# Pharmaceutical Manufacturing Category Effluent Limitations Guidelines, Pretreatment Standards, and New Source Performance Standards; Final Rule

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/fr%3A98-21027

## Record

- **Collection:** Federal Register
- **Document type:** Rule
- **Published:** September 21, 1998
- **Citation:** 63 FR 50388

## Text

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Parts 136 and 439

[FRL-6135-7]
RIN 2040-AA13

Pharmaceutical Manufacturing Category Effluent Limitations
Guidelines, Pretreatment Standards, and New Source Performance
Standards; Final Rule

AGENCY: Environmental Protection Agency.

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: This final regulation limits the discharge of pollutants into
navigable waters of the United States and into publicly owned treatment
works (POTWs) by existing and new pharmaceutical manufacturing
facilities. This regulation revises limitations and standards for four
subcategories of the pharmaceutical manufacturing Point Source
Category: Subcategory A (Fermentation), Subcategory B (Extraction),
Subcategory C (Chemical Synthesis): and Subcategory D (Mixing,
Compounding, and Formulating); and reformats and clarifies language
without revision to certain specified provisions of these four
subcategories and a fifth subcategory: Subcategory E (Research). This
regulation establishes effluent limitations guidelines and standards
under the Clean Water Act including ``best conventional pollutant
control technology (BCT) and ``best available technology economically
achievable (BAT)'' for existing direct dischargers, ``new source
performance standards (NSPS)'' for new direct dischargers and
pretreatment standards for existing and new indirect dischargers (PSES
and PSNS). This regulation also amends and clarifies some of the
limitations based on ``best practicable control technology (BPT)'' for
pharmaceutical manufacturing facilities and establishes analytical
methods for certain organic pollutants contained in this regulation.
EPA is today also publishing final Maximum Available Control Technology
(MACT) standards under the Clean Air Act (CAA) for the pharmaceutical
manufacturing industry elsewhere in today's Federal Register. The MACT
standards final rule will control emissions of hazardous air pollutants
(HAPs) from pharmaceutical manufacturing emission sources including
wastewater collection and treatment systems. The Offices of Water and
Air and Radiation have coordinated the development of these regulations
and have used a common technology basis in developing limitations and
standards for the volatile organic compounds (VOCs).
The final MACT standards and effluent limitations guidelines and
standards rules will benefit the environment by removing a total of
85.4 million pounds per year of conventional, nonconventional and toxic
(priority) pollutants from water discharges. The effluent limitations
guidelines and standards portion of those removals is 13.9 million
pounds per year of nonconventional and 16.0 million pounds per year of
organic pollutants including VOCs.

DATES: This regulation shall become effective November 20, 1998. The
incorportion by reference of certain publications listed in Part 136 is
approved by the Director of the Federal Register as of November 20,
1998.

ADDRESSES: For additional technical information write to Dr. Frank H.
Hund, Engineering and Analysis Division (4303), U.S. EPA, East Tower,
401 M Street SW, Washington, D.C. 20460 or send E-mail to:
[email protected] or call at (202) 260-7182. For additional
economic information contact Mr. William Anderson at the address above
or by calling (202) 260-5131 or send E-mail to:
[email protected].
The complete record (excluding confidential business information)
for this Clean Water Act rulemaking is available for review at EPA's
Water Docket, Room EB57; 401 M Street, SW, Washington, DC 20460. For
access to Docket materials, call (202) 260-3027 between 9 a.m. and 3:30
p.m. for an appointment. The EPA public information regulation (40 CFR
part 2) provides that a reasonable fee may be charged for copying.
The Technical Development Document and Economic Impact Analysis
supporting today's final water rule may be obtained by writing to the
EPA Office of Water Resource Center (RC-4100), 401 M Street SW.,
Washington, DC 20460, or calling (202) 260-7786.

FOR FURTHER INFORMATION CONTACT: For additional technical information
call Dr. Frank H. Hund at (202) 260-7182. For additional information on
the economic impact analyses contact Mr. William Anderson at (202) 260-
5131.
SUPPLEMENTARY INFORMATION:

Judicial Review

In accordance with 40 CFR 23.2, the rule will be considered
promulgated for purposes of judicial review at 1:00 p.m. Eastern time
on October 5, 1998. Under section 509(b)(1) of the Act, judicial review
of this regulation can be obtained only by filing a petition for review
in the United States Court of Appeals within 120 days after the
regulation is considered promulgated for purposes of judicial review.
Under section 509 (b)(2) of the Act, the requirements in this
regulation may not be challenged later in civil or criminal proceedings
brought by EPA to enforce these requirements.

Regulated Entities

Entities potentially regulated by this action include:

------------------------------------------------------------------------
Category Examples of regulated entities
------------------------------------------------------------------------
Industry.......................... Facilities that generate process
wastewater from the manufacture of
pharmaceutical products and/or
pharmaceutical intermediates by
fermentation, extraction, chemical
synthesis and/or mixing,
compounding and formulating.
------------------------------------------------------------------------

This table is not intended to be exhaustive, but rather provides a
guide for readers regarding entities likely to be regulated by this
action. This table lists the types of entities that EPA is now aware
could potentially be regulated by this action. Other types of entities
not listed in the table could also be regulated. To determine whether
your facility is regulated by this action, you should carefully examine
the applicability criteria in Secs. 439.1, 439.10, 439.20, 439.30,
439.40 and 439.50 of this final rule. If you have questions regarding
the applicability of this action to a particular entity, consult the
technical information person listed in the preceding FOR FURTHER
INFORMATION CONTACT section.

Compliance Dates

The compliance date for PSES is as soon as possible, but no later
than September 21, 2001. The compliance dates for NSPS and PSNS are the
dates the new source commences discharging.

[[Page 50389]]

Deadlines for compliance with BPT, BCT, and BAT are established in the
National Pollutant Discharge Elimination System (NPDES) permits.

Organization of This Document

I. Legal Authority
II. Background
A. Clean Water Act
1. Best Practicable Control Technology Currently Available (BPT)
2. Best Available Technology Economically Achievable (BAT)
(Section 304(b)(2) of the Act)
3. Best Conventional Pollutant Control Technology (BCT) (Section
304(b)(4) of the Act)
4. New Source Performance Standards (NSPS) (Section 306 of the
Act)
5. Pretreatment Standards for Existing Sources (PSES) (Section
307(b) of the Act)
6. Pretreatment Standards for New Sources (PSNS) (Section 307(b)
of the Act)
B. Section 304(m) Requirements and the Pollution Prevention Act
C. Updated Profile of the Industry
D. Existing and Proposed Rules
1. Clean Water Act Proposal
2. Clean Air Act Proposal
3. Clean Water Act Federal Register Notice of Availability
E. Discussion of Final Clean Air Act Rule Published Elsewhere in
Today's Federal Register
F. Relationship Between the MACT and CWA Rules
G. Final Clean Water Act Effluent Limitations Guidelines and
Standards Rule
III. Summary of Most Significant Changes to Water Rules From
Proposal
A. Limitations and Standards for Volatile Organic Compounds
B. Change in BAT Technology Basis for Organic Pollutants
C. BPT and BAT/BCT Limitation Changes
D. Pollutant Selection
IV. The Final Clean Water Act Regulation
A. Applicability and Scope of the Final Rule
B. Options Selection
C. Best Practicable Control Technology Currently Available (BPT)
D. Best Available Technology Economically Achievable
E. Pretreatment Standards for Existing Sources (PSES)
F. New Source Performance Standards (NSPS)
G. Pretreatment Standards for New Sources (PSNS)
H. Best Conventional Pollutant Control Technology (BCT)
V. Assessment of Costs and Impacts for the Final Pharmaceutical
Regulations
A. Introduction
B. Summary of Economic Analysis Methodology and Data
C. Changes to the Economic Analysis Since Proposal
D. Estimated Economic Impacts
1. Costs of Compliance
2. Economic Impacts on Facilities
3. Economic Impacts on Firms
4. Impacts on Output and Employment
5. Other Secondary Impacts
6. Impacts on New Sources
E. Regulatory Flexibility Analysis
F. Cost-Benefit Analysis
G Cost-Effectiveness Analysis
VI. Environmental Benefits
VII. Non-Water Quality Environmental Impacts
A. Air Pollution
B. Solid Waste
C. Energy Requirements
VIII. Regulatory Implementation
A. Implementation of the Limitations and Standards
B. Upset and Bypass Provisions
C. Variances and Modifications
1. Fundamentally Different Factors Variances
2. Removal Credits
D. Analytical Methods
IX. Regulatory Assessment Requirements
A. Executive Order 12866
B. Regulatory Flexibility Act and the Small Business Regulatory
Enforcement Fairness Act of 1996 (SBREFA)
C. Submission to Congress and the General Accounting Office
D. Paperwork Reduction Act
E. Unfunded Mandates Reform Act
F. Executive Order 12875 Enhancing Intergovernmental Partnership
G. National Technology Transfer and Advancement Act
H. Executive Order 13045 and Protecting Children's Health
X. Summary of Public Participation
A. Summary of Proposal Comments and Responses
B. Summary of Notice of Availability Comments and Responses
Appendix A to the Preamble--List of Abbreviations, Acronyms,
Definitions and Other Terms Used in This Document

I. Legal Authority

This final regulation establishes effluent limitations guidelines
and standards of performance and analytical methods for the
pharmaceutical manufacturing point source category under the
authorities of sections 301, 304, 306, 307, 308, 402 and 501 of the
Clean Water Act (``the Act''), 33 U.S.C. 1311, 1314, 1316, 1317, 1318,
1342 and 1361.

