at SOCMI
facilities may be addressed by regulations such as RCRA.
Comment; One commenter (A-90-19: IV-D-73) suggested
changing the requirement to destroy 99 percent of the total
HAP mass in residuals to 99 percent of the total VOHAP mass.
5-12
-------
Response ; The EPA has not made the change suggested by
the commenter in the final regulation because it would involve
an incorrect use of the term "VOHAP". However, the EPA has
clarified that the requirement to treat the residual to
destroy the total combined HAP mass flow rate by 99 percent or
more is determined by the procedures specified in §63. 145 (c)
or (d) . The requirement applies only to those HAP's listed in
tables 8 and 9 of the final regulation. Refer to section 2.0
of this BID volume for discussion on the correct use of the
term "VOHAP concentration."
One commenter (A-90-19: IV-D-64) expressed
concern that the requirements in §63.138 for 99 percent
reduction in a treatment process and for 99 percent
destruction of each residual could be extended to conventional
control technologies, such as flares, incinerators, process
heaters, and condensers. The commenter (A-90-19: IV-D-64)
stated that those devices do not always achieve 99 percent
removal or destruction, yet would be considered satisfactory
control devices for closed-vent systems as regulated in
§63.139. The commenter (A-90-19: IV-D-64) requested that the
EPA either clarify the definitions of "residual" and
"treatment process" so that such control devices could not be
considered treatment processes or lower the 99 percent
destruction requirement to 95 percent. The commenter
(A-90-19: IV-D-64) stated that part of the problem is the use
of a waste incinerator example in the definition of "treatment
process." The commenter (A-90-19: IV-D-64) stated that the
EPA seems to indicate that treatment processes apply to
liquids and control technologies apply to gases. The
commenter (A-90-19: IV-D-64) requested that the EPA
distinguish the differences.
Response; In the final rule, the EPA continues to allow
(1) several options for controlling HAP emissions from
wastewater in §§63.138(b), (c) , (d) , and (e) of subpart G,
which includes the requirement for a 99 percent reduction of
HAP emissions; and (2) three options for controlling HAP
emissions from residuals in §63. 138 (h), one of which requires
5-13
-------
the treatment of residuals to destroy the total combined HAP
mass flow rate by at least 99 percent. The commenter seems
confused about the relationship between the control
requirements for treatment processes, which are by definition
used to comply with §63.138, and control devices which are
used to comply with §63.139. Combustion devices which are
used in conjunction with waste management units must achieve
either a 95 percent reduction in the total organic compound
emissions or an outlet total organic concentration of 20 ppmv.
Any other control device used in conjunction with waste
management units must achieve a 95-percent reduction in the
total organic compound emissions. The EPA clarifies that both
treatment processes used to comply with §63.138 and control
devices used to comply with §63.139 may be used to treat both
liquids and gases. The HON does not specify that certain
technologies must be used to treat specific materials.
Comment; One commenter (A-90-19: IV-D-54) contended
that §63.138(e)(1), which states that recycled wastewater
streams or residuals "shall not be exposed to the atmosphere,"
is overly broad and unnecessary. The commenter (A-90-19:
IV-D-54) stated that this requirement is unnecessary because
§63.138(e)(2) requires these streams to comply Vith §63.133
through §63.137. The commenter (A-90-19: IV-D-54) requested
that the intent of §63.138(e)(1) either be clarified or the
paragraph be deleted.
Response; The provisions in §63.138(f)(1) of the final
rule [which was §63.138(e)(1) in the proposed rule] are
intentionally broad because the manner in which recycling of
wastewater streams or residuals is conducted may vary
considerably from facility to facility and may not involve
management in waste management units, including those
regulated under §63.133 through §63.137. The inclusion of the
phrase "shall not be exposed t? the atmosphere" is intended, to
provide for suppression and control of management units other
than those regulated by §63.133 through §63.137.
C9Ttiffl«j»TTt• one commenter (A-90-19: IV-D-78) recommended
that facilities be allowed to treat the overheads from a
5-14
-------
design steam stripper as a process vent and apply the
associated RCT (or otherwise achieve 98 percent HAP reduction)
instead of being required to install a condenser.
Response; The requirement to install a condenser as part
of the design steam stripper provisions in §63.138(g) has been
removed from the final regulation. The EPA clarifies that
each treatment process or waste management unit that receives,
manages, or treats a Group 1 wastewater stream or residual
removed from a Group 1 wastewater stream must comply with
§63.138(1)(3)(i) through (i)(3)(iv) of the final rule. The
emissions from the primary condenser on the steam stripper
overheads must be routed to a control device designed and
operated in accordance with §63.139.
5.4 AVAILABILITY OF SERVICE FIRMS
Comment: One commenter (A-90-23: IV-D-28) expressed
concern regarding the lack of discussion in the HON of the
services that commercial firms can provide to help SOCMI
facilities comply with the HON. The commenter (A-90-23:
IV-D-28) claimed that commercial firms can collect and treat
wastewaters on or offsite and treat newly generated and
historical wastes. The commenter (A-90-19: IV-D-28) was
unsure of the control, monitoring, recordkeepirig, and
reporting requirements that would apply to onsite or offsite
commercial firms which manage SOCMI wastewaters. The
commenter (A-90-23: IV-D-28) suggested clarifying the
requirements for commercial firms in the preamble to prevent
rule violations.
Response; The EPA clarifies that the HON does not -apply
to service firms. If a SOCMI plant owner or operator elects
to contract with a commercial firm, the SOCMI plant owner or
operator is still responsible for ensuring that any Group 1
wastewater sent offsite for treatment is managed in accordance
with the HON.
5.5 BIOLOGICAL TREATMENT
gQTOjH«?nt; One commenter (A-90-19: IV-D-32) stated that
the EPA's mathematical formulation for biological degradation
in the WATER? model and similar calculations in other
5-15
-------
acceptable simulation models correctly assume that sorption of
HAP's to the biological solids is a negligible removal
pathway. The commenter (A-90-19: IV-D-32) provided chemical-
specific sorption data substantiating this statement.
However, another commenter (A-90-19: IV-D-85) stated
that the EPA should account for VOC sorption onto sludge,
which may cause air emissions during disposal of the sludge.
Response ; The EPA agrees that sorption of VOC's onto
sludge generated by biological treatment units is negligible.
Therefore, VOC emissions resulting from the disposal of sludge
generated by biological treatment units is insignificant.
However, VOC contained in sludges generated by other sources,
such as API separator's, where no biodegradation occurs, may
be significant. Such sludges are defined as residuals in the
rule, and must be treated to destroy 99 percent of the organic
HAP content in the sludge. Therefore, emissions of HAP's from
sludges are subject to the control requirements of the HON.
5.6 PROCESS UNIT ALTERNATIVE
one commenter (A-90-23: IV-D-2) claimed that
the 10 ppm concentration threshold specified by the process
unit alternative in §63. 138 (d) should be replaced by a vapor
pressure threshold. The commenter (A-90-23: IV-D-2)
indicated that streams with concentrations as low as 10 ppm
would have very low emissions.
Response ; The EPA disagrees with the commenter 's
suggestion to use a vapor pressure threshold instead of the
10 ppmw threshold in the process unit alternative of
§63. 138 (d) . The EPA established the process unit alternative
control option to provide greater flexibility to facilities
for complying with the HON. Also, if an owner or operator
chooses to comply with the HON using this option, a
Group I/Group 2 determination is not necessary because all
wastewater streams from the process unit are controlled.
Changing the alternative control option to be based on a vapor
pressure rather than a concentration is not consistent with
the other compliance thresholds that are established in the
rule. To comply with the process unit alternative, the EPA
5-16
-------
therefore continues to require the owner or operator to
achieve a total VOHAP average concentration of 10 ppmw for
each process wastewater stream exiting a process unit.
5-17
-------
-------
6.0 COMPLIANCE DEMONSTRATIONS
6.1 BIOLOGICAL TREATMENT
Comment; One commenter (A-90-23: IV-D-20) indicated
that proposed §63,145(i)(1), which discusses the procedure for
determining compliance, does not address the situation where a
biological treatment system is used in conjunction with other
treatment systems or when a biological treatment system is
used but is vented to a control device. The commenter
(A-90-23: IV-D-20) stated that such treatment options are
currently in use at some SOCMI facilities and should therefore
be addressed in the rule.
Response: Situations where a biological treatment system
is used in conjunction with other treatment technologies are
covered by §63.145. If the biological treatment unit meets
the required mass removal provisions of §63.138(b)(1)(iii)(C),
(c)(1)(iii)(D), or (e), then the owner or operator must
demonstrate compliance by the corresponding procedures in
§63.145.
A biological treatment unit which meets the required mass
removal provisions is not required to be covered and vented to
a control device.
In the example cited by the commenter, the biological
treatment unit is vented to a control device. Therefore, the
owner or operator must demonstrate compliance by the
procedures in §63.145(f) and (h)(1), for wastewater streams
which are Group l for table 8 HAP's, or by the procedures in
§63.145(g) and (h)(1) for wastewater streams which are Group 1
for table 9 HAP's. These procedures may be used to
demonstrate compliance with the required mass removal
provisions for treatment devices other than biological
6-1
-------
treatment units. It should also be noted that the control
device to which the biological treatment unit is vented must
be in compliance with §63.139.
Comment; One commenter (A-90-19: IV-D-85) stated that
the EPA should either delete the equations in §63.145(1)(2},
which allows for the use of biological treatment in lieu of
steam stripping, or make sure that the equation cannot lead to
increased emissions. The commenter (A-90-19: IV-D-85)
suggested that if the equations remain available for use, the
EPA should further evaluate them in terms of likely
performance with complex mixed streams. The commenter
(A-90-19: IV-D-85) suggested that the equations should not be
available for conditions that the EPA is unable to validate;.
The commenter (A-90-19: IV-D-85) objected to the use of
total HAP's as the single parameter governing the equation and
stated that all high-risk pollutants should be considered.
The commenter (A-90-19: IV-D-85) argued that no SOCMI
facility should be able to use the equation to demonstrate
compliance unless the wastewater treatment system is
completely covered up to the biological treatment system.
Response; The equation referred to by the commenter
appears at §63.145(i)(2) of the proposed rule, and at
§63.145(h)(2) of the final rule. The equation cannot lead to
HAP emission reductions less than those achieved by steam
stripping because the equation is used to demonstrate that HAP
emission reductions achieved are equivalent to or greater than
those achieved by steam stripping, as provided for in the
provisions stated in §63.145(h)(2). Additionally, the rule
requires that all Group 1 vastevater streams be covered up
until all treatment requirements in §63.138 are achieved.
The commenter is correct that the equation at
§63.145(h)(2) estimates the total HAP emission reduction, and
not the HAP emission reduction of individual HAP's.
Section 112(d) of the CAA requires the EPA to develop
technology-based standards which obtain the maximum reduction
in HAP emissions. The CAA does not require individual
speciation of each HAP and the cost of such a demonstration
6-2
-------
and the additional monitoring, reporting, and recordkeeping
requirements are not warranted.
The EPA clarifies that the equation in §63.145(h) (2) can
account for multi -component interactions. For example, the
use of WATER? with site-specific biokinetic parameters
determined by the procedures in appendix C of part 63 will
account for multi-component interactions.
6.1.1 Method 304
One commenter (A-90-19: IV-D-32) stated that
the regulation should allow the use of biodegradation kinetic
coefficients predicted from respirometric studies
(i.e., UNIFAC fragment approach) and chemical structure in the
biological treatment unit simulation models. Two commenter s
(A-90-19: IV-D-32), (A-90-23: IV-D-20) contended that the
EPA should allow kinetic constants predicted by this
methodology to be substituted for the default constants in
WATER?, or to be used in other acceptable biological treatment
simulation models to predict the relative fractions of
volatilization and biodegradation in full-scale treatment
systems for the purpose of demonstrating equivalency to the
RCT.
Response: The EPA has revised the final rule to provide
more flexibility in determining site-specific biodegradation
constants, and has included these provisions in appendix C of
part 63. These provisions allow facilities to perform site-
specific testing as an alternative to Method 304A or 304B.
eom^i«»n^ ; Several commenters (A-90-19: IV-D-32; IV-D-33;
IV-D-34; IV-D-75) (A-90-23: IV-D-20) recommended that plants
should be allowed to use procedures other than Method 304 and
WATER?, which are still under development, for determining
biodegradation kinetics and demonstrating that biological
treatment provides effective control of HAP's. Two commenters
(A-90-19: IV-D-32; IV-D-75) proposed that a table of
acceptable procedures be included in the rule and that a
separate document describing these procedures be published.
Two commenters (A-90-19: IV-D-32; IV-D-75) claimed that there
are better methods than Method 304 to predict biodegradation
6-3
-------
kinetics for WATER?. The conunenters (A-90-19: IV-D-32;
IV-D-75) stated that the inclusion of Method 304 and WATER7 in
the regulation will discourage efforts to develop more
reliable and less labor-intensive methods.
One commenter (A-90-19: IV-D-32) recommended the use of
an alternative, direct method for determining the fraction of
a HAP that is biodegraded and the fraction that is emitted to
the air on a site-specific basis. One commenter (A-90-19:
IV-D-34) stated that Method 304 should be deleted as a
required test method to demonstrate biodegradation, and
alternate test methods recommended by the CMA should be
included.
One commenter (A-90-19: IV-D-75) claimed that neither
Method 304 nor any other method should be required for
biological treatment, because a biological treatment unit that
is operated within the ranges of certain parameters has stable
removal efficiency. The commenter (A-90-19: IV-D-75) stated
that these parameters are sufficient for demonstrating
compliance and achieving efficient removal of HAP's.
Response: The EPA is allowing two other options in
addition to Methods 304A and 304B for demonstrating effective
treatment with biological systems. These are outlined in
appendix C of part 63, "Determination of Fraction Biodegraded
(Fbic/ in a Biological Treatment Unit." When an option to
determine Fbio requires a model to be used, BASTE and TOXCHEM
will be allowed, as well as WATER7. The options in appendix C
are discussed more fully in section 4.3 of BID volume 2E.
Operating parameters are sufficient to show compliance
once Fbio is determined (i.e., once the mass removal is
determined to be based on biodegradation rather than
volatilization). However, the EPA disagrees with the
commenter's assertion that monitoring alone is sufficient for
demonstrating compliance for biological systems. The EPA
emphasizes that after the owner or operator demonstrates that
compliance is achieved through biodegradation not
volatilization, the operating parameters, which are based on
6-4
-------
operating conditions during the performance test, are adequate
to show compliance.
The EPA is considering the comment er ' s suggestion to
write a guidance document as a companion to appendix C of
part 63.
One commenter (A-90-19: IV-D-32) stated that
the regulation should be clarified to state that for new
treatment systems, an engineering estimate of the design
hydraulic retention time should be used in Method 304 when the
full-scale system is not in operation.
Response; The EPA clarifies that the hydraulic retention
time used for new systems not yet in operation must be the
same as the hydraulic retention time of the system as it will
be operated.
6.1.2 Compliance Issues
Comment ; One commenter (A-90-19: IV-D-97) recommended
that the EPA simplify compliance demonstrations for biological
treatment units so that both large and small SOCMI facilities
can continue to use their existing biological treatment units.
Response; The commenter did not discuss which aspects of
compliance demonstration for biological units should be
simplified or how the compliance demonstration requirements
should be revised. The proposed and final wastewater
provisions allow SOCMI facilities to use existing biological
units for treating Group 1 process wastewater streams,
provided that the level of treatment achieved is equivalent to
the reference control technology.
co]ifln«an-h •- one commenter (A-90-19: IV-D-85) stated that
since the efficiency of biological treatment units is based on
many variables, the EPA should require a high degree of proof
from industry to show that a biological treatment unit can
achieve an equivalent level of treatment as the level achieved
by the design steam stripper.
Response ; The EPA believes the final rule does require a
high degree of proof from industry to show that a biological
treatment unit can achieve a level of treatment equivalent to
the RCT. The rule requires that the actual HAP mass removal
6-5
-------
achieved by the biological treatment unit as determined by the
procedures in §63.145(h)(2) is equal to or exceeds the
required mass removal as determined by the procedures in
§63.145(f) for new sources or §63.145(g) for new and existing
sources. The required mass removal is the mass removal that
would be achieved by the RCT. Alternatively, the owner or
operator can demonstrate the biological treatment unit is
achieving 95 percent HAP mass reduction by the procedures in
§63.145(i) . Additionally, the owner or operator must select
parameters to be monitored which will insure that the
biological treatment unit will remain in compliance.
6.2 MONOD EQUATION AND ALTERNATIVE KINETICS FORMULAS
qqmffliafit; One commenter (A-90-19: IV-D-32) supported the
EPA's selection of the Monod equation to simulate
biodegradation kinetics in the WATER? model. Three commenters
(A-90-19: IV-D-32; IV-D-34; IV-D-75) requested that
simulation models such as PAVE, TOXCHEM, BASTE, and CINCI be
acceptable methods for demonstrating that an enhanced
biological treatment system complies with the HON.
Response; The EPA agrees that alternative kinetics
formulations can be used to simulate biodegradation kinetics
when such a formulation is found to provide reasonable site-
specific emission estimates. The EPA has added appendix C to
40 CFR part 63 to provide detailed guidance for demonstrating
compliance with the provisions for biological treatment in the
final regulation.
The EPA has not included the PAVE model in appendix C.
The Henry's law value is an important input parameter for
estimating emissions from wastewater. The PAVE model does not
allow for the input of the Henry's law constant, and it is
unclear if or how the PAVE model estimates this parameter. In
addition, PAVE is designed to evaluate one chemical at a time
and it calculates the biomass as an output. Models used in
appendix C need to be able to evaluate multicomponent streams,
and input the biomass for the system.
6-6
-------
6.3 PERFORMANCE TESTING
Cf^mpent:; One conunenter (A-90-19: IV-D-85) stated that
the proposed HON may not provide a reliable basis for
evaluating the equivalence of biological treatment units. The
commenter (A-90-19: IV-D-85) stated that the models used for
demonstrating compliance of biological treatment units in the
proposed HON are based on the Monod equation, which may not be
an appropriate basis for evaluating biodegradation of toxic
streams. The commenter (A-90-19: IV-D-85) referenced an
article authored by Clay, S.G., Boud, Jr., A.F.; Rozich, A.F.;
Moran, N.R., titled "Using Respirometry to Assess Waste
Streams and Set Surcharges." Water Environment and
Technology. June 1992, pages 60-65, which indicated that the
Monod equation may be an unreliable predictor of the rate of
biodegradation.
Response: The article cited by the commenter suggests
that the Monod equation will not accurately predict the
specific growth rate of biomass when inhibitory substrates are
present. Biomass growth rate inhibition may occur when the
biomass is exposed to unexpectedly high concentrations of
organic compounds or organic compounds to which the biomass is
not climatized. The article referenced by the commenter
suggests that the Haldane equation can be used to predict the
specific growth rate of biomass when inhibitory substrates are
present.
Facilities in the SOCMI will operate their biological
treatment units in a manner consistent with wastewater
discharge permit requirements and will avoid upset conditions.
Upset conditions, if they occurred, could contribute to
inhibition of the biomass growth rate, resulting in violations
of permitted discharge limits. Inhibition of the biomass
growth rate is not expected to occur except upon infrequent
abnormal operations.
Also, the owner or operator of a biological treatment
unit used to comply with the HON must use the procedures
specified in appendix C of part 63 to ensure compliance with
the HAP emission control requirements in §63.138 of subpart G.
6-7
-------
comment:; One commenter (A-90-19: IV-D-85) stated that
proposed §63.145(a)(1), which requires performance tests for
demonstrating compliance with the wastewater treatment
provisions of §63.138 in the proposed regulation, is
inadequate because plant operators are not required to conduct
performance tests to estimate future credits and debits.
The commenter (A-90-19: IV-D-85) disagreed with the use
of either "process knowledge" or "records" of the mass
concentrations and flow rates to estimate emissions debits and
credits because this approach allows plant operators to pick
and choose the method which suits them best, rather than
requiring operators to choose the most accurate technique or
to verify the accuracy of a record.
Another commenter (A-90-19: IV-D-45 and IV-F-7.7)
opposed the use of engineering calculations instead of
requiring initial performance testing for determining
wastewater treatment plant compliance with the provisions of
§63.138.
Response; In allowing the use of process knowledge the
EPA took into consideration that different methods for
determining debits and credits for emissions averaging or for
demonstrating compliance with §63.138 have different
uncertainties associated with them in terms of accuracy. The
EPA also took into consideration that, for some facilities,
sampling and testing may be impractical, unsafe, or too
costly. In some cases, the availability of wastewater flow
measurement data, analytical data, or design data may make
sampling and testing unnecessary for all wastewater streams.
Therefore, the EPA has provided facilities the option of using
process knowledge and records for purposes of determining
credits and debits for emissions averaging and for
demonstrating compliance with §63.138.
Comment1 Two commenters (A-90-19: IV-D-32; IV-D-77)
recommended that the EPA clarify in proposed §63.138(f) that
performance testing is not required for a design steam
stripper.
6-8
-------
Response; There is no language in the proposed or final
rules to suggest that performance testing is required to
demonstrate compliance with the design steam stripper
provisions [§63. 138 (f) in the proposed rule and §63. 138 (g) in
the final rule] . Therefore, the EPA maintains that
clarification is not necessary.
TWO commenters (A-90-19: IV-D-70; IV-D-99)
asserted that design steam strippers should be subject to at
least one performance test to establish control efficiency and
design parameters to be monitored. The commenters (A-90-19:
IV-D-70; IV-D-99) claimed that average parameter values should
be determined hourly and should be based on data gathered
every 15 minutes.
Response : The EPA's analysis shows that the design steam
stripper will achieve the required treatment efficiencies
required in the final rule. Therefore, performance tests are
not required. The EPA clarifies the rule provides for
monitoring of design parameters, including wastewater feed
temperature, steam flow rate, and wastewater feed rate.
fp9PP«»p^ • one commenter (A-90-19: IV-D-64) supported the
acceptance of design analyses and documentation in proposed
§63. 138 (i) (1) as an alternative to performance tests.
Response; The EPA clarifies that the acceptance of
design analyses and documentation in 63.138(j)(l) of the final
rule as an alternative to demonstrating compliance through
testing applies only to 63 . 138 (b) (1) , (c) (1) and (d) . If a
biological treatment unit is used to comply with the HON, an
owner or operator must follow the procedures in appendix C of
part 63 to ensure compliance. The EPA does not intend for
design analysis to be used for biological treatment units.
6.4 METHODS 25D AND 305
one commenter (A-90-19: IV-D-50) was concerned
with the applicability of Method 25D and Method 305 because
validation studies have not been released. The commenter
(A-90-19: IV-D-50) specifically expressed concern with the
detection abilities of these two methods in a wastewater
6-9
-------
containing VOC's other than HAP's, and the availability of
labs to run the analysis.
Response; The draft validation study for Method 25D is
available and is titled "Method 25D Recovery Factors,"
contract no. 68D90055, October 1991.
Comment; One conunenter (A-90-19: IV-D-77) stated that
the EPA should clarify that Method 305 is limited to testing
for table 9 HAP's. The commenter (A-90-19: IV-D-77) also
stated that the EPA needs to clarify §63.144(b), which
provides options on how to determine whether HAP's are
present, because 1- 'hod 305 seems to test for more than just
table 9 HAP's.
Response; Ir rder to determine if a wastewater stream
is subject to the ^astewater provisions of the HON, the annual
average flow rate, and the annual average VOHAP concentration
must be determined. There are three options for measuring
VOHAP concentrations.
The first option is to directly measure the VOHAP
concentration of each individual hazardous air pollutant (HAP)
in the wastewater using Method 305. The total VOHAP
concentration is the sum of the individual compound VOHAP
concentrations.
A second option is to use Method 250 and the total
volatile organic (VO) concentration as a surrogate for VOHAP
concentration. Method 25D does not provide speciation, and
will measure both HAP and non-HAP compounds. The result is a
single concentration which represents the total volatile
organic concentration in the wastewater. Under this option,
there is no speciation and it is assumed that the VO
concentration equals the VOHAP concentration. Therefore, this
option makes the most sense for wastewater streams containing
only HAP's regulated under subpart G, or when the ratio of
non-HAP's to HAP's is low.
The third option is to use a method other than Method 305
which measures individual organic HAP concentrations in the
wastewater. The individual concentrations, however, can be
6-10
-------
corrected to their concentrations as if they had been measured
by Method 305, by multiplying each concentration by the
compound-specific fraction measured (Fm) values in table 34 of
subpart G.
The applicability of Method 305 is not limited to testing
for table 9 HAP's. Method 305 will detect organic compounds
other than those specified in table 9 of subpart G. However,
Method 305 will speciate, so that compounds listed on table 9
of subpart G can be identified. Only table 9 HAP's are
subject to regulation under the wastewater provisions of the
HON.
6.5 TESTING AT PEAK LEVELS FOR COMPLIANCE DEMONSTRATION
Comment: Several commenters (A-90-19: IV-D-32; IV-D-33;
IV-D-110; IV-D-112) stated that the requirement in
§63.145(a)(1) to test for compliance in treatment processes
and waste management units when flow rates and VOHAP
concentrations are at peak levels is technically infeasible
and does not represent annual average conditions. One
commenter (A-90-19: IV-D-112) suggested that compliance
testing be performed when the SOCMI process is operating at
the production rate or annual average flow rate determined
pursuant to §63.144. One commenter (A-90-23: IV-D-20)
indicated that if some facility processes only operated at a
particular time of the year and others were only run at peak
production capacity once every 2 years, conducting a
compliance test on peak VOHAP concentration generation levels
could be very difficult. The commenter (A-90-23: IV-D-20)
suggested that if the most difficult compliance conditions are
not reasonably available for performance testing, then the
most difficult available conditions should be used.