II. Background

A. Clean Water Act

The Federal Water Pollution Control Act Amendments of 1972
established a comprehensive program to ``restore and maintain the
chemical, physical, and biological integrity of the Nation's waters,''
(section 101(a)). To implement the Act, EPA is to issue effluent
limitations guidelines, pretreatment standards and new source
performance standards for industrial dischargers.
These guidelines and standards are summarized briefly below:
1. Best Practicable Control Technology Currently Available (BPT)
(Section 304(b)(1) of the Act)
BPT effluent limitations apply to all discharges from existing
direct dischargers. BPT effluent limitations guidelines are generally
based on the average of the best existing performance by plants of
various sizes, ages, and unit processes within the category or
subcategory for control of pollutants.
In establishing BPT effluent limitations guidelines, EPA considers
the total cost of achieving effluent reductions in relation to the
effluent reduction benefits, the age of equipment and facilities
involved, the processes employed, process changes required, engineering
aspects of the control technologies, non-water quality environmental
impacts (including energy requirements) and other factors as the EPA
Administrator deems appropriate (Section 304(b)(1)(B) of the Act). The
Agency considers the category or subcategory-wide cost of applying the
technology in relation to the effluent reduction benefits. Where
existing performance is uniformly inadequate within a category or
subcategory, BPT may be transferred from a different subcategory or
category.
2. Best Available Technology Economically Achievable (BAT) (Section
304(b)(2) of the Act)
In general, BAT effluent limitations represent the best existing
economically achievable performance of plants in the industrial
subcategory or category, based upon available technology. The Act
establishes BAT as the principal national means of controlling the
direct discharge of toxic and nonconventional pollutants to navigable
waters. The factors considered in assessing BAT include the age of
equipment and facilities involved, the process employed, potential
process changes, and non-water quality environmental impacts (including
energy requirements) (Section 304(b)(2)(B)). The Agency retains
considerable discretion in assigning the weight to be accorded these
factors. As with BPT, where existing performance is uniformly
inadequate within a category or subcategory, BAT may be transferred
from a different subcategory or category. BAT may include process
changes or internal controls, even when these technologies are not
common industry practice.

[[Page 50390]]

3. Best Conventional Pollutant Control Technology (BCT) (Section
304(b)(4) of the Act)
The 1977 Amendments to the Act established BCT for discharges of
conventional pollutants from existing industrial point sources. Section
304(a)(4) designated the following as conventional pollutants:
Biochemical oxygen demanding pollutants (BOD5), total
suspended solids (TSS), fecal coliform, pH, and any additional
pollutants defined by the Administrator as conventional. The
Administrator designated oil and grease as an additional conventional
pollutant on July 30, 1979 (44 FR 44501).
BCT is not an additional limitation, but replaces BAT for the
control of conventional pollutants. In addition to other factors
specified in Section 304(b)(4)(B), the Act requires that BCT
limitations be established in light of a two part ``cost-
reasonableness'' test. American Paper Institute v. EPA, 660 F.2d 954
(4th Cir. 1981). EPA's current methodology for the general development
of BCT limitations was issued in 1986 (51 FR 24974; July 9, 1986).
4. New Source Performance Standards (NSPS) (Section 306 of the Act)
NSPS are based on the best available demonstrated control
technology. New plants have the opportunity to install the best and
most efficient production processes and wastewater treatment
technologies. As a result, NSPS should represent the most stringent
numerical values attainable through the application of the best
available control technology for all pollutants (e.g., conventional,
nonconventional, and toxic pollutants). In establishing NSPS, EPA is
directed to take into consideration the cost of achieving the effluent
reduction and any non-water quality environmental impacts and energy
requirements.
5. Pretreatment Standards for Existing Sources (PSES) (Section 307(b)
of the Act)
PSES are designed to prevent the discharge of pollutants that pass
through, interfere with, or are otherwise incompatible with the
operation of publicly owned treatment works (POTWs). The Act authorizes
EPA to establish pretreatment standards for pollutants that pass
through POTWs or interfere with POTWs' treatment processes or sludge
disposal methods. The legislative history of the 1977 Act indicates
that pretreatment standards are to be technology-based and analogous to
the BAT effluent limitations guidelines for removal of toxic
pollutants. For the purpose of determining whether to promulgate
national category-wide pretreatment standards, EPA generally determines
that there is pass through of a pollutant and thus a need for
categorical standards if the nation-wide average percent removal of a
pollutant removed by well-operated POTWs achieving secondary treatment
is less than the percent removed by the BAT model treatment system.
The General Pretreatment Regulations, which set forth the framework
for the implementation of categorical pretreatment standards, are found
at 40 CFR Part 403. (Those regulations contain a definition of pass
through that addresses localized rather than national instances of pass
through and does not use the percent removal comparison test described
above. See 52 FR 1586, January 14, 1987.)
6. Pretreatment Standards for New Sources (PSNS) (Section 307(b) of the
Act)
Like PSES, PSNS are designed to prevent the discharges of
pollutants that pass through, interfere with, or are otherwise
incompatible with the operation of POTWs. PSNS are to be issued at the
same time as NSPS. New indirect dischargers, like new direct
dischargers, have the opportunity to incorporate into their plants the
best available demonstrated technologies. The Agency considers the same
factors in promulgating PSNS as it considers in promulgating NSPS.

B. Section 304(m) Requirements and the Pollution Prevention Act

Section 304(m) of the Clean Water Act (33 U.S.C. 1314(m)), added by
the Water Quality Act of 1987, requires EPA to establish schedules for
(i) reviewing and revising existing effluent limitations guidelines and
standards (``effluent guidelines''), and (ii) promulgating new effluent
guidelines. On January 2, 1990, EPA published an Effluent Guidelines
Plan (55 FR 80), in which schedules were established for developing new
and revised effluent guidelines for several industry categories. One of
the industries for which the Agency established a schedule was the
Pharmaceutical Manufacturing Point Source Category.
Natural Resources Defense Council, Inc. (NRDC) and Public Citizen,
Inc., challenged the Effluent Guidelines Plan in a suit filed in U.S.
District Court for the District of Columbia (NRDC et al v. Reilly, Civ.
No. 89-2980). The plaintiffs charged that EPA's plan did not meet the
requirements of sec. 304(m). A Consent Decree in this litigation was
entered by the Court on January 31, 1992. The terms of the Consent
Decree are reflected in the Effluent Guidelines Plan published on
September 8, 1992 (57 FR 41000). This plan, as modified, required,
among other things, that EPA propose effluent guidelines for the
pharmaceutical manufacturing category by February, 1995 and take final
action on these effluent guidelines by April, 1998. Recently EPA filed
an unopposed motion requesting an extension of time until July 30, 1998
for the Administrator to sign the final rule.
The Pollution Prevention Act of 1990 (PPA) (42 U.S.C. 13101 et
seq., Pub. L. 101-508, November 5, 1990) ``declares it to be the
national policy of the United States that pollution should be prevented
or reduced whenever feasible; pollution that cannot be prevented should
be recycled in an environmentally safe manner, whenever feasible;
pollution that cannot be prevented or recycled should be treated in an
environmentally safe manner whenever feasible; and disposal or release
into the environment should be employed only as a last resort...''
(Sec. 6602; 42 U.S.C. 13101(b). In short, preventing pollution before
it is created is preferable to trying to manage, treat or dispose of it
after it is created. This effluent guideline was reviewed for its
incorporation of pollution prevention as part of this Agency effort.
According to the PPA, source reduction reduces the generation and
release of hazardous substances, pollutants, wastes, contaminants or
residuals at the source, usually within a process. The term source
reduction ``include[s] equipment or technology modifications, process
or procedure modifications, reformulation or redesign of products,
substitution of raw materials, and improvements in housekeeping,
maintenance, training, or inventory control.'' The term ``source
reduction'' does not include any practice which alters the physical,
chemical, or biological characteristics or the volume of a hazardous
substance, pollutant, or contaminant through a process or activity
which itself is not integral to or necessary for the production of a
product or the providing of a service.'' 42 U.S.C. 13102(5) In effect,
source reduction means reducing the amount of a pollutant that enters a
waste stream or that is otherwise released into the environment prior
to out-of-process recycling, treatment, or disposal.
The PPA directs the Agency to, among other things, ``review
regulations of the Agency prior and subsequent to their proposal to
determine their effect on