Response; The EPA continues to maintain the same
regulatory language in §63.145(a)(1), but clarifies that an
owner or operator may use the most difficult available
conditions and provide rationale through extrapolation for how
compliance with the HON shall be achieved under the most
difficult conditions.
6-11
-------
6.6 USE OF MODELS TO SHOW COMPLIANCE FOR ALTERNATIVE CONTROL
TECHNOLOGY
Comment; One commenter (A-90-19: IV-D-75) supported the
use of models to determine whether an alternative control
technology meets the RCT treatment requirements. The
commenter (A-90-19: IV-D-75) claimed that the models are
reliable when used with appropriate physical property
information and that models were used to establish RCT
performance and as a basis for EPA's economic evaluation.
Several commenters (A-90-19: IV-D-32; IV-D-108) (A-90-23:
IV-D-20) argued that facilities should be allowed to use ASPEN
simulations to demonstrate the equivalency of alternative
steam stripper designs with the RCT, since the EPA has based
their design on ASPEN simulations. One commenter (A-90-19:
IV-D-32) provided data in appendix N of the comment letter
which contains examples of simulation model results. One
commenter (A-90-19: IV-D-64) encouraged the EPA to streamline
the process for approving alternative stripper designs under
§63.143(d). One commenter (A-90-19: IV-D-75) said that the
requirements for performance testing and monitoring are an
unnecessary expense.
One commenter (A-90-19: IV-D-73) claimed that simulation
models adequately determine HAP removal efficiency of steam
strippers and also identify the critical parameters for
monitoring, recordkeeping, and reporting.
Response: It is unclear what one commenter (A-90-19:
IV-D-75) means by "alternate control technology". The EPA
clarifies that any control technology can be used as a
treatment device to meet the provisions of §63.138 if the
technology achieves HAP emission reductions equivalent to the
wastewater RCT. The demonstration of compliance can be made
by a design analysis and supporting documentation as provided
in §63.138(j)(1) or by conducting performance tests using the
test methods and procedures in §63.145 as referenced by
§63.138(j)(2).
Table 12 of the final rule provides for monitoring of
alternative parameters for treatment processes. The request
6-12
-------
to the implementing agency for monitoring alternative
parameters must include a description of the methods used to
monitor. These methods may include the use of simulation
models.
6.7 AVAILABILITY OF COMBUSTION TECHNOLOGIES
Comment; One commenter (A-90-19: IV-D-107) stated that
several combustion or recovery technologies are available for
use in meeting the requirements of the rule and each should be
allowed to compete in the market place.
Response; The EPA clarifies that any control technology
can be used to comply with the rule if the technology achieves
HAP emission reductions equivalent to the RCT and meets the
requirements for control devices specified in the rule.
6.8 USE OF EPA-APPROVED METHODS
Cg^mpnen-t; Several commenters (A-90-19: IV-D-32; IV-D-33;
IV-D-53; IV-D-79) asserted that methods previously approved by
the EPA should not need to be validated using Method 301. One
commenter (A-90-19: IV-D-33) stated that in addition to the
analytical methods listed in §63.143(b) for the parameters
that must be monitored, the HON should allow a facility to use
any relevant method approved by the EPA for compliance with
CWA requirements, and the facility should not be required to
validate the method because the EPA will have already
determined the method to be valid. One commenter (A-90-19:
IV-D-32) included a preliminary list of EPA-validated methods,
which industry currently uses when conducting performance
tests. One commenter (A-90-19: IV-D-53) recommended that the
EPA include a list of methods which do not have to be
validated using Method 301. The commenter (A-90-19: IV-D-53)
said that the list should at least include Methods 8020, 8021,
8240, 8260, 602, and 604. One commenter (A-90-19: IV-D-33)
suggested language to amend §66.144(b)(3)(iii)(B). One
commenter (A-90-19: IV-D-32) asserted that alternate testing
and analytical procedures on which the CMA has commented
should not require validation and approval by the EPA before
they are used. The commenter (A-90-19: IV-D-32) stated that
these methods, which the CMA has deemed equivalent or more
6-13
-------
appropriate than the proposed procedures, be cited as
acceptable alternatives for monitoring for each situation and
process to which they are applicable.
Response ; The methods that the commenter suggested are
OSW (SW-846) and OW methods. These methods were developed for
different types of source categories for small subsets of
compounds which are on the HON target list. Both those
offices have less stringent acceptance criteria for when
percent recovery is acceptable. For example, OSW methods
allow 50 to 150 percent recovery of target compounds as
acceptable, while Method 301 allows 70 to 130 percent recovery
and a correction procedure. An owner or operator would have
to validate the method, using Method 301, as if no other valid
method existed. They validate the method as measuring the
target compound in the water, then correct with the Fm factor.
Therefore, no PEG sampling is required. The final rule
continues to allow an owner or operator to use any method as
long as it is verified by Method 301. In the proposal, the
EPA requested evaluation data on other methods and did not
receive any. Therefore, the EPA continues to require
Method 301 for validation.
6.9 MONITORING REQUIREMENTS FOR RECYCLED STREAMS
One commenter (A-90-19: IV-D-54) stated that
the monitoring requirements in proposed §63.143 in table 11 do
not seem appropriate for wastewater streams that are recycled.
The commenter (A-90-19: IV-D-54) stated that recycle streams
should not be subject to the monitoring requirements in
§63. 143 (b) and table 11 because the information that is
obtained would not serve any useful regulatory purpose. One
commenter (A-90-19: IV-D-32) stated that wastewaters that, are
recycled wholly within a SOCMI process, and which are not
exposed to the atmosphere, represent no potential to emit and
should not be subject to wastewater monitoring requirements.
Response i The EPA clarifies the monitoring requirements
in table 11 of the proposed rule (table 12 of the final rule)
did not apply to recycled streams. The monitoring
6-14
-------
requirements for waste management units in table 11 of the
final rule do apply to recycled streams.
The treatment requirements in §63.138 allow an owner or
operator to comply by either using one of the specified
treatment processes for wastewater and residuals or by
recycling the wastewater or residuals to the process. The
provisions for recycled streams state that: (1) the wastewater
stream or residual must not be exposed to the atmosphere; and
(2) each waste management unit that treats the recycled
residual or recycled wastewater, prior to or during recycle,
must meet the requirements of §63.133 through 63.137 of
subpart G. Sections 63.133 through 63.137 contain the
inspection and monitoring requirements for waste management
units. These requirements are listed in table 11 in the final
rule.
6.10 VENDORS
Comment : One vendor (A-90-19: IV-D-8) provided
information to the EPA on a leak detection device to be used
instead of Method 21 for compliance with the inspection
provisions for collection and treatment systems.
Response ; The EPA has provided a discussion on this
alternate leak detection device in section 5.0 'of BID
Volume 2A.
6.11 INSPECTIONS
One commenter (A-90-19: IV-D-73) supported the
storage vessel floating roof inspection provisions for
wastewater tanks that meet the Group 1 wastewater tank
criteria in §63.133(c) and (d) . The commenter (A-90-19:
IV-D-73) stated that the semi-annual inspection requirement
for wastewater tanks is excessive and should be replaced with
the requirement for storage tank floating roof inspections.
The commenter (A-90-19: IV-D-73) recommended deleting
§63.133(f) .
Response ; The EPA has corrected an error in proposed
§63. 133 (f) to clarify the original intent of the regulation.
In the final rule, the EPA intends for improper work practices
associated with wastewater tanks and for one type of control
6-15
-------
equipment failure in §63.133(g)(l)(ix) (i.e., cracked, gapped,
or broken gaskets, joints, lids, covers, or doors) to be
visually inspected initially and semi-annually thereafter.
Regarding the inspection provisions for the remainder of the
control equipment failures for wastewater tanks, the EPA has
corrected the wording of the provisions in §63.133(f) to refer
owners and operators to the inspection schedule for storage
vessel floating roofs. The final rule no longer requires
semi-annual inspections for such failures.
Comment; One commenter (A-90-19: IV-D-73) suggested
establishing a uniform annual inspection frequency for all
surface impoundments, containers, and individual drain
systems.
Response; The EPA maintains that the inspection
requirements for surface impoundments, containers, and
individual drain systems are consistent and include: (1) an
inspection to detect leaks in covers; and (2) inspections for
improper work practices and control equipment failures. The
leak inspection provisions in §63.148 of the final rule for
surface impoundments, containers, and individual drain systems
include an initial inspection of covers using Method 21 and
semi-annual visual inspections of covers for visible, audible,
or olfactory indications of leaks. Inspection of surface
impoundments, containers, and individual drain systems for
improper work practices and control equipment failures is
required initially, and semi-annually thereafter. Semi-annual
inspection for improper work practices and control equipment
failures is required because these types of failures
(e.g., leaving the cover off a container) would cause greater
emissions than a leak in a cover.
qnmimajTfr« one commenter (A-90-19: IV-D-117) suggested
that operations personnel visually inspect drain covers at a
minimum of once per day instead of semi-annually claiming that
this would n~u cause an increase in control costs and would
help prevent emissions.
Response: The EPA disagrees with the commenter (A-90-19:
IV-D-117) that daily inspections would not increase the
6-16
-------
control costs for wastewater. The annual cost of compliance
with the HON includes a labor cost which is equal to the labor
wage rate ($/hr) multiplied by the number of hours per year
the laborer spends to keep the facility in compliance with the
HON. Increasing the number of hours per year that the laborer
spends on inspections will increase the annual cost, and thus,
increase the control cost.
cement; One commenter (A-90-19: IV-D-64) stated that
the EPA should reconsider whether an LDAR program is justified
in as many places as it appears in subpart G. The commenter
(A-90-19: IV-D-64) expressed particular concern about the
LDAR program as it applies to low pressure and low temperature
closed-vent systems. The commenter (A-90-19: IV-D-64)
contended that an LDAR program may not be necessary for this
type of control device. The commenter (A-90-19: IV-D-64)
stated that the EPA should set a de minimis VOHAP
concentration in- wastewater for the LDAR provisions below
which, monitoring would not be required. The commenter
(A-90-19: IV-D-64) stated that monitoring should not be
required for equipment in which the total VOHAP concentration
is less than 5 percent by weight and contended that
engineering assessments should be adequate for determination
of total VOHAP concentration. The commenter (A-90-19:
IV-D-64) contended that the delay of repair provisions in
§63.171 of subpart H should also be included in the LDAR
provisions in subpart G.
Response; The EPA has determined that a total VOHAP
concentration in wastewater lower than 5 percent by weight can
produce a concentration in the vapor space of a tank,
container, etc., of 500 parts per million by volume above
background from a leak. Therefore, the EPA has not added a
de minimis VOHAP concentration for vastewater below which
monitoring would not be required.
The leak inspection provisions for subpart G in
§63.148(e) of the final rule specify that delay of repair can
be invoked if the repair is technically infeasible without a
6-17
-------
process unit shutdown or if the emissions resulting from
immediate repair would be greater than the fugitive emissions
resulting from delay of repair.
gomjmeyTt; One commenter (A-90-19: IV-D-73) requested
that the inspection and monitoring requirements of wastewater
closed-vent systems be regulated under the closed-vent
provisions in §63.160(a) and §63.172 of subpart H, because
common closed-vent systems often serve various types of
emission points. The commenter (A-90-19: IV-D-73) agreed
with the exclusions for bleeds and drains in §63.139(h), and
for waiving component inspections if unsafe in §63.139(e).
The commenter (A-90-19: IV-D-73) also agreed with the delay
of repair provisions in §63.171 and favored retaining these
three sections in the closed-vent system provisions.
Response; The EPA assumed by "drains," the commenter was
referring to low leg drains. The EPA considered placing all
of the inspection provisions that apply to closed-vent systems
in §63.172 of subpart H; however, the EPA concluded that the
regulation would be more clear if the closed-vent system
requirements that apply to the SOCMI are consolidated within
subpart G. Therefore, the EPA has combined the requirements
for closed-vent systems into §63.148, which was a reserved
section in the proposed rule. The reason for consolidating
the closed-vent system requirements into one section is
because common closed-vent systems often serve different
emission points. This reorganization also reduces repetition
within the regulation.
r^l^n^; One commenter (A-90-19: IV-D-64) supported the
provisions in §63.140 for delay of leak repairs when repair is
infeasible without a process shutdown, but suggested that
these provisions be moved to the general standards in §63.102
and be applied to all LOAR programs in subpart 6.
Response: The leak inspection provisions for all of
subpart 6 have been moved to §63.148 of subpart G in the final
rule. Section 63.148(e) of subpart G allows delay of repair
for leaks if the repair is technically infeasible without a
process unit shutdown or if the emissions from immediate
6-18
-------
repair of the leak are greater than the emissions that would
result from delay of repair. Since the provision in §63.148
of subpart G only apply to subpart G, they were not moved to
the general standards provisions in §63.102 of subpart F in
the final rule because subpart F applies to subpart H as well
as subpart G.