[[Page 50391]]

source reduction'' (Sec. 6604; 42 U.S.C. 13103(b)(2). This directive
led the Agency to implement a pilot project called the Source Reduction
Review Project that would facilitate the integration of source
reduction in the Agency's regulations, including the technology-based
effluent guidelines and standards.
In the preamble to the proposed regulations, EPA discussed the
possible pollution prevention alternatives available in pharmaceutical
manufacturing. At that time, EPA indicated that pollution prevention
opportunities were limited in the active ingredient manufacturing
subcategories (namely, fermentation, natural extraction and chemical
synthesis) but the use of water-based coatings in the formulation
subcategory operations was a viable pollution prevention approach which
eliminates the need for solvents in tablet coating operations. This
approach may only be applicable to some and not most tablet coating
operations, however. Since the proposal, EPA has received two
suggestions for incorporating pollution prevention into the final
regulations which were discussed in the August 8, 1997 Notice of
Availability at 62 FR 42720. One suggestion presented to the Agency was
that Subcategories B and D dischargers that incorporate best management
practices (BMPs), which reduce their discharge of any of the regulated
pollutants should not have to monitor for the specific regulated
pollutants, and possibly only monitor for the conventional pollutants
and COD. This pollution prevention approach is similar to the one
adopted in the Pesticide Formulators, Packagers and Repackagers (PFPR)
final regulation which was published in the Federal Register on
November 6, 1996 at 61 FR 57518. (It should be noted that PFPR
facilities that use the promulgated pollution prevention option have to
assess their wastewater and may be required to treat wastewater prior
to discharge.) EPA evaluated this suggestion and decided that since EPA
is not promulgating BAT limitations for specific organic pollutants,
this pollution prevention suggestion was not relevant to compliance by
subcategory B and D direct dischargers with final BAT limitations. For
PSES, EPA believes the suggestion may be workable for indirect
dischargers, since standards for specific organic pollutants are
contained in the final rule; however, no information was submitted to
identify the pollution prevention practices that would be incorporated
into the rule, and EPA has been unable to identify any.
Another pollution prevention approach suggested to EPA was that
Subcategories A and C facilities that can demonstrate a reduction in
the use of a regulated pollutant and resultant lowered air emissions or
water discharges should receive a higher effluent discharge limitation.
As suggested, the higher effluent discharge limitation would be
directly proportional to the amount of reduction achieved in the use of
the regulated pollutant. Along with this suggestion, the commenters
provided examples of how this pollution prevention suggestion could
work in individual instances.
In evaluating this suggestion including the examples provided, EPA
was concerned about the amount and type of process information that
would have to be obtained from facilities and the methodology for
estimating the pollutant reductions as the result of any pollution
prevention practices. Another concern of the Agency had to do with the
determination of when, in the new product development phase of work,
the practice represents a pollution prevention activity or is just part
of normal process development work in bringing a new product process to
full scale production. EPA was also concerned that pollutant discharge
or emission reductions achieved in the bench scale or pilot scale
product development activities may not be realized during full scale
production operations. In the period following publication of the NOA,
the Agency did not receive sufficient information relative to these
concerns to enable it to develop a viable pollution prevention
alternative based on this suggestion.

C. Updated Profile of the Industry

The pharmaceutical manufacturing industry covered by this
rulemaking is made up of 566 facilities located in 39 states, Puerto
Rico and the Virgin Islands. EPA estimates that 304 of these facilities
could be affected by today's final rule. The major concentrations of
manufacturing facilities are located in the Northeast, the Midwest and
Puerto Rico.
The pharmaceutical manufacturing industry is defined by four types
of manufacturing operations or processes. These activities result in
subcategorization for purposes of this rulemaking. The four
subcategories are referred to as:
Subcategory A: Fermentation
Subcategory B: Natural Extraction
Subcategory C: Chemical Synthesis
Subcategory D: Formulating, Mixing and Compounding
A complete discussion of each subcategory's manufacturing
operations and wastewater characteristics may be found in Sections 3
and 5 of the final Technical Development Document (TDD), ``Development
Document for Final Effluent Limitations Guidelines and Standards for
the Pharmaceutical Manufacturing Point Source Category'' (EPA 821-R-98-
005).
A fifth subcategory, Subcategory E: Research, was excluded from
regulation beyond the existing BPT regulation promulgated on October
27, 1983 at 48 FR 49808. The Research subcategory is defined by bench-
scale activities or operations related to the research on and
development of pharmaceutical products. BAT/BCT limitations for this
subcategory are determined on a case by case best professional judgment
(BPJ) basis. For indirect dischargers, the general prohibition in 40
CFR part 403 apply; in addition POTWs will establish local pretreatment
limits on a case by case basis as necessary.

D. Existing and Proposed Rules

EPA promulgated interim final BPT regulations for the
pharmaceutical manufacturing point source category on November 17, 1976
(41 FR 50676; 40 CFR Part 439, Subparts A through E). The five
subcategories of the pharmaceutical manufacturing industry (40 CFR part
439) were defined at that time as:
Subpart A--Fermentation Products Subcategory
Subpart B--Extraction Products Subcategory
Subpart C--Chemical Synthesis Subcategory
Subpart D--Mixing, Compounding, and Formulating
Subcategory
Subpart E--Research Subcategory
The 1976 BPT regulations set monthly limitations for biochemical
oxygen demand (BOD5) and chemical oxygen demand (COD) based
on percent removal for all subcategories. No daily maximum effluent
limitations were established for these parameters. The pH was set
within the range of 6.0 to 9.0 standard units. The regulations also set
maximum 30 day average concentration-based limitations for total
suspended solids (TSS) for subcategories B, D and E. No TSS limitations
were established for subcategories A and C. Subpart A was amended (42
FR 6813) on February 4, 1977, to improve the language referring to
separable mycelia and solvent recovery. The amendment also allowed the
inclusion of spent beers (broths) in the calculation of raw waste loads
for Subpart A in those instances where the spent beer is actually
treated in the wastewater treatment system.

[[Page 50392]]

On October 27, 1983, at 48 FR 49808, EPA revised the subcategory
names to those currently applicable and promulgated revised BPT, BAT,
PSES and PSNS for Subparts A thru D to cover the toxic pollutant
cyanide, conventional pollutants BOD5, TSS and pH, and the
nonconventional pollutant COD. The 1983 regulations kept intact the
percent reduction regulations for BOD5 and COD established in 1976 but
added floor concentration-based limitations for these parameters
applicable to subcategories B, D and E. The revisions for TSS consisted
of deriving the limitations by the use of a multiplication factor of
1.7 times each plant's BOD5 discharge. EPA also promulgated
BPT, BAT, PSES and PSNS for pH (6.0-9.0) and BAT concentration-based
limitations controlling the discharge of cyanide for subcategory A
through D. The Agency also proposed NSPS for BOD5, TSS and
pH in the October 1983 notice, but did not publish final NSPS for these
parameters.
On December 16, 1986, at 51 FR 45094, EPA promulgated BCT effluent
limitations guidelines for BOD5, TSS and pH for
subcategories A thru D. That final rule set BCT effluent limitations
equal to the existing BPT effluent limitations guidelines for
BOD5, TSS, and pH.
1. Clean Water Act Proposal
On May 2, 1995 at 60 FR 21592, EPA proposed revised BPT
concentration based limitations for BO5, COD and TSS based
on advanced biological treatment for all subcategories and cyanide
limitations based on hydrogen peroxide oxidation technology for the A
(Fermentation) and C (Chemical Synthesis) subcategories. For BAT, EPA
proposed end-of-pipe limitations for 53 organic pollutants plus
ammonia, cyanide and COD for subcategories A and C. For subcategories B
(Natural Extraction) and D (Formulating, Mixing and Compounding), EPA
proposed BAT limitations for 53 organic pollutants and COD. The
technology basis for the volatile organic compounds (VOCs) limitations
was steam stripping plus advanced biological treatment for
subcategories A and C and advanced biological treatment for
subcategories B and D. The technology basis for the non-volatile
organics was advanced biological treatment only, and the proposed
ammonia limitations were based on nitrification. The proposed BAT
cyanide limitations were equivalent to the BPT limitations, and the BCT
limitations were also proposed equal to BPT for all manufacturing
subcategories.
For NSPS, EPA proposed end-of-pipe standards for 53 organic
pollutants plus ammonia, BO5, TSS, cyanide and COD for
subcategories A and C and end-of-pipe standards for 53 organic
pollutants plus BO5, TSS, and COD for subcategories B and D.
The BO5, COD, and TSS standards were based on two sets of
performance data from the best performing plants in each of the A or C
and B or D subcategories. The end-of-pipe VOC limitations were based on
steam stripping with distillation and advanced biological treatment.
For PSES EPA detailed two coproposals (A and B) to control VOCs in
all subcategories. Coproposal A had pretreatment standards for 12
highly volatile organic compounds and 33 less volatile organic
compounds. To show compliance with the pretreatment standards,
monitoring for the 12 highly volatile compounds would have been
required in-plant. Coproposal B had only the pretreatment standards for
the 12 highly volatile compounds. In addition, EPA proposed cyanide
(identical to BPT) and ammonia standards (based on steam stripping) for
subcategories A and C. The proposed PSNS differed from PSES in that the
standards for all volatile organic compounds were based on steam
stripping plus distillation technologies.
Finally, EPA proposed that pilot plant wastewater would not be
regulated by Subcategory E (Research) limitations but under appropriate
manufacturing subcategory limitations.
2. Clean Air Act Proposal
On April 2, 1997 at 62 FR 15753, EPA proposed National Emission
Standards for Hazardous Air Pollutants (NESHAPs) for the
Pharmaceuticals Production Source Category. In that proposed rule, the
Agency proposed Maximum Available Control Technology (MACT) standards
for controlling emissions of hazardous air pollutants (HAPs) from
process vents, storage tanks, equipment leaks, wastewater collection
and treatment systems and heat exchange systems at pharmaceutical
manufacturing facilities that are determined to be major sources of
HAPs.
The proposed MACT standards for wastewater emission sources
contained two alternative formats for achieving compliance, a percent
removal and a reference control technology. Applicability
determination, definitions, and control requirements were similar to
the Hazardous Organic NESHAPs (HON) MACT standards for wastewater. The
proposed standard required facilities to control wastewater streams
that exceed the concentration cutoff where the process wastewater
stream exits the pharmaceutical process equipment identified as the
point of determination (POD). The proposed concentration cutoffs were
1,300 parts per million by weight (ppmw) for partially soluble HAPs and
5,200 ppmw for total HAPs at processes or PODs with annual HAP loads of
1 megagram per year or metric ton per year (Mg/yr).
Also, the proposed standard required all streams having a HAP
concentration of 10,000 ppmw to be controlled at facilities with annual
HAP loads of 1 Mg/yr or greater.
The proposed standards required that the control of wastewater
emissions be accomplished in one of the following manners: (1) Using a
design biotreatment system for soluble HAPs; (2) Demonstrating removals
achieving 99 percent by weight of partially soluble HAPs and 90 percent
by weight of soluble HAPs from treatment systems; or (3) Demonstrating
a removal of 95 percent by weight of total organic HAP from the
treatment system. The MACT standard proposal also discussed options for
CWA controls in light of the CAA MACT standard proposal for controlling
emissions from wastewater streams at pharmaceutical facilities being
covered by the proposed effluent limitations guidelines and standards.
EPA's intent was that the effluent limitations guidelines and standards
build on the MACT standards, and the discussion suggested several
options to accomplish this.
3. Clean Water Act Federal Register Notice of Availability
EPA published a Notice of Availability (NOA) in the Federal
Register on August 8, 1997 at 62 FR 42720. EPA published this Notice in
order to: allow public comment on the data received since the May 2,
1995 CWA proposal, further develop and revise options for the control
of the VOCs that were presented in the April 2, 1997 CAA MACT proposal,
and suggest responses to some comments on the 1995 CWA proposal.
In section II of the NOA, EPA provided the results of an EPA
sampling study designed to provide information concerning the pass
through analysis for water soluble organic pollutants such as methanol
and discussed the pass through analysis that EPA would be performing
with respect to these and other pollutants.
In section III, EPA presented revisions of the pretreatment options
which were earlier described in the MACT proposal, and presented
options for reducing the discharge loadings of VOCs not controlled by
the proposed MACT

[[Page 50393]]

standards. One option was compliance with the proposed MACT standards
together with additional PSES requirements for all VOCs except alcohols
and related compounds based on the performance database used in the
1995 proposal. A second option included coverage of additional
pollutants including alcohols and related compounds. EPA also presented
costs and loadings for two scenarios involving these two options. One
scenario would exclude facilities that discharged less than 10,000
pounds per year of pollutants of concern, while the other scenario
would not exclude them.
In section IV, EPA presented the results of analyses with respect
to the proposed data base for NSPS requirements for the conventional
pollutants, COD and ammonia, pollutant exclusions, use of surrogate
pollutants for compliance monitoring, small facility exclusion and
changes to engineering costs and loadings removal estimates. In
addition, EPA presented data editing criteria and methodologies for
deriving BPT and BAT effluent limitations and PSES. On pages 42722-
42724 of the NOA, EPA presented BPT, BAT limitations and PSES being
considered.

E. Discussion of Final Clean Air Act Rule Published Elsewhere in
Today's Federal Register

EPA received a number of comments on the proposed MACT standards
for wastewater streams. While certain changes were made (see the final
MACT rule published elsewhere in today's Federal Register) the controls
required by the proposed MACT standards have not changed. As proposed,
the final MACT incorporates the HON wastewater standards, thereby
clarifying the MACT requirements for off-site treatment of wastewater.
Under specified conditions, a source can transfer affected wastewater
streams containing soluble HAPs and less than 50 ppmw partially soluble
HAPs off-site for treatment. In addition, if the off-site treatment
facility is a POTW with uncovered headworks (grit chamber, primary
settling tanks, etc.) a demonstration that less than five percent of
the total soluble HAPs are emitted is required. For POTWs with
completely covered headworks, the final rule does not require a
demonstration that less than five percent of the total soluble HAPs are
emitted.

F. Relationship Between the MACT and CWA Rules

As noted above, the CAA MACT rule being promulgated today sets
emission standards for HAPs from wastewater collection and treatment
systems at major source pharmaceutical manufacturing facilities. The
CWA final effluent limitations guidelines and standards control the
discharge of toxic, conventional and nonconventional pollutants in
wastewater discharges from pharmaceutical manufacturing facilities.
Some of the water pollutants being controlled by today's effluent
guidelines and standards are also HAPs and thus these pollutants are
being controlled by both the MACT and CWA final rules. The extent of
the coverage of waterborne HAPs by the air and water rules will be
discussed in subsequent sections, as will the joint economic analysis
and environmental benefits assessment that were conducted for the two
rules.

G. Final Clean Water Act Effluent Guidelines Limitations and Standards
Rule

Today EPA is promulgating revised BPT limitations only for COD
based on advanced biological treatment for all four subcategories.
For subcategories A and C, EPA is promulgating BAT limitations for
COD equal to the revised BPT limitations and for 30 organic pollutants,
including 28 VOCs (of which 13 are HAPS) based on advanced biological
treatment identified as a basis for the revised COD limitations. In
addition, for subcategories A and C, EPA is promulgating BAT ammonia
limitations based on nitrification technology, and is modifying the BAT
compliance monitoring requirements for the existing cyanide
limitations.
For subcategories B and D, EPA is adding BAT limitations for COD
equal to the revised BPT requirements, and is withdrawing the existing
BPT and BAT cyanide limitations since the facilities in these
subcategories do not generate cyanide in their wastewaters.
The Agency is promulgating PSES for 23 VOCs (10 of which are HAPs)
plus ammonia for subcategories A and C, and is also clarifying the
compliance requirements for the existing cyanide pretreatment
standards. For subcategories B and D, EPA is promulgating PSES for the
5 VOCs (1 of which is a HAP) and, for the same reason given above, is
withdrawing the existing cyanide standards. Subcategories A and C
facilities must continue to comply with the cyanide standards, and
achieve compliance with the standards for ammonia and the 23 organic
pollutants within three years. Subcategories B and D facilities must
achieve compliance with the 5 organic pollutant standards within three
years. The compliance times of up to three years is being given because
of the design and installation of technologies used as a basis for the
standards, such as steam stripping and nitrification require sufficient
lead times for implementation.
EPA is promulgating NSPS for subcategories A and C equal to the BAT
limitations for COD, ammonia and the organic pollutants, including the
VOCs, and revised limitations for BOD5 and TSS based on
advanced biological treatment. EPA is also promulgating NSPS for
subcategories B and D equal to BAT for COD and revised limitations for
BOD5 and TSS based on advanced biological treatment, and is
withdrawing the existing cyanide NSPS for these two subcategories.
For PSNS EPA is promulgating standards equal to PSES for all
pollutants and subcategories and is withdrawing the existing cyanide
PSNS for subcategories B and D. Finally, EPA is promulgating BCT
limitations equal to the existing BPT limitations for BOD5,
TSS and pH.
In today's rule, EPA has republished many parts of the existing
guideline in Part 439 to make the changes made today easier to
understand, and also reformated the guideline to make it more clear and
easier to use. The republication or reformatting of existing
requirements is not intended to introduce substantive changes to these
regulatory provisions. For that reason, EPA believes prior notice and
comment on these provisions is unnecessary.