6.12 MONITORING
Comment; Two commenters (A-90-19: IV-D-34) (A-90-23:
IV-D-20) stated that the ranges required in tables 14a, 14b,
15a, 15b, and 16 should be deleted. One commenter (A-90-19:
IV-D-34) stated that the requirements should be deleted
because the concept of requiring ranges is not defined by the
methods which are necessary to determine the data. The other
commenter (A-90-23: IV-D-20) stated that §63.146, which
requires a source to report the VOHAP concentration range
provides no benefit because sources are required to make a
Group 1 or Group 2 determination reflecting annual averages
and ranges.
Response; In the final rule, the requirement to provide
VOHAP concentration ranges in tables 14a, 14b, I5a, 15b, and
16 of subpart G has been eliminated. Furthermore, there are
not specific methods required for the determination of VOHAP
concentration. Section 63.144(b) of subpart G in the final
rule allows knowledge of wastewater, bench-scale or pilot-
scale test data, or sampling measurements to determine VOHAP
concentration. Sampling measurements may be analyzed to
determine VOHAP concentration using Method 305, Method 25D, or
any other method validated according to section 5.1 or 5.3 of
Method 301.
6.12.1 Treatment Processes
gQ1W**T1t'; One commenter (A-90-23: IV-D-20) stated that
the monitoring methods in table 11 in §63.143 should be made
consistent by allowing Method 305 and any other applicable
method, which has been validated using section 5.1 or 5.3 of
Method 301, to be used to monitor items 1 and 3 of the table.
Responsei The proposed rule did allow what the commenter
has requested. The monitoring methods in table 11 in §63.143
6-19
-------
of the proposed rule specified that Method 305 and any other
applicable method which has been validated using section 5.1
or 5.3 of Method 301 could be used for items 1 and 3. In the
final rule, monitoring items 1 through 6 have been eliminated
from the monitoring requirements for treatment processes
because these items are actually performance tests that can be
used for demonstrating compliance. Section 63.145 of the
final rule contains the provisions for determining compliance
and these provisions allow the use of Method 305 or any other
method that has been validated according to section 5.1 or 5.3
of Method 301 for the measurement of VOHAP concentration. The
monitorir.g requirements for treatment processes are in
table 12 of subpart G in the final rule.
One commenter (A-90-23: IV-D-20) stated that
§63 . 145 (c) (3) (i) , which requires that flow meters be used on
both the inlet and outlet flow of treatment processes, should
be modified to allow a SOCMI source the option of placing a
single flow meter at the inlet when the treatment process has
an outlet flow that is not greater than the inlet (such as
adsorption or biological treatment) .
Response; The EPA agrees that in instances when the
outlet flow is not greater than the inlet flow,' a flow meter
at either the inlet or outlet is sufficient for determining
the flow rate.
gmrnnant ; One commenter (A-90-19: IV-D-73) suggested
x .t the EPA reduce the frequency of monitoring for treatment
alternatives 1 - 7 in table 11 of §63.143, and that the
monitoring schedule should be based on the probability of
emission exceedance estimated from one year of monthly
monitoring data. The commenter (A-90-19: IV-D-73) provided a
statistical approach to determine monitoring frequency. The
c imenter (A-90-19: IV-D-73) claimed that this approach would
reward processes with high efficiencies and encourage
facilities to install processes with high efficiencies to
decrease required sampling frequency. One commenter (A-90-19:
IV-D-89) suggested using quarterly monitoring requirements in
table 11 of proposed subpart G. The commenter (A-90-19:
6-20
-------
IV-D-89) recommended keeping continuous data on critical
operating variables and considering a QA/QC program similar to
the valve QA/QC program listed in subpart H.
Response: The monitoring requirements for treatment
alternatives 1-6 in table 11 of the proposed subpart G have
been eliminated. Proposed table 11 is table 12 in the final
rule. The EPA has determined that these monitoring
requirements were performance tests that are required by
§63.145 of subpart G to demonstrate compliance. The
monitoring requirements for treatment alternative 7 in
table 11 of the proposed rule have been increased from monthly
monitoring requirements to continuous monitoring requirements
in order to be consistent with, and as stringent as, the
monitoring requirements for steam stripping. For
alternative 7 from table 11 of the proposed rule, Method 304A
or 304B is required initially, and those parameters that are
monitored upon approval from the permitting authority must be
monitored continuously.
6.12.2 Waste Management Units
co^«»nt; one commenter (A-90-19: IV-D-32) claimed that
annual monitoring, semi-annual visual inspection, and repair
may not be possible for all portable containers. The
commenter (A-90-19: IV-D-32) reasoned that some portable
containers may no longer be onsite or may not be owned by the
plant owner. Thus, the commenter (A-90-19: IV-D-32)
concluded that these requirements should be deleted from the
rule.
Response; The EPA recognizes that many containers, which
by definition must be portable, may be sent offsite or may be
owned by someone other than the owner or operator of a SOCMI
plant. In the final rule, the EPA has reduced the inspection
requirements for containers. The EPA has revised table 11 of
subpart G of the final rule to require the owner or operator
to perform Method 21 testing initially but the EPA has deleted
the requirement to inspect containers semi-annually for
improper work practices. In fact, only certain large
6-21
-------
containers still require any monitoring. Refer to
section 4.1.7 of this BID volume for more information.
Comment; Two commenters (A-90-19: IV-D-32), (A-90-23:
IV-D-20) stated that the EPA should exempt surface
impoundments that are operated under a vacuum from leak test
requirements in §63.134.
Response; In the final rule, §63.134(b)(4) of subpart G
exempts from the leak detection requirements any cover on a
surface impoundment and the corresponding control device if
the cover and control device are operated and maintained under
negative pressure.
Comment; One commenter (A-90-19: IV-D-34) stated that
the period of repair for floating roofs on oil-water
separators is too short and would be difficult to complete
within 15 days. The commenter (A-90-19: IV-D-34) suggested
that 45 days be provided, which is the same time frame for
wastewater tank repairs. Two commenters (A-90-19: IV-D-64)
(A-90-23: IV-D-20) stated that allowing only 15 days for
final repair of a leak in a surface impoundment in §63.134(d)
and 45 days for final repair of a leak in a wastewater tank in
§63.133(g) seems inconsistent. The commenters (A-90-19:
IV-D-64) (A-90-23: IV-D-20) requested that the EPA modify the
leak repair provisions for surface impoundments to be
consistent with the requirements for wastewater tanks.
Response; The EPA believes that an error in the proposed
rule has led to confusion regarding the period allowed for
making repairs. The proposed §63.133(g) implied that 45 days
would be allowed for repair of improper work practices,
control equipment failures, and leaks. This was not the EPA's
intent. The EPA intended to allow 15 dava for repair of leaks
from all waste management units. In the final rule, this has
been clarified by placing all subpart G provisions for leak
detection and repair in a new section (§63.148). This new
section includes provisions in §63.148(e) for delay of repair
if the repair is technically infeasible without a process unit
shutdown or if the emissions from immediate repair would be
greater than the emissions from delay of repair. It should be
6-22
-------
noted that a 15-day period for repair of leaks is consistent
with subpart H of the HON and with previous rules for
equipment leaks.
The EPA intended to allow 45 days for repair of control
equipment failures (e.g., repair of a floating roof) because
such repairs would likely take longer than 15 days to
complete. In the proposed rule, §63.133(g) indicated that
45 days would be allowed for repair of control equipment
failures for wastewater tanks. This repair period was
selected because it is consistent with the provisions for
repair of storage vessels. This allowance was inadvertently
not included in the proposed provisions for surface
impoundments and oil-water separators.
Because the repair provisions for control equipment
failures were specified in the same paragraph in the proposed
rule as those for leaks and improper work practices, it
appeared that the 15-day period for leak repair for surface
impoundments and oil-water separators would also apply for
repair of control equipment failures for these units. This
was not the EPA's intent.
In the final rule, it has been clarified that control
equipment failures for wastewater tanks, surface impoundments,
and oil-water separators must be repaired within 45 days.
It should be noted that both the proposed and final rule
include provisions in §63.140 for delay of repair of control
equipment failures and improper work practices if the repair
is technically infeasible without a process unit shutdown or
if the emissions from immediate repair would be greater than
the emissions from delay of repair.
qfflifflitpni;; One commenter (A-90-19: IV-D-64) expressed
concern about why an LOAR program should be required for the
roof and roof fittings of a vastewater tank, surface
impoundment, or oil-water separator when it is not required
for a fixed roof or internal floating roof storage vessel.
The commenter ( A-90-19: IV-D-64) stated that the LDAR
requirements should be removed from those appropriate sections
of the proposed wastewater regulation.
6-23
-------
Response; An LDAR program is not required for a fixed
roof or internal floating roof storage vessel, because such
vessels contain valuable product, and the EPA has determined
that leaks in these types of tanks will be carefully monitored
by the owner or operator to reduce product loss. Furthermore,
the leak detection and repair provisions for fixed roofs on
wastewater tanks, covers on surface impoundments, and fixed
roofs on oil-water separators have been amended for the final
rule. An inspection of the fixed roof or cover using
Method 21 is required initially in the final rule, as
proposed. However, the final rule requires an annual visual
i pection for visible, audible, or olfactory indications of
)cs, instead of an annual Method 21 inspection.
comment; TWO commenters (A-90-19: IV-D-32; IV-D-75)
argued that allowing visual monitoring of p-traps and s-traps
would eliminate unnecessary water flows and would be
consistent with pollution prevention practices without
increasing emissions. One commenter (A-90-19: IV-D-32)
stated that drains which are regularly used for discharges
should not be routinely inspected. The commenter (A-90-19::
IV-D-32) stated that the regulation should allow the use of
non-volatile organic liquids (e.g., glycols) to be used as a
vapor barrier in p-trap and s-trap drains. One commenter
(A-90-19: IV-D-73) suggested replacing the monitoring
requirements for drains with a general duty requirement to
operate traps and seals as designed.
Response; The requirements in §63.136(e)(1) of subpart G
in the final rule require the owner or operator to ensure that
water is maintained in a p-trap or s-trap. Verifying the
continuous flow of w.^ter to the trap is only an example of how
an owner or operator would verify the continuous presence of
water in a trap. Therefore, an owner or operator may choose
to visually monitor *the p-traps and s-traps to verify the
presence of water i :ead of monitoring the continuous flow of
water to the traps. :he monitoring requirements for p-traps
and s-traps in table 11 of subpart G have been modified to
clarify that monitoring the continuous flow of water to the
6-24
-------
traps is only one example of how to verify the presence of
water in a trap.
Routine inspection of drains is not required. Rather,
semi-annual inspections are required. It is unclear what the
commenter (A-90-19: IV-D-32) means by "drains that are
regularly used for discharges". If the owner or operator can
verify the continuous flow of water to the p-traps and
s-traps, then such drains would not require inspection.
It is not clear what the commenter (A-90-19: IV-D-73)
means by a "general duty requirement to operate traps and
seals as designed". However, if a "general duty requirement"
can verify the continuous presence of water in the drain, then
the traps and seals themselves need not be inspected.
The EPA does recommend "non-volatile" organic liquids
such as glycols to be used as vapor barriers in p-trap and
s-trap drains because water will work sufficiently as a vapor
barrier. Furthermore, some glycols are regulated by the
wastewater provisions of the HON.
6.12.3 Control Devices
Comment; One commenter (A-90-19: IV-D-32) stated that
certain wastewater monitoring requirements for closed-vent
systems in table 12 and §63.143 should be modified to make
them more appropriate for the range of control options that
may be used. The commenter (A-90-19: IV-D-32) suggested that
scrubbers (absorbers) are an example of a closed-vent control
device which is not listed in the monitoring requirements.
Response: The control devices listed in table 12 of
proposed subpart G (table 13 of the final rule) are only
examples of the control devices that can be used in
conjunction with closed-vent systems. Section 63.139(c)(l) of
the final rule states that an enclosed combustion device can
include but is not limited to vapor incinerators, boilers, or
process heaters, and §63.139(c)(2) states that a recovery
device can include but is not limited to a carbon adsorption
system or condenser. Furthermore, §63.139(c)(5) of the final
rule allows the use of any control device for a closed-vent
system that reduces emissions by 95 percent. The EPA realizes
6-25
-------
that the control devices listed in table 12 of the proposed
rule are only examples, but the EPA cannot list the monitoring
requirements for all possible control devices. If an owner or
operator chooses to use a closed-vent control device other
than those listed in table 13 of the final rule, the owner or
operator must obtain approval from the permitting authority
for the parameters that the owner or operator wishes to
monitor.
The EPA did not include absorbers as an example of a
closed-vent control device, because the most widely used
scrubbing medium for absorbers is water. The EPA assumed that
an owner or operator would not use an absorber with water as
the scrubbing medium to control the emissions of HAP's that
had previously been removed from wastewater. However,
absorbers can be used as closed-vent control devices as long
as they achieve an emission destruction of 95 percent.
Comment: One commenter (A-90-19: IV-D-89) suggested
that the monitoring frequency for non-regenerative carbon
adsorbers specified in table 12 of §63.143 should be extended
to 50 percent of the design replacement interval if there are
carbon adsorbers in series. The commenter (A-90-19: IV-D-89)
stated that since a second canister is on line,' the monitoring
frequency can be extended beyond the normal single canister
replacement interval.
Response; Regarding the monitoring frequency for non-
regenerative carbon adsorbers as specified in table 13 of
subpart G in the final rule, the EPA continues to require
organic compound concentration monitoring of the adsorber
exhaust either daily or at intervals no greater than
20 percent of the design carbon replacement interval,
whichever is greater. The EPA continues to include as an
alternative to this monitoring the option for the owner or
operator to replace the carbon in the carbon adsorption system
at a regular predetermined interval that is less than the
carbon replacement interval. The owner or operator must
consider (1) the maximum design flow rate and (2) the organic
concentration in the gas stream that is vented to the carbon
6-26
-------
adsorber when determining how often to replace the carbon.
Because the final rule allows for scheduled replacement of
carbon in lieu of monitoring for non-regenerative carbon
adsorption systems, the EPA maintains that the monitoring
frequency stated in the proposed rule remains appropriate for
owners or operators who elect to monitor.
6.12.4 Method 21
comp«=»r^• One commenter (A-90-19: IV-D-34) stated that
many of the requirements for the use of containers in
wastewater service are not reflected in any of the floor
determinations and are difficult and expensive to achieve.
The commenter (A-90-19: IV-D-34) expressed concern that the
proposed rule requires for existing equipment, which meets the
proposed definition of container, that all covers and openings
for each container be "designed for and operated without leaks
at the 500 ppmv level as determined by Method 21." The
commenter (A-90-19: IV-D-34) recommended that the EPA develop
work practice standards that focus on keeping containers
closed when in use rather than a complex monitoring and
replacement strategy. The commenter (A-90-19: IV-D-34)
provided alternate regulatory language which incorporated all
recommendations to the EPA.
Response; The floor for the control of wastewater
emissions from containers is no control, because the EPA has
determined that at proposal, covers, control devices, and
submerged fill pipes were not used by industry for containers.
Furthermore, the MACT floor for control of wastewater
emissions from any waste management unit or drain system is no
control. However, the Administrator determined that it was
appropriate to establish wastewater requirements above the
floor.
The inspection requirements for leaks in covers have been
changed in the final rule. Section 63.148 of the final rule
only requires an initial inspection using Method 21. Annual
visual inspections for visible, audible, or olfactory
indications of a leak are required in the final rule, instead
of annual inspections using Method 21 as proposed.
6-27
-------
One commenter (A-90-19: IV-D-33) stated that
the monitoring requirements for treatment processes in
proposed §63.143(b) are overly burdensome and should be
modified to allow for greater flexibility and to minimize
redundancy between the Act and the CWA monitoring requirements
on the same waste streams.
Another commenter (A-90-19: IV-D-33) stated that
facilities should be allowed to monitor surrogate parameters
or monitor less frequently if a different parameter or reduced
frequency is allowed by their CWA permit, which is issued
pursuant to NPDES or an industrial pretreatment program. The
commenter (A-90-19: IV-D-33) stated that the monitoring
frequency is established on a case-by-case basis, taking into
account such factors as toxicity, expected treatment
efficiency, demonstrated performance of the treatment process,
facility compliance history, sampling and analytical costs,
and the resulting burden on the regulator to review records
and process reports from the facility. The commenter
(A-90-19: IV-D-33) stated that such case-by-case factors are
relevant to wastewater monitoring for the HON.
Response; The EPA agrees with certain points raised by
the commenter and has reduced the monitoring requirements for
treatment processes. After determining that the performance
criteria in §63.145 were sufficient to ensure compliance with
the wastewater treatment requirements in the HON, the EPA
deleted the monthly monitoring requirements that were in
table 11 of subpart G of the proposed rule. The remaining
monitoring requirements in table 12 of subpart G of the final
rule (i.e., table 11 in the proposed rule) require continuous
monitoring for certain operating parameters associated with
the design steam stripper and biological treatment systems.
Comment; One commenter (A-90-19: IV-D-92) suggested
only regulating surface impoundments with emissions greater
than 500 ppmv aiove background, because this quantity of
emissions is al wed from openings. Another commenter
(A-90-19: IV-C -) suggested using a mass threshold or a
percentage of total facility emissions threshold for control
6-28
-------
of surface impoundments and individual drain systems to avoid
controlling systems with low emissions.
Response: In response to the commenter's suggestion not
to require emission controls on surface impoundments that emit
less than 500 ppmv above background, the EPA clarifies that
the 500 ppmv determination is a criterion for the inspection
provisions. It is not an allowable emission rate, but rather
an indication of whether a system has adequate emission
suppression. The EPA requires emission controls on all
surface impoundments managing wastewater streams that are
subject to regulation.
The purpose of the equipment standard is to ensure that
air emissions are suppressed. Emissions from wastewater are
directly proportional to the exposed surface area. For this
reason, modifying the regulatory requirements would result in
substantially higher emissions than the control requirements
of the proposed HON regulation. Therefore, the EPA has not
implemented this suggestion in the final HON regulation.
To provide greater flexibility, the EPA has added a
provision which allows an owner or operator to demonstrate
through a pressure test that the surface impoundment and
associated closed-vent system are under negative pressure.
This type of demonstration satisfies the monitoring
requirement so that the owner or operator is not also required
to perform Method 21. In addition, both the applicability
criteria (i.e., VOHAP concentration and flow rate) and the
1 Mg/yr sourcewide compliance option in §63.138(c)(5) and (6)
are intended to exempt from the control requirements of
§63.138 wastewater streams with low emissions relative to the
cost of control. Therefore, the final rule avoids the
unnecessary control of waste streams and wastewater streams
that have a low potential for emissions.
The EPA clarifies that owners and operators must comply
with §63.133 through §63.137 only when the wastewater
collection and treatment units regulated under these parts of
the HON regulation are used to receive, manage, or treat a
6-29
-------
Group 1 wastewater stream or a residual removed from a Group 1
wastewater stream.
comment; Two commenters (A-90-19: IV-D-32; IV-D-73)
recommended that fugitive emissions monitoring requirements
based on Method 21 be deleted from subpart G and suggested
that if the EPA must include fugitive emissions testing
requirements for wastewater management units, these sources
should be included in subpart H. One commenter (A-90-19:
IV-D-32) stated that the EPA has not performed an analysis of
the cost and environmental benefits associated with requiring
leak testing to be performed on sources of fugitive emissions.
Two commenters (A-90-19: IV-D-32; IV-D-102) stated that
Method 21 is inappropriate for VOHAP measurements because it
measures total VOC content, not just VOHAP1s, and it is
ineffective for measuring low levels of volatile organics.
Two commenters (A-90-19: IV-D-97; IV-D-102) recommended that
all references to the use of Method 21 for wastewater streams
be deleted from the HON and replaced by visual inspection
only. One commenter (A-90-19: IV-D-73) was unsure if
Method 21 was valid for detecting leaks from fixed roof
wastewater tanks and pointed out that no provisions are made
for repair if a leak is found in a fixed roof.'
One commenter (A-90-19: IV-D-31) claimed that the
ability to measure and repair small leaks less than 500 ppmv
is not practical. The commenter (A-90-19: IV-D-31) claimed
that Method 21 was originally intended to evaluate leaks of
pure compounds at levels of 10,000 ppmv. The commenter
(A-90-19: IV-D-31) asserted that wastewater tanks with low to
moderate concentrations may produce a vapor content much less
than 500 ppmv, and Method 21 leak detection testing would be
useless. The commenter (A-90-19: IV-D-31) supported
exempting tanks with low concentrations of wastewater from
Method 21 testing to avoid needless expenses.
Response: The EPA reviewed the option of consolidating
all fugitive emission testing in subpart H. However, due to
the structure of subpart H and to the different compliance
schedules for subparts G and H, incorporating the leak
6-30
-------
inspection requirements from subpart G into subpart H would
have generated additional confusion in the regulated
community. In particular, the leak inspection provisions
associated with wastewater management were not easily
incorporated into subpart H. The EPA agrees that the leak
inspection requirements which were located in separate
sections for each emission point in subpart G should be
condensed into a single section. Therefore, in the final
rule, the EPA incorporated all leak inspection provisions for
subpart G into §63.148.
In response to the technical comments about the use of
Method 21, the EPA asserts that the method was designed to
detect leaks from equipment. Method 21 is not used for
measuring emission rates. Many existing rules have
incorporated similar requirements. The EPA continues to
require at least the initial use of Method 21 for leak
detection followed by annual visual inspections for most waste
management units. The EPA has incorporated all provisions for
repairing any leaks detected by Method 21 in §63.148 of
subpart G.
The EPA points out that Method 21 is effective for
detecting concentrations of 500 ppmv VOC's in the air. For
example, Method 21 testing will indicate that 500 ppmv VOC's
is present in the air above an open wastewater tank when a
concentration as low as five percent VOC's is present in the
wastewater. For additional discussion about the capacity and
vapor pressure .thresholds for wastewater tanks that were
incorporated into the final rule, refer to section 4.1.9 of
this BID volume.
one commenter (A-90-23: IV-D-31) stated that
it was unclear whether or not the two conditions described in
subpart H as "unsafe to screen" and "inaccessible" for
Method 21 leak detection would apply to subpart G. The
commenter (A-90-19: IV-D-31) claimed that these exemptions
would eliminate wastewater tanks and wastewater tank roofs
where components cannot be reached safely.
6-31
-------
gesponse: The Method 21 requirements from the proposed
w.^tewater provisions have been moved to §63.148 of subpart G.
ithin §63.148 are provisions that are written to ensure that
equipment that is "unsafe to inspect" is exempt from the
initial Method 21 inspection requirements. There are also
provisions in §63.148 that are written to ensure that
equipment that is "difficult to inspect" is exempt from the
initial Method 21 inspection requirements. Equipment that is
"unsafe to inspect" or "difficult to inspect" is only subject
*••> annual visible, audible, and olfactory inspection
jquirements. In this case, "difficult to inspect"
encompasses any piece of equipment - at is inaccessible. The
Method 21 requirements in §63.148 >c subpart G apply to
wastewater tanks, as well as surface impoundments, containers,
individual drain systems, oil-water separators, and closed-
vent systems.
Comment; One commenter (A-90-19: IV-D-31) suggested
that Method 21 leak detection should not be required for fixed
roof wastewater tanks under a continuous negative pressure.
The commenter (A-90-19: IV-D-31) claimed that no leaks can
occur under these conditions and recommended adding a
measurement of static pressure to the Method 21 applicability
cr_ :eria for fixed roof tanks.
Response; In the final rule, §63.133(b) 4) of subpart G
exempts any fixed roof wastewater tank and closed-vent system
that is operated and maintained under negative pressure from
leak inspections using Method 21.
£7T1UH?lrt; One commenter (A-90-19: IV-D-31) claimed that
restricting the Method 21 calibration gas to a mixture of
methane in air limits Method 21 to the use of an instrument
with an FID or NDIR detector, because a PID will not respond
to methane. The comc^nter (A-90-19: IV-D-31) claimed that
Method 21 can be used with several reference gases for which
response factors of the affected HAP have been determined
and/or published and hat the PID has the optimum response for
some HAP's.
6-32
-------
Response; The EPA clarifies that Method 21 does not
restrict the calibration gas to a mixture of methane and air,
but rather requires an adjustment of the readings to a methane
basis. Therefore, PID may be used, but must be adjusted to a
methane basis. The reason that all must be adjusted to a
methane base is because having a single base makes
measurements from all instruments regardless of calibration
gas comparable. Refer to chapter 5.0 of BIO volume 2A for
additional discussion of the issue.
Comment; One commenter (A-90-19: IV-D-31) claimed that
using the predominant HAP in the wastewater stream to
determine the Method 21 response factor could cause the
screening values to be high. The commenter (A-90-19:
IV-D-31) also claimed that the predominant HAP in the
wastewater may not necessarily be the predominant HAP in the
vapor stream.
Response; The EPA agrees that the predominant HAP in the
wastewater may not be the predominant HAP in the vapor stream.
Therefore, in the final rule, the EPA no longer requires
response factor adjustments. For additional discussion of
actual monitoring requirements, refer to chapter 5.0 of BID
volume 2A.
Comment: One commenter (A-90-19: IV-D-64) requested
that the EPA clarify language in §63.133 through §63.138,
which states that a roof or cover "shall be designed and
operated without leaks as indicated by an instrument reading
of less than 500 ppm by volume..." The commenter (A-90-19:
IV-D-64) stated that the EPA must specify which points on the
roof or cover must be monitored because monitoring the entire
surface would be unreasonable.
Response; The leak inspection provisions from §63.133
through §63.138 of the proposed rule have been moved to
§63.148 in the final rule. In §63.148 of the final rule, the
leak inspection provisions and the Method 21 requirements for
all of subpart G are clarified. Section 63.148(c)(6) of the
final rule specifies which points on the roof or cover must be
monitored and includes "all potential leak interfaces".