III. Summary of Most Significant Changes to Water Rules From
Proposal

This section describes the most significant changes to the rule
since proposal. Many of these changes have resulted from the comments
that are discussed below (see section X). This section will discuss the
major changes in the rule concerning revisions to the limitations and
standards for VOCs, changes in the BAT technology basis and changes in
the BPT and BAT limitations for pollutants other than the VOCs. More
detailed explanations for changes may be found in the comment response
document in the record of the final rule.

A. Limitations and Standards for Volatile Compounds

In today's final rule, EPA is not requiring that the limitations
for VOCs be measured in-plant as proposed. For all four subcategories,
BAT, NSPS, PSES, and PSNS limitations and standards, except for cyanide
limitations and standards in subcategories A and C, this rule does not
alter the generally applicable rule

[[Page 50394]]

(122.45(h) or 403.6(e)) that limitations generally are measured at the
end-of-pipe discharge point. This rule provides clarification of the
existing in-plant monitoring for cyanide as discussed in the
Implementation Section of this preamble (see section VIII A).
At proposal, EPA proposed PSES for 13 alcohols and related
pollutants (compounds) under coproposal B. These pollutants were
methanol, ethanol, n-propanol, isopropanol, n-butyl alcohol, tert-butyl
alcohol, amyl alcohol, formamide, N,N-dimethylaniline, pyridine, 1,4-
dioxane, aniline, and petroleum naphtha. No PSES/PSNS are being
promulgated for these pollutants today because EPA determined these
pollutants do not pass through POTWs or interfere with the treatment
works. (See section IV.E for a discussion of the passthrough analysis
for these pollutants).

B. Change in BAT Technology Basis for Organic Pollutants

In the August 8, 1997 NOA, EPA discussed changing the technology
basis for BAT organic pollutant limitations for subcategories A and C
facilities from in-plant steam stripping and advanced biological
treatment to advanced biological treatment only. EPA received comments
supporting this change in technology basis. The final MACT standards
being promulgated today will control most emissions of VOCs from
wastewaters at subcategories A and C direct discharging facilities
based on the use of steam stripping technology. Accordingly, EPA
believes that it is not necessary or appropriate to include this
technology in the BAT technology basis; the CWA limitations and
standards are calculated from a data base representing advanced
biological treatment only. Thus, EPA is promulgating BAT limitations
for all of the 30 organic pollutants for subcategories A and C
facilities based on advanced biological treatment only. EPA notes that
one facility not covered by the MACT standards would need to install
steam stripping technology in order to achieve the effluent limitations
following the biological treatment system.

C. BPT and BAT/BCT Limitation Changes

Based on the receipt of new data from commenters, proposed
limitations were revised for the nonconventional pollutants COD and
ammonia and a number of the organic pollutants. In addition, commenters
on the proposed limitations for the conventional pollutants BOD5 and
TSS, as well as COD, indicated that EPA should eliminate all non-
process wastewater in the calculation of limitations for these
parameters. In developing limitations for the proposal, EPA did not
back out the estimated non-process wastewater from the total wastewater
flow and adjust the concentration accordingly because the non-process
flow data provided by facilities in the data sets were only gross
estimates and were not based on daily measurements of non-process flow.
Despite requesting more precise information (such as daily non-process
flow data) from facilities that generated the data sets used to
calculate the proposed limitations for BOD5, TSS and COD, EPA did not
obtain this information. However, in the NOA, EPA presented revised
proposed limitations for BOD5 and TSS and COD that were calculated from
the existing plant data sets using the gross estimates of non-process
flow, as described below, to adjust the concentrations in addition to
several new data sets from plants other than those used for the
proposal.
In a previous EPA effluent limitations guidelines and standards
rulemaking for the Organic Chemicals, Plastics and Synthetic Fibers
(OCPSF) industry (52 FR 42522), only plant data sets that contained
less than 25 percent non-process wastewater through treatment were used
in calculating limitations. Thus, the 25 percent level of non-process
wastewater dilution was determined as a benchmark in order to evaluate
biological treatment performance. For the purposes of the NOA, in cases
where the non-process flow was estimated to be more than 25 percent of
the total flow, the non-process wastewater was backed out of the total
flow volume and the parameters corrected for the absence of this non-
process wastewater. However, for the final rule, limitations for COD
are developed from data sets in which the reported flow volume contains
less than 25 percent non-process wastewater and the limitations are
calculated without correcting the data sets for the non-process flow
dilution. This change is discussed further in section IV.D below. As
further discussed below, limitations for BOD5 and some of the remaining
TSS are not being revised at this time since the revised COD limits
requiring advanced biological treatment will incidentally remove a
large portion of the remaining BOD5 and TSS.
Another change to the proposal involved the limitations and
standards proposed for cyanide. EPA proposed BPT, BAT, NSPS, PSES and
PSNS limitations and standards for cyanide based on the performance of
hydrogen peroxide oxidation technology. Following the proposal, EPA
received comments indicating that the use of the hydrogen peroxide
technology to destroy cyanide could possibly result in equipment
explosions with certain types of wastewater. Other commenters indicated
that hydrogen peroxide technology may not be an appropriate cyanide
destruction technology for all treatment situations. Along with these
comments, EPA received additional data on the performance of alkaline
chlorination technology in destroying cyanide. Based on these comments
and the new performance data, EPA indicated in the NOA that it was
considering promulgating two sets of cyanide limitations, one based on
the performance of hydrogen peroxide technology and the other based on
the performance of alkaline chlorination technology. In the NOA, EPA
indicated that only those facilities that could demonstrate that a
potential safety hazard could result from their use of hydrogen
peroxide technology would be subject to the alkaline chlorination
limitations and standards. EPA also solicited information and comments
regarding wastestreams with high organic content as evidenced by high
COD or total organic carbon (TOC) levels, and at what levels these
pollutants would indicate that the wastestream(s) high organic content
would present a safety concern and would more appropriately be
controlled by limitations based on alkaline chlorination. After
consideration of the information provided in response to the
solicitation in the NOA, particularly new performance data representing
current (post 1990 base year) loadings, EPA has decided not to revise
the existing limitations and standards for cyanide based on the small
amount of cyanide discharge loadings that would be removed. However,
the final rule continues to require compliance with the cyanide
limitations be established in-plant, prior to commingling the cyanide
bearing wastestreams with non-cyanide wastestreams for those facilities
where the cyanide levels would be below the level of detection at the
end-of-pipe monitoring location.
Along with comments on its proposed numerical limitations and
standards for ammonia and organic pollutants, EPA received data
concerning the performance of steam strippers, advanced biological
treatment and nitrification in connection with these proposed
limitations. EPA evaluated these data, and provided revised numerical
limitations and standards in the NOA for ammonia, several organic
pollutants controlled by BAT technology (advanced biological treatment)
and several VOCs controlled

[[Page 50395]]

by steam stripping technology for PSES. As the result of the data
received and evaluated, along with comments on the NOA, EPA has changed
the numerical BAT limitations for ammonia. In response to comments in
the NOA indicating that indirect dischargers should be able to achieve
the PSES ammonia limitations using either two-step nitrification
technology or steam stripping, EPA has decided to set the PSES ammonia
limitations equal to the BAT ammonia limitations, and to provide that
indirect discharging subcategories A and C facilities discharging to
POTWs with nitrification capability need not comply with the
categorical limit for ammonia. EPA has also changed the numerical BAT
limitations and PSES for several organic pollutants based on its
analysis of data received in response to the proposal.