6-33
-------
One commenter (A-90-19: IV-D-34) suggested
that the EPA should delete the annual monitoring requirement
in §63.133(b) (1) (ii) and replace it with a provision to repair
equipment if there is sensory evidence (visual, olfactory, or
audible) of a leak.
Response; The annual Method 21 inspection requirements
in §63.133 (b) (1) (ii) of the proposed rule have been deleted.
As described in §63.148 of subpart G and table 11 of subpart G
of the final rule, the owner or operator must conduct an
initial inspection using Method 21 and semi-annual visual
inspections for visible, audible, or olfactory indications of
leaks in fixed-roof tanks.
6.12.5 ?at Exchange Systems
Cor ant : Several commenters (A-90-19: IV-D-32; IV-D-33;
IV-D-34; IV-D-36; IV-D-53; IV-D-67 ; IV-D-110; IV-D-112)
(A-90-23: IV-D-4) stated that the sampling provisions in
§63. 102 (b) (2) (ii) seem to require sampling cooling water at
the entrance and exit of each heat exchanger system. One
commenter (A-90-19: IV-D-89) claimed that most heat exchange
systems are piped in parallel.
One commenter (A-90-19: IV-D-89) provided a figure to
help clarify where sample ports should be located, showing
sample ports at the cooling water supply and the cooling water
return .
Several commenters (A-90-19: IV-D-32; IV-D-33; IV-D-34;
IV-D-36; IV-D-53; IV-D-67; IV-D-110; IV-D-112) (A-90-23:
IV-D-4) disagreed with requiring sampling of each heat
exchanger and recommended that the EPA rewrite the provision
to require sampling at the entrance and exit of each cooling
tower system that services a unit.
One commenter (A-90-19: IV-D-38) claimed that cooling
towers are the only source of emissions in heat exchange
systems and further claimed that it should be specified in the
regulation that monitoring is required for the return water
anc not the individual heat exchangers. One commenter
(A-90-19: IV-D-50) claimed that it was unclear where cooLing
water samples are to be taken.
6-34
-------
Response: The EPA requires sampling at the entrance and
exit of each heat exchange system. A heat exchange system is
not a heat exchanger. The EPA has defined a heat exchange
system as any cooling tower system or once-through cooling
water system (e.g., river or pond water). Therefore, sampling
is not required at the entrance and exit of each heat
exchanger. Rather, sampling is required at the entrance and
exit of each cooling tower for recirculating systems, or the
points at which the cooling water enters and exits the once-
through cooling water system for nonrecirculating systems.
Sampling of both the cooling water supply and the cooling
water return is necessary in order to determine the emissions
from the cooling tower. Sampling only the return water would
not demonstrate when there is a concentration differential
across the tower, and would therefore not indicate when
compounds are volatilizing from a heat exchange system.
Comment; One commenter (A-90-19: IV-D-86) argued that
the sampling requirements for cooling towers are too
burdensome for multi-purpose batch operations due to the
variety of compounds in the cooling water.
Response; The monitoring requirements for heat exchange
systems do not require speciation of HAP's. Facilities can
monitor for speciated HAP, total HAP, total VOC concentration,
or TOG for semi-volatile HAP's. Therefore, a variety of
compounds in the cooling water will not overburden facilities
when complying with monitoring requirements.
rrnrnnqn-fr; one commenter (A-90-19: IV-D-38) suggested two
different options for monitoring heat exchange systems. The
commenter (A-90-19: IV-D-38) recommended speciation of HAP's
monthly for 6 months and then quarterly for the remainder of
two years and then using an average concentration determined
from this data to speciate any future leaks that are detected
using more conventional methods. The commenter (A-90-19:
IV-D-38) recommended using a monitoring frequency determined
by the facility according to historical needs.
Response; The EPA will allow monitoring of speciated
HAP, total HAP, total VOC concentration, or TOG for semi-
6-35
-------
volatile HAP's to detect leaks in a heat exchange system. The
monitoring requirements for heat exchange systems do not
include speciation of the inlet and outlet samples.
Conventional methods may indicate when the average
concentration in the cooling water increases, but conventional
methods do not provide any information on the magnitude of the
concentration differential across the cooling tower or the
magnitude of emissions from the cooling tower. Furthermore,
historical needs cannot indicate when a leak will occur in a
heat exchange system. As heat exchange system equipment
becomes older, it is more likely to develop a leak. A
facility with relatively new equipment will probably have had
few leaks, but as the facility becomes older, the equipment
may develop more leaks.
Comment: One commenter (A-90-19: IV-D-89) suggested
supplying a reference or guidance which clarifies the basis
for the methodology used to determine leaks in heat exchange
systems and requested that EPA specify which methods are
acceptable to determine HAP concentration in the cooling
water. One commenter (A-90-19: IV-D-38) presented a list of
test methods and devices that the commenter (A-90-19:
IV-D-38) claimed can detect a leak, determine its magnitude,
and provide characteristics of the contaminant.
Response; The EPA allows several methods to detect leaks
from cooling water, but has not provided a list in this BID
volume.
co^pent: One commenter (A-90-19: IV-D-89) favored
allowing 30 days from initial knowledge of a heat exchanger
leak until isolation, repair, or delay of repair is required,
because of sample turnaround time. One commenter (A-90-19:
IV-D-73) recommended that a 60-day repair period be provided
and an additional 60-day extension be allowed for repairing a
heat exchanger leak. The commenter (A-90-19: IV-D-73)
claimed that it takes several days to determine which heat
exchanger is leaking and that heat exchangers usually do not
have a backup so shutdown is therefore required.
6-36
-------
One commenter (A-90-19: IV-D-33) requested that the EPA
clarify at what point the 15 calendar days for repair of a
leaking heat exchanger begin. The commenter (A-90-19:
IV-D-33) recommended that the 15 days begin when the results
of any necessary analyses are known by the owner or operator
of the facility.
One commenter (A-90-19: IV-D-34) stated that the
proposed repair periods for heat exchanger leaks are
impractical. The commenter (A-90-19: IV-D-34) contended that
the 15 calendar days in which to repair a detected leak
specified in §63.102(b)(2)(v) should be extended to 90 days
because special parts may be needed and maintenance schedules
may require adjustment. The commenter (A-90-19: IV-D-34)
added that because the delay of repair provisions in
§63.102(b)(3) reference a process unit shutdown, the term
"process unit shutdown" should be defined in §63.101 rather
than §63.161. The commenter (A-90-19: IV-D-34) contended
that only planned process unit shutdowns and not emergency or
unplanned shutdowns should trigger the requirement for repair.
Response: Based on comments received by the EPA, the
amount of time that a facility has to repair a leak in a heat
exchange system has been extended from 15 days to 45 days.
The 45 days to repair a leak begins when the results of the
monitoring tests indicate that a leak is present (i.e., when a
1 ppm differential across a heat exchange system is detected).
A definition of process unit shutdown has been added to
§63.101. The EPA has elected to keep the definition of
process unit shutdown in §63.111 and §63.161 also. If a heat
exchanger cannot be repaired without a process unit shutdown,
a shutdown is required to repair the leak, unless the owner or
operator can show that a shutdown would cause more emissions
than the leak. Unplanned shutdowns are required for leaks in
a heat exchange system, because large quantities of emissions
can be released from an unrepaired leak in the system. For
example, an average-size cooling tower (15,000 gpm) with a
leak of only 1 ppm can emit almost 3 tons in one month if left
unrepaired.
6-37
-------
Extending the repair period for a leak in a heat exchange
system by 30 days will allow a sufficient amount of time for a
facility to determine which heat exchanger is leaking. The
extension also allows enough time to adjust maintenance
schedules and order special parts.
Comment; One commenter (A-90-19: IV-D-38) agreed with
the delay of repair provisions for cooling water systems.
Response; Based on comments received, the EPA has
extended the amount of time that a facility has to repair a
leak in a heat exchange system from 15 to 45 days. The EPA
has determined that 15 days is an insufficient amount of time
for a facility to repair a leak in all cases. In certain
cases, an owner or operator may have trouble identifying which
heat exchanger is leaking, or may have to adjust maintenance
schedules, or order special parts. Furthermore, a facility
must now shut down if the leak cannot be repaired in 45 days,
unless the owner or operator can demonstrate that the
emissions from shutdown are greater than the emissions from
the leak. This provision was added because the EPA has
determined that a significant amount of emissions can occur
from a cooling tower if the leak is left unrepaired
(Memorandum from Kristine Pelt, Radian, to Mary Tom Kissel1,
EPA/SDB, "Leaks from a Heat Exchange System," November 23,
1993) .
6.12.5.1 Cooling Tower Systems
Comment; Several commenters (A-90-19: IV-D-53; IV-D-73;
IV-D-38) claimed that a 1 percent or 1 ppm variation of TOG
levels in cooling water systems cannot be detected or
duplicated because of the low VOHAP concentrations typically
present in cooling water systems. Because of the inherent
uncertainty of analytical methods, two commenters (A-90-19:
IV-D-53; IV-D-110) recommended that the EPA use analytical
method performance data to determine when a concentration
increase indicates a leak. One commenter (A-90-19: IV-D-53)
claimed that even the best analytical methods have precisions
of about 9 or 10 percent and recommended dropping the
1 percent criterion to determine a leak from the rule. One
6-38
-------
commenter (A-90-19: IV-D-50) claimed that the definition of
leak does not have any basis. Two commenters (A-90-19:
IV-D-36) (A-90-20: IV-D-20) stated that the EPA should not
consider a change in a reading of one part per mill-ion or one
percent to be a cooling water system leak. The commenters
(A-90-19: IV-D-36) (A-90-20: IV-D-20) stated that if the
outlet stream had a low flow rate, a concentration of one part
per million or one percent would not be a concern, and thus,
the EPA should set action levels based on the size of the flow
exiting the tower.
Two commenters (A-90-19: IV-D-32; IV-D-54) stated that
the wastewater VOHAP concentration that is used to identify a
leak in a cooling water system should be based on the
potential to emit. Three commenters (A-90-19: IV-D-32;
IV-D-54; IV-D-110) contended that the proposed leak detection
action criteria in §63.102(f)(2)(iv) may be appropriate for
cooling systems using large volumes of water for heat
exchange, but are unnecessarily restrictive for smaller
cooling systems since the potential to emit significant
amounts of HAP's is proportionately smaller. One commenter
(A-90-19: IV-D-32) provided a table of recommended action
levels expressed as the concentration of total'VOHAP in the
wastewater which are dependent on water flow rate.
Response; A 1 ppm variation in concentration is the
lowest variation that can be measured. The EPA defines a leak
as a statistically significant difference of at least 1 ppm in
speciated HAP, total HAP, or total VOC concentration at the
95 percent confidence level. The 95 percent confidence level
allows for variation at low concentration levels. The
one-percent variation in total HAP levels has been eliminated
as a leak criterion.
Even for cooling towers with low flow rates, a 1 ppm
variation across the cooling tower can cause significant
emissions. For example, a 1 ppm variation across a cooling
tower with a flow rate of only 2,000 gallons per minute will
result in over 2 tons of emissions if left undetected for
6-39
-------
6 months. Therefore, it is necessary to monitor cooling
towers with low flow rates on a quarterly basis.
The EPA is allowing TOC as a monitoring parameter for
semi-volatile HAP's listed in Method 625, but not for volatile"
HAP's.
Comment; Two commenters (A-90-19: IV-D-38; IV-D-89)
claimed that the cooling water monitoring requirements for the
large list of HAP's will be expensive, costing approximately
$300 - $400 per sample analyzed. One commenter (A-90-19:
IV-D-89) suggested performing the cheaper Total Purgeable
Organic Carbon tests on the cooling water and only requiring a
sampling program if the return water carbon is over 10 percent
higher than the supply water at a 95 percent confidence limit.
One commenter (A-90-19: IV-D-38) suggested that each
facility be allowed to develop a site-specific monitoring
program for heat exchange systems. The commenter (A-90-19:
IV-D-38) suggested using a TOC test to determine the "normal"
level of organic material found in a cooling water system and
using this "baseline" to determine system changes. The
commenter (A-90-19: IV-D-38) also provided a list of
"conventional ways" to determine a leak, including an increase
in TOC, loss of heat transfer, oil sheen on the water surface,
etc. One commenter (A-90-19: IV-D-53) claimed that process
knowledge can be used to determine a heat exchanger leak.
Several commenters (A-90-19: IV-D-32; IV-D-54; IV-D-ll.2)
suggested that the EPA allow a surrogate parameter for routine
testing and require more extensive testing if the surrogate
parameter indicates a leak.
Response; The monitoring requirements for cooling towers
in §63.104 of subpart F have been changed to allow testing of
speciated HAP, total HAP, TOC for semi-volatile compounds, or
total VOC concentration. A leak will be indicated by a
statistically significant difference in speciated HAP, total
HAP, or total VOC concentration of 1 ppm at the 95 percent
confidence level. The one percent increase of total HAP
concentration as a criterion for a leak has been eliminated
from the final rule.