D. Pollutant Selection

EPA received several comments concerning the reasoning behind the
regulation of certain pollutants as well as the overall rationale for
selecting pollutants for regulation. In the NOA, EPA indicated that it
had reviewed the loadings bases of all the pollutants selected for
regulation and had determined that in the case of eight pollutants,
insufficient amounts of the pollutants are being discharged to justify
national regulation. These pollutants are diethyl ether, cyclohexane,
chloromethane, dimethylamine, methylamine, furfural, 2-methylpyridine
and trichlorofluoromethane. Since the NOA, EPA has reevaluated its
final loadings database and has determined that the exclusion of these
pollutants along with an additional 15 pollutants is appropriate. The
additional 15 pollutants are excluded from the BAT regulation based on
the lack of removals from current discharge or the control of
discharges of the pollutant by other regulated pollutant parameters.
These pollutants are butanone, formaldehyde, n-butanol, tertiary
butanol, n-propanol, ethylene glycol, polyethylene glycol 600, aniline,
petroleum naphtha, 1,4-dioxane, formamide and dimethyl formamide,
dimethylaniline, dimethylacetamide and pyridine.
EPA proposed PSES for 45 organic pollutants, 37 of which are VOCs,
under co-proposal A with compliance for the standards for 12 of the
VOCs to be monitored in-plant, and compliance for the standards for the
remaining 33 organics to be monitored at the end-of-pipe. In the NOA,
EPA presented two revised PSES options, under which EPA would
promulgate pretreatment standards for VOCs with end-of-pipe monitoring.
The pollutants not regulated under one of these PSES options include
water soluble alcohols such as methanol and related compounds. After
consideration of comments and evaluating the results of the
Barcelonetta POTW study and its implications on the final pass through
analysis (see further discussion of pass through analysis in section IV
E below) and further evaluation of incidental removals and the amount
of or discharge removals for the pollutants, EPA is promulgating PSES
and PSNS for 23 VOCs for subcategories A and C and 5 VOCs for
subcategories B and D. The PSES and PSNS do not include the alcohols
and related compounds, and are based on monitoring at the end-of-pipe
unless the POTW determines it to be impractical per 40 CFR 403.6(e).

IV. The Final Clean Water Act Regulation

This section discusses the applicability of the final rule,
regulatory options considered and the rationale for the selected
options for BPT, BCT, BAT, PSES, PSNS and NSPS.

A. Applicability and Scope of the Final Rule

Today's final effluent limitations guidelines and standards are
intended to cover pollutants in process wastewater discharges from
existing and new pharmaceutical manufacturing facilities. Based on
comments, EPA has revised the proposed scope of the rule. This final
rule contains revisions to the effluent limitations guidelines and
standards in four subcategories (A thru D) of the pharmaceutical
manufacturing point source category, EPA is not revising the scope of
the applicability for the fifth subcategory (Subcategory E-Research).
With regard to subcategory E facilities, EPA proposed to revise the
description of the research subcategory in the applicability section of
the existing subcategory E regulations to exclude pilot or full-scale
operations that generate wastewater using fermentation, extraction,
chemical synthesis or mixing, compounding and formulating from the
scope of subpart E, and these operations were proposed to be covered by
the appropriate subcategory A through D. After considering the comments
received concerning the regulation of wastewaters from pilot-scale
operations, EPA has decided not to change the existing description of
the research subcategory in the applicability section. EPA believes
that it does not have sufficient information concerning subcategory E
generated wastewaters to change the existing description. Subpart E
facilities remain subject to the BPT limitations in the existing
guidelines. If pilot scale operations occur at either stand alone
research facilities or during operations at manufacturing facilities,
then BAT and BCT limits for these wastewaters can be determined by
permit writers on a best professional judgment (BPJ) basis, or
similarly, such wastewater generated at indirect discharging facilities
may be addressed by the regulations found at 40 CFR 403.5 and by local
limits on a case-by-case basis.
Pharmaceutical manufacturers use many different raw materials and
manufacturing processes to create a wide range of products. These
products include medicinal and feed grades of all organic chemicals
having therapeutic value, whether obtained by chemical synthesis,
fermentation, extraction from naturally occurring plant or animal
substances, or by refining a technical grade product.
The pharmaceutical products, processes and activities covered by
the manufacturing subcategories in this final regulation include, but
are not limited to:
a. Biological products covered by the U.S. Department of Commerce,
Bureau of the Census Standard Industrial Classification (SIC) Code No.
2836, with the exception of diagnostic substances. (Products covered by
SIC Code No. 2836 were formerly covered under the 1977 SIC Code No.
2831.)
b. Medicinal chemicals and botanical products covered by SIC Code
No. 2833;
c. Pharmaceutical products covered by SIC Code No. 2834;
d. All fermentation, biological and natural extraction, chemical
synthesis and formulation products considered to be pharmaceutically
active ingredients by the Food and Drug Administration that are not
covered by SIC Code Nos. 2833, 2834, and 2836;
e. Multiple end-use products derived from pharmaceutical
manufacturing operations (e.g., components of formulations,
intermediates, or final products, provided that the primary use of the
product is intended for pharmaceutical purposes);
f. Products not covered by SIC Code Nos. 2833, 2834, and 2836 or
other categorical limitations and standards if they are manufactured by
a pharmaceutical manufacturer by processes that generate wastewaters
that in turn closely correspond to those of pharmaceutical products.
(An example of such a product is citric acid.)
g. Cosmetic preparations covered by SIC Code No. 2844 that contain
pharmaceutically active ingredients or

[[Page 50396]]

ingredients intended for treatment of some skin condition. (This group
of preparations does not include products such as lipsticks or perfumes
that serve to enhance appearance or to provide a pleasing odor, but do
not provide skin care. In general, this also excludes deodorants,
manicure preparations, shaving preparations and non-medicated shampoos
that do not function primarily as a skin treatment.)
A number of products and/or activities such as surgical and medical
manufacturing and medical laboratory activity are not part of the
pharmaceutical manufacturing category. A descriptive listing of the
products and activities that are specifically excluded from the
pharmaceutical manufacturing category are contained in the
applicability provision of the final rule and in sections 2 and 3 of
the final TDD.
In the NOA, EPA indicated that it was considering excluding from
the scope of the regulation organic chemical manufacturers covered by
the OCPSF regulation (40 CFR, Part 414) that manufacture pharmaceutical
intermediates and active ingredients provided that the pharmaceutical
portion of the process wastewater is less than 50 percent of the total
process wastewater. EPA received no adverse comments concerning this,
and has decided to promulgate this exclusion as described in the NOA.
Thus facilities will be covered by the existing OCPSF regulation for
both their OCPSF and pharmaceutical manufacturing process wastewaters
provided that the pharmaceutical portion of the process wastewater at
the facility is less than 50 percent of the total.

B. Options Selection

EPA evaluated final technology options for BPT, BAT, BCT, NSPS,
PSES and PSNS limitations and standards for all four subcategories A
thru D. The options considered for each level of control are discussed
below in sections IV.C thru H.

C. Best Practicable Control Technology Currently Available (BPT)

EPA proposed to revise BPT for the conventional pollutants
BOD5 and TSS, the nonconventional pollutant COD, and the
toxic pollutant cyanide for subcategories A and C, and for
subcategories B and D, proposed to revise BPT limitations for
BOD5, TSS, and COD and to withdraw the cyanide limitations.
In response to this proposal, EPA received comments claiming that EPA
lacks the legal authority to revise BPT for the conventional pollutants
since the proposed revised BPT limitations did not pass the BCT cost-
reasonableness test. EPA also received comments claiming that COD and
cyanide should not be regulated at BPT but only at the BAT level.
In today's rulemaking, EPA is revising BPT limitations only as to
COD. The current BPT limitations for BOD5, TSS and cyanide
will continue to apply (except for subcategories B and D where EPA is
withdrawing the BPT limitations for cyanide). Accordingly, issues
raised by commenters regarding EPA's legal authority to revise BPT for
BOD5, TSS, or cyanide do not need to be addressed in this
rulemaking. Nonetheless, EPA continues to believe that it has the legal
authority to revise BPT limitations as appropriate. EPA further
believes it can do so for conventional pollutants without having to
apply the BCT cost-reasonableness test. Because EPA's authority to
revise BPT limitations for conventional pollutants or cyanide is no
longer an issue in this rulemaking, EPA is providing only a general
statement of its statutory authority to revise BPT. For example,
section 304(b) of the CWA directs EPA to revise all effluent limitation
guidelines, including those based on BPT, at least annually if
appropriate. Similarly, section 304(m) directs EPA to establish a
schedule ``for the annual review and revision of promulgated effluent
guidelines, in accordance with subsection (b) of this section.'' EPA
does not believe that the addition of the BCT provisions to the CWA
supplanted the BPT provisions. When enacting the more recent BCT
provisions, Congress did not strip EPA of its explicit authority to
revise or update BPT as necessary and appropriate. Moreover, the
different purposes of BPT and BCT limitations would support an EPA
decision to promulgate best ``practicable'' control technology for
conventional pollutant control (represented by BPT), rather than the
higher ``best available'' standard (represented by BCT).
Similarly, it is the Agency's position that it is not required to
regulate COD or cyanide only at the BAT level. As noted above, section
304(b) of the CWA as well as section 304(m) directs EPA to revise all
effluent limitations guidelines, including those based on BPT, at least
annually if necessary and appropriate. It is EPA's view that the
addition of BAT provisions to the CWA did not supplant the BPT
provisions. When enacting the more recent BAT provisions, Congress did
not strip EPA of its authority to revise or update BPT as necessary and
appropriate. Further, the different purposes of BPT and BAT limitations
would support an EPA decision to promulgate revised effluent limitation
guidelines for nonconventional or toxic pollutants that reflect simply
the next generation of best ``practicable'' control technology
(represented by BPT), rather than the higher ``best available''
standard (represented by BAT).
Since EPA is not revising BPT limitations for cyanide (but rather
is modifying the compliance monitoring requirements for cyanide for
subcategories A and C, and withdrawing the limitations as to
subcategories B and D), the issue need not be addressed further in this
rulemaking.
EPA believes that the decision of whether or not to revise BPT for
nonconventional pollutants should be made based upon consideration of a
number of factors, including, but not necessarily limited to, cost, the
technology being considered and the relative performance being achieved
(best ``practicable'' versus best ``available''), the anticipated
pollutant reductions, and implementation burden on permit writers.
In this case, EPA has made a determination that the costs and
removals associated with the implementation of advanced biological
treatment at a best ``practicable'' level warrant revision of COD at
BPT. This is in part due to the relatively high concentrations of COD
in the effluent that are allowed under the existing percent removal BPT
limitations which are unique to this industry. In other cases, the
Agency has decided not to revise BPT (see, for example, Effluent
Limitation Guidelines for the Pulp, Paper, and Paperboard Category,
subparts B and E, 63 FR 18534, April 15, 1998).
As noted above, EPA proposed to revise BPT for the conventional
pollutants BOD5 and TSS, the nonconventional pollutant COD,
and the toxic pollutant cyanide for subcategories A and D, and for
subcategories B and D, to revise BPT limitations for BOD5,
TSS, and COD and to withdraw the existing cyanide limitations. The
technology basis of the proposed BPT limitations was advanced
biological treatment. EPA also determined that the level of performance
necessary for a plant to be considered as a best performer at the best
``practicable'' level was full compliance with the existing BPT
limitations. Of the plants considered as best performers at proposal,
EPA selected five A and C subcategory plants and two B and D
subcategory plants. The Agency then calculated long-term average
performance concentrations for regulated pollutants from the best
performing A and C and B and D plants.
In developing the final BPT limitations, EPA has essentially

[[Page 50397]]

followed the proposal methodology except that EPA used only data sets
representing less than 25 percent non-process wastewater through
treatment and included the additional data sets received since proposal
in its final limitations determinations. Except for one facility which
adds non-process wastewater after treatment but before the end-of-pipe
sample point, the BPT data sets were not corrected for non-process
wastewater and the final limitations were calculated using the plant
flow that included some non-process wastewater.
EPA did not back out the estimated non-process wastewater in
developing the proposed BPT concentration based limitations because
non-process flow data available at that time were only gross estimates
not identified in sufficient detail and were not based on daily
measurements of non-process flow. Regarding the proposed BPT
limitations, commenters indicated that EPA should eliminate all non-
process wastewater from the calculation of BPT limitations. EPA did not
have information such as daily non-process flow data from facilities
that generated the data sets used in the calculation of BPT and BAT
limitations for BOD5, TSS and COD to allow adjustment. In
the recent NOA, EPA presented BPT limitations for BOD5 and
TSS and BAT COD limitations that were calculated from plant data sets
which included the additional data submissions obtained since proposal
from which the non-process wastewater had been backed out. In cases
where the non-process flow was estimated by EPA to be more than 25
percent of the total flow using the available data, the fraction of the
non-process to process flow volume was used to calculate a correction
factor and the long-term average concentration values for each of the
BPT parameters were adjusted to reflect the parameters absence of this
non-process wastewater. No corrections were made to data sets where the
non-process flow was estimated to be less than 25 percent of the total
flow.
EPA received no adverse comments regarding these adjusted
limitations. However, based on further analysis, EPA believes that it
is more appropriate to follow the methodology used in developing the
final Organic Chemicals, Plastics, and Synthetic Fibers (OCPSF)
regulation (52 FR 52522) final BPT limitations. In that rule, only
plant data sets that contained less than 25 percent non-process
wastewater through treatment were used in the calculation of BPT
limits, and the effluent data were not adjusted to take into account
plant data sets that contained more than 25 percent non-process
wastewater through treatment. EPA selected this approach in calculating
the final BPT limitations in this rule for the same two reasons used
during development of the OCPSF rule. (See 52 FR 42522). First, using
data sets with greater than 25 percent non-process wastewater through
treatment introduces considerable uncertainty into the limitation
calculations because the flow data that would be used are only in part
based on daily flow measurements whereas the concentration-based
limitations are calculated from the long term average of daily
measurements over long periods of time (12-24 months). Second, the
final limitations should represent as much as possible the performance
of treatment technology on process wastewater. In determining permit
mass limits, permit writers and, where applicable, pretreatment control
authorities should identify the amount of non-process wastewater being
treated. The flow volume representing 25 percent or less of the total
flow should be included in the volume used to calculate allowable mass
discharges. Any additional volume would have to be evaluated on a case-
by-case basis to determine what, if any, mass allowances are
appropriate.
EPA considered four options for the final BPT limitations. Under
the first option, EPA would not revise the existing BPT limitations for
BOD5, TSS, COD and cyanide. No costs or removals are
associated with this option. Under the second option, EPA would revise
the BPT limitations based on advanced biological treatment only for
COD, and revise the monitoring requirements for the existing cyanide
limitations. Under option three, EPA would revise BPT limitations for
BOD5 and TSS based on advanced biological treatment and
revise the monitoring requirements for the existing cyanide
limitations. Under the fourth option, EPA would revise BPT limitations
for BOD5, TSS, and COD based on advanced biological
treatment, and revise the monitoring requirements for the existing
cyanide limitations. The options for all subcategories are the same,
except as to cyanide where the option for subcategories B and D
contains the option to withdraw the cyanide limitations rather than
just modify the monitoring requirements.
The pretax total annualized costs, pollutant removals, and costs
per pound removed associated with the options, except the ``no action''
option, are shown below in Table IV.C.1.

Table IV.C.1.--BPT Pretax Option Costs, Pollutant Removals and Cost per Pound Removed
----------------------------------------------------------------------------------------------------------------
Total
annualized Pollutant Cost per pound
Treatment option cost ($ removals ($1996/lb)
million 1997) (lbs)
----------------------------------------------------------------------------------------------------------------
A/C Subcategory
----------------------------------------------------------------------------------------------------------------
Clarify cyanide monitoring, revise COD only..................... $2.48 14,352,000 $0.17
Clarify cyanide monitoring, revise BOD5 & TSS................... 2.61 4,692,000 0.56
Clarify cyanide monitoring, revise BOD5, TSS, & COD............. 3.10 15,731,000 0.20
----------------------------------------------------------------------------------------------------------------
B/D Subcategory
----------------------------------------------------------------------------------------------------------------
Withdraw cyanide, revise COD only............................... $1.38 539,000 $2.56
Withdraw cyanide, revise BOD5 & TSS............................. 1.89 588,000 3.21
Withdraw cyanide, revise BOD5, TSS, & COD....................... 2.16 598,000 3.62
----------------------------------------------------------------------------------------------------------------

In selecting these treatment options, EPA considered the total cost
in relation to the effluent reduction benefits, the age of equipment
and facilities involved, the processes employed, process changes
required, engineering aspects of the control technologies, non-water
quality environmental impacts (including energy requirements) and

[[Page 50398]]