6-40
-------
Performing a TOG test to determine the "normal" level or
organic material does not guarantee that a leak will be
detected. If the TOG test is performed when a heat exchanger
is leaking, the "normal" level of organic material will be
elevated. Furthermore, determining the "normal" level of TOG
in cooling water does not provide any information on the
concentration differential across the cooling tower or the
quantity of emissions generated by the cooling tower.
Conventional ways of determining a leak or surrogate
parameters cannot predict the magnitude of the leak and do not
provide information on the concentration differential across
the cooling tower. Conventional methods or surrogate
parameters can help determine when a heat exchanger is
leaking. However, the EPA's definition of a leak in a heat
exchange system does not always coincide with a leak in a heat
exchanger, unless the leaking compounds volatilize in the
cooling tower.
The EPA is allowing the TOC test for only semi-volatile
HAP's listed in Method 625. The EPA is not allowing TOC for
volatile compounds as specified in Method 624 because too much
of the volatile HAP may be lost during the handling of the
sample. The method does not safeguard against emission losses
when transferring the sample. In contrast, methods such as
Methods 624 and 8020 require sealed caps and other sample
preserving techniques. Method 301 may be used to validate
other methods used to monitor volatile HAP's.
The EPA has no fundamental objection to using TOC as a
monitoring parameter, but it is not appropriate for volatile
HAP's. The EPA is allowing TOC as a monitoring parameter for
semi-volatile HAP's because such HAP's are less likely to
volatilize during sampling. Because the TOC test is less
costly than a total HAP or speciated HAP test, the EPA has
provided a more cost-effective method for owners or operators
with semi-volatile HAP's.
comi^nt• Several commenters (A-90-19: IV-D-32; IV-D-33;
IV-D-54; IV-D-112) disagreed with the requirement in §63.102
which requires testing for total VOHAP concentrations in
6-41
-------
cooling water. The commenters (A-90-19: IV-D-32; IV-D-33;
IV-D-54; IV-D-112) suggested that the EPA specify an action
level based on HAP's in table 9 as the basis for implementing
leak detection requirements.
One commenter (A-90-19: IV-D-73) concurred that
monitoring of cooling water should be limited to table 9 HAP's
because these HAP's will volatilize from water. One commenter
(A-90-19: IV-D-110) stated that the EPA should require
testing for total VOHAP concentration rather than total HAP
concentration because only table 9 HAP's are subject to the
wastewater provisions. One commenter (A-90-19: IV-D-53)
claimed that treatment chemicals and variation of intake water
quality could interfere with leak detection if the regulation
requires testing of total HAP's. One commenter (A-90-19:
IV-D-73) stated that cooling towers should only be monitored
for HAP's present in the unit(s) being serviced by the cooling
tower.
Response; The monitoring requirements for heat exchange
systems allow for sampling of speciated HAP, total HAP, total
VOC concentration, or TOG for semi-volatile HAP's. A leak is
detected in a recirculating cooling system if the influent
concentration to the cooling tower is at least 1 ppm higher
than the effluent concentration from the cooling tower.
Therefore, a leak is detected only if there are compounds
volatilizing from the cooling tower. Compounds that do not
readily volatilize from water (HAP's not listed on table 9)
will not cause a concentration differential across the cooling
tower. Therefore, the definition of leak is based on whether
or not the compounds in the cooling water are volatile, and
repair of leaks is only required when the compounds in the
cooling water are volatile.
Comrofnt1 One commenter (A-90-19: IV-D-34) suggested
that if the EPA is going to regulate new cooling tower
emissions, a design standard such as the ASME code for heat
exchange systems should be considered rather than a LDAR
standard for new heat exchangers.
6-42
-------
Response; A design standard such as the ASME code for
heat exchange systems does not guarantee that the heat
exchanger will not leak. ASME codes are written for design
and safety purposes. They ensure that a piece of equipment,
such as a heat exchanger, achieves the desired performance
level and operates safely. Furthermore, it is the actual
construction and not the construction code that will determine
if the heat exchanger will leak. For example, the heat
exchanger may be defective or the material of construction may
corrode due to old age or due to the types of chemicals being
processed. Therefore, a leak detection and repair program is
still necessary to ensure that a heat exchanger is not
leaking.
Comment: One commenter (A-90-19: IV-F-7.43 and
IV-D-117) claimed that the cooling water monitoring
requirements in §63.102(b)(2) provide a loophole which allows
large emissions of volatile HAP's and other VOC's. The
commenter (A-90-19: IV-F-7.43 and IV-D-117) claimed that lack
of maintenance occurs with cooling towers. The commenter
(A-90-19: IV-F-7.43 and IV-D-117) suggested requiring the
installation of continuous TOC monitoring devices on all
cooling water equipment in HAP service. The commenter
(A-90-19: IV-F-7.43 and IV-D-117) further suggested that if
the TOC reading of the cooling water reaches 15 ppm or
greater, then a sample of the cooling water should be
submitted for analysis. The commenter (A-90-19: IV-F-7.43
and IV-D-117) suggested that if TOC levels of 20 ppm and above
are reached, the piece of equipment should be taken out of
service and repaired as soon as possible. The commenter
(A-90-19: IV-D-117) suggested that a reading of 15 ppm or
greater should trigger periodic sampling of cooling tower
stacks.
Response; Monitoring of cooling tower influent and
effluent concentrations to detect leaks in a heat exchange
system is required monthly for the first 6 months and
quarterly thereafter. A leak in a heat exchange system is
defined as a difference in concentration of 1 ppm at a
6-43
-------
95 percent confidence level. The EPA has written these
requirements to prevent large emissions of HAP's and other
VOC's from occurring at the cooling tower. Lack of
maintenance will not occur with cooling towers, because if a
leak is detected, it must be repaired no later than 45 days
after it is detected. If the leak cannot be repaired without
process unit shutdown, the facility is required to shut the
process down, unless the owner or operator can demonstrate
that a shutdown will cause more emissions than the leak. The
facility also has the option to isolate the leaking process
equipment from HAP service until it is repaired.
Installations of continuous TOG monitoring devices would
be prohibitively expensive to install on all cooling water
equipment in HAP service, which would include every heat
exchanger. Furthermore, a reading of 15 or 20 ppm on a piece
of cooling water equipment does not indicate emissions from a
cooling tower. Emissions from a cooling tower are indicated
by a concentration differential across the cooling tower. For
example, if the influent and effluent concentrations of a
cooling tower are both 15 ppm, there are no emissions
occurring from the cooling tower. However, an influent
concentration of 15 ppm and an effluent concentration of
10 ppm indicate that emissions are occurring from a cooling
tower. Therefore, monitoring the influent and effluent of a
cooling tower is sufficient to determin- vhen leaks are
occurring.
C?fflffi?nt; Several commenters (A-90-19: IV-D-32; IV-D-53;
IV-D-112) supported §63.102(b)(4) which exempts from
monitoring non-contact cooling water systems which operate at
water pressures exceeding process fluid pressures.
Response: Non-contact heat exchange systems which
operate at water pressures exceeding process fluid pressures
were exempted from monitoring requirements because any leaks
would occur into the process fluid and not into the cooling
water.
6-44
-------
6.12.5.2 Once-Throuah Cooling Water
Comment: One commenter (A-90-19: IV-D-73) argued that
monitoring should only be required for recirculating cooling
water systems that are open to the atmosphere. Another
commenter (A-90-19: IV-D-34) suggested that requirements for
control of "once-through" cooling water systems should be
deleted. One commenter (A-90-19: IV-D-53) urged the EPA to
exempt once-through cooling water systems from the HON. One
commenter (A-90-19: IV-D-34) stated that once-through cooling
water is currently regulated under CWA regulations and the air
emissions are insignificant because the potential for HAP's to
enter the water is low and the driving force for
volatilization is very small. Several commenters (A-90-19:
IV-D-36; IV-D-53; IV-D-54; IV-D-73) (A-90-23: IV-D-20)
claimed that once-through cooling water systems are already
subject to NPDES wastewater discharge permit monitoring
requirements and should therefore not be subject to the HON.
One commenter (A-90-19: IV-D-34) provided data from
several NPDES permits, which document allowable discharge
limits ranging from 4 ppm with continuous monitoring in place
to 0.75 ppm with cooling water leak detection and repair as
part of best management practices.
Response; Once-through cooling water systems with
effluent discharge limits of less than 1 ppm are no longer
subject to the HON monitoring requirements for heat exchange
systems. A leak in a heat exchange system is defined as a
1 ppm differential in concentration across the heat exchange
system at a 95 percent confidence level. When a heat
exchanger in a once-through cooling water system is leaking,
the effluent concentration will be higher than the influent
concentration. Therefore, an effluent concentration limit of
less than 1 ppm guarantees that the variation in concentration
across a once-through heat exchange system is less than 1 ppm
if a heat exchanger is leaking. For once-through cooling
water systems with effluent discharge limits greater than or
equal to 1 ppm, it is impossible to guarantee that the
variation across the system is less than 1 ppm unless the
6-45
-------
influent concentration is monitored. Therefore, once-through
cooling water systems with effluent discharge limits greater
than or equal to 1 ppm are not exempt from the RON monitoring
requirements for heat exchange systems.
6-46
-------
7.0 RECORDKEEPING AND REPORTING
Two commenters (A-90-19: IV-D-70; IV-D-99)
requested that the Administrator be notified when a heat
exchanger is leaking, and the Administrator should have the
option to require a unit shutdown and repair before the next
scheduled shutdown. One commenter (A-90-19: IV-D-70)
provided a copy of text from an air permit which details
appropriate action levels and time required for repairs when
cooling towers are emitting butadiene.
Response; The heat exchange system provisions in
subpart F have been amended such that an owner or operator can
no longer invoke delay of repair for a leaking heat exchange
system if the repair is technically infeasible without a
process unit shutdown as previously stated in §63.102(b)(3)(i)
of subpart F of the proposed rule. In the final rule, a
process unit shutdown is required to repair a leak in a heat
exchange system, unless the owner or operator can demonstrate
that a process unit shutdown would cause greater emissions
than the emissions from the leaking heat exchange system until
the next planned shutdown. The EPA has determined that
significant emissions can occur from a leaking heat exchange
system between planned process unit shutdowns and has
determined that process unit shutdown is the appropriate
"action level" and time required for repairs (Memorandum from
Kristine Pelt, Radian, to Mary Tom Kissell, EPA/SDB, "Leaks
from a Heat Exchange System," November 23, 1993.).
qmnffler^; one commenter (A-90-19: IV-D-32) stated that
the recordkeeping and reporting requirements for wastewater
subject to the HON should not be required for RCRA-permitted
7-1
-------
treatment units because RCRA already specifies sufficient
monitoring, reporting, and recordkeeping.
Response; The EPA recognizes that recordkeeping and
reporting overlap exists between the HON and RCRA for RCRA-
permitted treatment units. In the final rule, the EPA has
addressed this issue by incorporating in §63.110(e)(2)(ii) of
subpart G an option for case-by-case determination of
requirements. This option allows owners or operators to work
with the Administrator to minimize any duplicative testing,
monitoring, recordkeeping, and reporting requirements.
c^ipe^t• one commenter (A-90-23: IV-D-20) stated that
the information required to document operating conditions
during the compliance test should be restricted to treatment
process information and should not include all process
information. The commenter (A-90-23: IV-D-20) stated that
§63.145(a)(4) should be altered to reflect such changes.
Response: The provisions in §63.145(a)(4) of subpart G
do not require an owner or operator to document all process
information. Rather, §63.145(a)(4) requires that an owner or
operator shall record all process information that is
necessary to document operating conditions during the test.
CoTHTO^TTfr; One commenter (A-90-19: IV-D-33) stated that
the EPA should explain why the recordkeeping requirements in
§63.102(b)(1) are necessary and what degree of detail is
required. The commenter (A-90-19: IV-D-33) stated that only
a brief explanation was included in the proposal preamble
(57 FR 62614). The commenter (A-90-19: IV-D-33) stated that
other regulations (e.g., NPDES and pretreatment requirements)
currently require paperwork for maintenance-related
wastewater, and thus, §63.102(b)(1) is not necessary and
should be deleted.
Response: The recordkeeping requirements for maintenance
wastewater have been moved from §63.102(b)(1) of subpart F of
the proposed rule to §63.105 of subpart F, entitled
maintenance wastewater requirements in the final rule. The
recordkeeping requirements for routine maintenance and
maintenance-turnaround wastewater are the same and these
7-2
-------
requirements will help ensure that procedures will be followed
to properly manage maintenance wastewater and control HAP
emission from maintenance wastewater to the atmosphere. The
level of detail for the recordkeeping requirements is not
specified in the rule in order to provide flexibility.
However, the owner or operator must provide a description of
maintenance activities which meets the requirements specified
in §63.105(b) of the final rule. The recordkeeping
requirements for NPDES and pretreatment permits are not
sufficient for compliance with the recordkeeping requirements
for maintenance wastewater regulated by the HON. These types
of permits only regulate the amount of organic material
present in the wastewater when it is discharged from the
facility. The maintenance requirements of the HON are written
to ensure the proper management of maintenance wastewater and
the control of HAP emissions to the atmosphere from
maintenance wastewater.
??TOT?*PTrt; One commenter (A-90-19: IV-D-33) stated that
the reporting requirements of §63.146 require submittal of
more information than is necessary to demonstrate compliance.
The commenter (A-90-19: IV-D-33) indicated that table 14a in
§63.146(a)(1) and table 14b in §63.146(a)(2) require almost
identical information for new facilities. The commenter
(A-90-19: IV-D-33) recommended that §63.146(a) be simplified
by eliminating subparagraph (a)(1) and table 14a, re-numbering
table 14b as 14, and re-numbering the subparagraphs. The
commenter (A-90-19: IV-D-33) contended that the same problem
arises with tables 15a and 15b in §63.146(b) and suggested the
same solution for deleting the redundancy.
Response: The commenter (A-90-19: IV-D-33) has
misinterpreted the reporting requirements for the
Implementation Plan and the Notification of Compliance Status
as listed in §63.146 of subpart G. The EPA did not intend for
identical information to be listed in tables 14a and 14b or in
tables 15a and 15b for new sources. The information in
tables 14a and 15a is to be submitted for table 8 compounds at
new sources. The information in tables 14b and 15b is to be
7-3
-------
submitted for table 9 compounds at new sources or for table 9
compounds at existing sources. The titles of tables 14a, 14b,
15a, and 15b and the text in §63.146 of subpart G have been
revised to clarify these reporting requirements.
Comment; One commenter (A-90-19: IV-D-33) suggested
that if an existing source, which elected to comply with the
process unit alternative of §63.138(d), completed table 16 in
§63.146(b)(3), then the facility should not also need to
complete table 15.
Response; The EPA agrees with the commenter (A-90-19:
IV-D-33). The provisions in §63.146(b)(3) of subpart G have
been clarified and state that if an owner or operator
completes table 16, then table 15b need not be completed.
Table 15a applies only to table 8 compounds for new sources.
epimner^; one commenter (A-90-19; IV-D-38) alleged that
the Administrative Authority should have the ability to
approve alternative heat exchanger and maintenance plans
subject to subparts G and H without having to publish notice
in the Federal Register.
Response; The EPA assumes that the commenter (A-90-19:
IV-D-38) is referring to the provisions in §63.102(c) of
subpart F in the proposed rule. These provisions stated that
the Director of the EPA's Office of Air Quality Planning and
Standards would determine when an alternative means of
compliance with subparts G or H is permitted and would publish
a notice to that effect in the Federal Register. The EPA
would like to point out that this authority rests with the
Administrator; thus the proposed rule contained an error.
Since the heat exchange system and maintenance wastewater
requirements are in subpart F, the provisions from §63.102(c)
of the proposed rule do not apply. In the final rule, the
general standards, heat exchange system, and maintenance
wastewater requirements have been moved to separate sections
in subpart F for clarity.
7-4
-------
8.0 WORDING OF THE PROVISIONS
One commenter (A-90-19: IV-D-34) stated that
the units for both MR and RMR in §63.145 must be the same in
order to compare the values.
Response ; The EPA has revised the equations and the
wording of the provisions at §63. 145 (g) and (h) , so that the
units for required mass removal (RMR) and actual mass removal
(MR) are consistent.
comment ! One commenter (A-90-19: IV-D-77) stated that
the EPA should correct §63. 133 (f) (2) of the proposed rule,
which addresses control equipment failures for wastewater
tanks, because the section incorrectly references
§§63.133(e) (2) (i)-(viii) when it should reference
§63.133(f) (2) (i)-(ix).
One commenter (A-90-19: IV-D-73) pointed out that the
preamble incorrectly references table 9 HAP's in §63.138 when
the table is actually in §63.131.
One commenter (A-90-19: IV-D-33) stated that the
reference in §63. 133 (a) should be changed from "(c)" to "(b)."
One commenter (A-90-19: IV-D-87) noted that
§63.138(d) (1) (i) is referenced in §63.145(b) (1) and that this
section does not exist.
One commenter (A-90-19: IV-D-85) stated that
§63.112(c) (1) (ii) should not refer to §63.132 (d) (4) but should
refer to §63. 138 (d) (4) . The commenter (A-90-19: IV-D-85)
pointed out that the way the proposed provisions were written,
sources are exempted from installing the RCT if they meet
certain monitoring and recordkeeping requirements.
Response ; The EPA agrees with the commenters that the
references were incorrect in the proposed rule. The
8-1
-------
references in the final rule have been corrected. However,
some paragraphs have been renumbered in the final rule, and,
therefore, the cross-references may have changed.
rngnt^p-h• one commenter (A-90-19: IV-D-73) suggested
deleting the third sentence in the definition of individual
drain system, because the sentence presents a design
requirement.
Response; The commenter is correct, and this change, has
been incorporated.
Comment; One commenter (A-90-19: IV-D-64) stated that
in proposed §63.138(g)(3), the 99 percent destruction should
be of total VOHAP, rather than HAP.
Response; The wastewater provisions in §63.138(g)(3) of
the proposed rule are found in §63.138(h)(3) of the final
rule. The EPA disagrees with the commenter (A-90-19:
IV-D-64) that the provisions in §63.138(h)(3) should refer to
99 percent destruction of VOHAP. These provisions refer to a
99 percent destruction of the total HAP mass flow rate. When
referring to the mass flow rate, the EPA refers to HAP. The
term "VOHAP" is used when referring to the concentration used
to determine applicability and the concentration used for
enforcement.
Comment; One commenter \-90-19: IV-D-64) stated that
"enclosed combustion device" is an undefined term.
Response; The EPA is not adding a definition of
"unenclosed combustion device" to the final rule because the
only type of combustion device that is not specifically
enclosed is a flare. In the final rule, requirements for
operating flares associated with the control of HAP emissions
from wastewater are located in §63.139 of subpart G.
Comment; One commenter (A-90-23: IV-D-20) stated that
the units (i.e., kg/yr and kg/hr) in §63.145(h) and (i) should
be expressed consistently in kg/hr.
Response; The EPA agrees with the commenter (A-90-19:
IV-D-20) that the units in §§63.145(h) and (i) should be
expressed consistently. Therefore, the mass flow rate units
8-2
-------
in §§63.145(h) and (i) have been changed and are all expressed
as kg/hr.
rgmpep-t;; one commenter (A-90-19: IV-D-33) stated that
the "and" after §63.132(d)(2) should be moved to after
§63.132(d)(4).
Response; The EPA agrees with the commenter (A-90-19:
IV-D-33) and the "and" after §63.132(d)(2) of the final rule
has been removed. However, an "and" was not placed after
§63.132(d)(4) of the final rule because paragraphs (1) through
(5) in §63.132(d) of the final rule are all independent
sentences.
Comment; One commenter (A-90-19: IV-D-33) stated that
the definition of "recovery device" appears in both §63.101
and §63.111 and should be moved from §63.101, which has a more
concise definition, to §63.111, which would eliminate the-need
for the definition in §63.101.
Response; The definition of "recovery device" has been
made consistent in §63.101 of subpart F and §63.111 of
subpart G. However, the EPA has decided to leave the
definition in §63.101 of subpart F and §63.111 of subpart G,
because the term is used frequently in both subparts and is
referred to by other definitions in both subparts.
Comment: One commenter (A-90-19: IV-D-33) stated that
the definitions of "closed-vent system," "control device,"
"process unit," and "process unit shutdown" should remain in
both §63.111 and §63.161, but the definitions should be made
consistent or the same, if possible.
Response; The EPA agrees with the commenter (A-90-19:
IV-D-33) that the definitions of "closed-vent system",
"control device", "process unit", and "process unit shutdown"
should remain in both §63.111 of subpart G and §63.161 of
subpart H. The definitions have been made consistent when
possible.
Comment; One commenter (A-90-19: IV-D-33) stated that
English units should be placed in parentheses after the metric
units throughout the HON in order to avoid confusion in
8-3
-------
converting metric units to English units and to be consistent
with the preamble.
Response; The regulation specifies only metric units,
because the EPA enforces standards based on the metric system.
Previous NESHAP and NSPS are based on metric units. Adding
English units would create confusion about which value is
enforceable due to rounding differences between the two
values .
Comment : One commenter (A-90-19: IV-D-33) stated that
the EPA should add "regulated" prior to "wastewater streams"
in §§63.131(a) (5) , (a) (6) , and (a) (7) .
Response: The EPA disagrees with the commenter (A--90-19:
IV-D-33) that the word "regulated" should be added prior to
the words "wastewater streams" in §63. 131 (g), (h) , and (i) in
the final rule. These paragraphs refer to figures 5, 6, and 7
which show the control options for those wastewater streams
subject to the. control requirements of the wastewater
provisions of the HON. These control options include
treatment of Group 1 wastewater streams, a combination of
Group 1 and Group 2 wastewater streams, or treatment of all
Group 1 and Group 2 wastewater streams (as required by the
process unit alternative illustrated in figure 8) . Control of
wastewater streams is required only if Group 1 wastewat«ir
streams are present at the facility. However, Group 2
wastewater streams are also "regulated" by the wastewater
provisions of the HON, although they do not require control.
The EPA is concerned that the change suggested by the
commenter could cause confusion if the term "regulated" were
misinterpreted to mean only Group 1 wastewater streams.
compefiti ; one commenter (A-90-19: IV-D-33) stated that
the definition of oil-water separator or organic-water
separator needs the word "equipment" added to the end.
Response ; The EPA agrees with the commenter (A-90-19:
IV-D-33) , and the word "equipment" has been added to the end
of tha definition of oil-water separator.
Two commenters (A-90-19: IV-G-10) (A-90-23:
IV-G-5) suggested that the EPA clarify the definition of
8-4
-------
wastewater in figure 2 of §63.131 by moving the term "flow
rate < 0.02 liter per minute" to the decision box containing
the term "<5 ppmw."
Response; The term "flow rate <0.02 liter per minute"
cannot be moved to the decision box containing the term
"<5 ppmw" without changing the meaning of the flow diagram and
the applicability criterion. If the term "flow rate
<0.02 liter per minute" is moved from the decision box
containing the term "concentration < 10,000 ppmw," wastewater
streams will be exempt from the HON wastewater provisions
simply by having a concentration less than 10,000 ppmw.
Wastewater streams having concentrations between 5 and
10,000 ppm are only exempt from the HON wastewater provisions
if their flow rate is less than 0.02 £pm. All wastewater
streams with concentrations less than 5 ppmw are exempt from
the HON wastewater provisions.
8-5
-------
-------
TECHNICAL REPORT DATA
'Please read Instruction* on ilic remrse be lore co'nntennz;
1 REPORT NO
EPA-453/R-94-003b
A. TITLE ANOSU8TITLE
H«i«rdoui Air Pollutant Emmions lra<* Pro
Ch«nucal Manufacturing. Industry — aackgroun
VoluM 2Bi CooMnti on Wa»t«wat«r
7 AUTHOR(S)
9 PERFORMING ORGANIZATION NAME AC
Office of Air Quality PI am
U.S. Environmental Protect
Research Triangle Park, Noi
2. 3 RECIPIENT S ACCESSION NO
5. REPORT DATE
=••• Unit* in th« Synthetic Organic March 1994
6. PERFORMING ORGANIZATION CODE
3. PERFORMING ORGANIZATION REPORT NO.
jo ADDRESS 10. PROGRAM ELEMENT NO
ling and Standards
Lon Agency 11 CONTRACT/GRANT NO
rth Carolina 27711 68010117
12. SPONSORING AGENCY MAME AND ADDRESS 13. TYPE OF REPORT AND PERIOD COVERED
Director, Office of Air Quality Planning and Standards
Office of Air and Radiation 14. SPONSORING AGENCY CODE
U.S. Environmental Protection Agency EPA/200/04
Research Triangle Park, North Carolina 27711
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A final rule for the regul
pollutants (HAP' a) from ch
manufacturing industry (SO
sections 112, 114, 116, an
emission standards were pr
(57 FR 62608). Public he a
published in the Federal B
volume of the background i
presents the agency's res;
17.
a. DESCRIPTORS
Air pollution
Pollution control
SOCMI
Hazardous air pollutant
National impacts
18. DISTRIBUTION STATEMENT
ation of emissions of organic hazardous air
emical processes of the synthetic organic chemical
•CMI) is being promulgated under the authority of
d 301 of the Clean Air Act, as amended in 1990. The
oposed in the Federal Reaister on December 31. 1992
rings were held. A supplemental notice was
eoister on October 15, 1993 (58 FR 53478). This
nformation document summarizes all comments and
onses on wastewater operations.
KEY WORDS AND DOCUMENT ANALYSIS
b.lDENTIPIERS/OPEN ENDED TERMS C. COSATI Fieid/GfOUp
Air pollution control
19. SECURITY CLASS /Tliti Report) 21 NO. OP PAGES
239
20. SECURITY CLASS , Tins page I 22. PRICE
UNCLASSIFIED
EPA Farm 2220-1 (R«v. 4-771 PREVIOUS EDITION 15 OBSOLETE
------- |