other factors in accordance with section 304(b)(1)(B) of the CWA.
EPA has determined to revise BPT effluent limitations only for COD.
EPA is also clarifying the compliance monitoring requirements for the
existing BPT limitations for cyanide for subcategories A and C, and
withdrawing the existing cyanide limitations for subcategories B and D.
As discussed above, EPA believes that it has the statutory authority to
revise BPT and that it has the discretion to determine whether to
revise BPT effluent limitations guidelines in particular circumstances.
The CWA requires EPA, when setting BPT, to examine the total cost of
treatment technologies in relation to the effluent reduction benefits
achieved. In addition, in determining whether to set BCT limitations,
the Agency needs to consider the reasonableness of the cost of reducing
conventional pollutants and compare the cost of removing those
pollutants by regulated plants and by POTWs. Accordingly, EPA examined
the use of advanced biological treatment as a basis for both BPT and
BCT limitations for BOD5 and TSS. The Agency found that the
reductions in these conventional contaminants achieved by this
technology were not commensurate with the costs, largely because of the
large operational costs associated with the removal of TSS. While it is
EPA's view that it can revise BPT limitations for conventional
pollutants without passing the BCT cost test (where the BPT effluent
reduction ratio is favorable), the Agency is not generally inclined to
do so unless the removals achieved by the existing BPT limitations are
significantly fewer than would be achieved through revision of BPT.
That was not the case here. Revising BPT (and BAT) for COD plants will
not only remove large amounts of COD, but also achieve significant
incidental removals of BOD5 and TSS. For this reason, EPA
has determined that it is not necessary to separately revise the BPT
limits for BOD5 and TSS in this case.
EPA has determined to revise BPT for COD because the biological
treatment technology used as a basis for the limitations really
represents BPT technology and is widely used in the industry.
The bulk parameter and nonconventional pollutant COD is an
indicator of organic matter in the wastestream that is susceptible to
strong oxidation, and as such would also measure organic material
susceptible to biochemical oxidation, as well as some that is more
difficult to oxidize biochemically. In addition, limited studies and
discharge monitoring data have identified toxicity associated with the
COD levels contained in effluents from pharmaceutical manufacturing
facilities. Further discussion of the toxicity levels measured in the
effluents from pharmaceutical manufacturing facilities is contained in
Section 6 of the TDD. The revised COD limitations are estimated to
remove approximately 14.9 million pounds annually, including incidental
removal of 2.7 million pounds of BOD at an annualized cost of $2.48
million ($1997).
The revised COD provisions require the use of either the new
effluent concentration limitations or the existing 74 percent reduction
requirement, depending upon which method determines the more stringent
plant permit limitation. This is being done in order to avoid back-
sliding issues for existing plants that because of low influent
concentration already meet lower effluent limits for COD.
With regard to cyanide, EPA is retaining the existing BPT
limitations for the A and C subcategories. Further revision of the BPT
cyanide limitations was not selected since the removals were estimated
to be less than 42 pounds per year, thus, determined not to be
beneficial in relation to the annualized costs of over $200,000
($1997).
However, EPA is modifying the requirements for compliance
monitoring (for subcategories A and C). The current limitations require
compliance monitoring after cyanide treatment and before dilution with
other wastestreams, or in the alternative, monitoring after mixing with
other wastestreams based on a standard dilution factor. Today's rule
does not change the prohibition on dilution to meet the effluent
limitations for cyanide. The rule continues to require monitoring for
compliance with the existing limitations in-plant, prior to the
commingling of cyanide-bearing wastestreams with non-cyanide bearing
wastestreams for those facilities where the cyanide levels would be
below the level of detection at the end-of-pipe monitoring location.
The only change in the monitoring requirements is to eliminate the
current dilution standard that applied industry-wide, and to allow
individual facilities to demonstrate that end-of-pipe monitoring for
cyanide is feasible (i.e., cyanide is detectable); those facilities may
continue to monitor at the end of pipe.
The ability of EPA to require in-plant monitoring has recently been
questioned in connection with the Great Lakes water quality guidance
program. American Iron and Steel Institute (AISI) v. EPA, 115 F.3d 979
(D.C. Cir. 1997). The Court held that although EPA has the authority to
require monitoring of internal wastestreams, see AISI, 115 F.3d at 995,
the CWA does not authorize EPA to require compliance with water quality
based effluent limitations at a point inside the facility and thereby
deprive a permittee of the ability to choose its own control system to
meet the limitations, see id. at 966. EPA does not believe that
decision controls here. The AISI court did not consider the question
whether EPA has authority to regulate internal wastestreams in the
context of technology-based controls such as BPT/BAT, PSES and NSPS/
PSNS. Unlike water quality-based effluent limitations, which are
calculated to ensure that water quality standards for the receiving
water are attained, technology-based limitations and standards are
derived to measure the performance of specific model technologies that
EPA is required by statute to identify. In identifying these
technologies, EPA is directed to consider precisely the type of
internal controls that are irrelevant to the development of water
quality-based effluent limitations, such as the processes employed,
process changes, and the engineering aspects of various types of
control techniques. EPA's technology-based effluent limitations are
intended to reflect, for each industrial category or subcategory, the
``base level'' of technology (including process changes) and to ensure
that ``in no case * * * should any plant be allowed to discharge more
pollutants per unit of production than is defined by that base level.''
E.I. du Pont de Nemours & Co. v. Train, 430 U.S. at 129 (1973).
EPA believes that it can require in-plant monitoring to demonstrate
compliance with technology-based effluent limitations in accordance
with the CWA and its regulations at 40 CFR 122.44(i), 122.45(h),
125.3(e) and 403.6(e). In today's rule, EPA is continuing to require
in-plant monitoring for cyanide except where cyanide can be detected in
the final effluent. Were EPA to require compliance monitoring of the
final effluent without adjustment for the amount of dilution in
cyanide-bearing waste streams, there would be no way to determine
whether the facility had adequately controlled for cyanide or whether
the effluent has simply been diluted below the analytical detection
level. Diluting pollutants in this manner rather than preventing their
discharge is inconsistent with achieving the removals represented by
the technology-based levels of control and hence with

[[Page 50399]]

the purposes of the limitations. It is also inconsistent with the goals
of the CWA in general.

D. Best Available Technology Economically Achievable

EPA proposed adding new end-of-pipe BAT limitations for 53 organic
pollutants plus ammonia, revising the existing cyanide limitations and
adding the BPT revised COD limitations for subcategories A and C. For
subcategories B and D, EPA proposed adding new end-of-pipe BAT
limitations for 53 organics, BPT revised COD limitations and
withdrawing the existing cyanide limitations. The technology basis for
the limitations for VOCs was steam stripping plus advanced biological
treatment for subcategories A and C and advanced biological treatment
for subcategories B and D. The technology basis for the ammonia
limitations was nitrification. The revised cyanide limitations for the
A and C subcategories were the same as the revised BPT proposed
limitations. For subcategories B and D cyanide limitations were
proposed to be withdrawn since facilities in these subcategories do not
use or generate cyanide in their wastewaters.
EPA received a number of comments indicating that steam stripping
technology was not appropriate for the treatment of VOCs and that
emissions of these pollutants from wastewater should be controlled by
CAA regulations. In the preamble to the proposed MACT standards, EPA
indicated that, in view of the MACT proposed wastewater standards, that
it was considering changing the BAT technology basis for subcategory A
and C VOCs limitations to end-of-pipe advanced biological treatment. In
the NOA, EPA reiterated this option and provided cost information which
compared the original proposal technology basis (steam stripping and
advanced biological treatment) to the advanced biological treatment
technology basis.
EPA also received comments on its proposed ammonia limitations.
Commenters indicated that the ammonia limitations were inadequately
supported by nitrification data. In the NOA, EPA indicated that after
reevaluating its nitrification data base, it intended to base the BAT
ammonia limitations on both one or two stage nitrification technology,
presented compliance costs estimates based on two stage nitrification
technology and revised limitations based on incorporating additional
data, including data representing two stage nitrification, into the
data base. In comments on the NOA, commenters indicated that some
plants employing the proposed technology basis did not believe that
they could achieve consistent compliance with the revised limitations.
In order to respond to these commenters, EPA evaluated additional
nitrification data received from facilities after the August 8, 1997
publication of the NOA. As a result of this evaluation, EPA has
recalculated the ammonia limitations that were presented in the NOA. In
doing so, EPA used only data that showed evidence that nitrification
was occurring and compared separate sets of limitations developed using
single-stage and two-stage nitrification data sets, respectively. The
results of this comparison gave final limitations less stringent than
those calculated for the NOA, but reflective of systems that nitrify
continuously whether they are one or two stage systems.
EPA considered three regulatory options as the basis for BAT
limitations for subcategory A and C facilities. All three options
modify the existing BAT regulations to parallel the BPT regulations and
to clarify the compliance monitoring point for the existing cyanide
limitations. The first option is a no cost revision which incorporates
the BPT clarification for cyanide and revised BPT limitations for COD.
The second option adds limitations for 30 organic pollutants based on
advanced biological treatment and revised limitations for COD equal to
the final BPT limitations and clarifies the compliance monitoring point
for cyanide. The third option adds limitations for 30 organic
pollutants based on advanced biological treatment, ammonia limitations
based on one or two stage biological nitrification technology,
incorporates the revised COD limitations and clarifies the compliance
monitoring point for cyanide. The pretax total annualized compliance
costs and pollutant removals associated with the second and third
options (only options incurring costs) are shown below in Table IV.D.1
for subcategories A and C:

Table IV.D.1--BAT Pretax Options Costs, and Pollutant Removals for
Subcategory A and C Direct Dischargers
------------------------------------------------------------------------
Total Pollutant
annualized removals
Regulatory option cost ($ (million lbs
million 1997) per yr)
------------------------------------------------------------------------
Add Organics and COD and clarify cyanide $2.3 1.4
Add Organics, Ammonia and COD and
clarify cyanide........................ 3.6 2.2
------------------------------------------------------------------------

EPA evaluated the costs and economic impacts associated with each
option and determined that all the options were economically
achievable. After considering the pollutant load removals, the costs,
as well as the non-water quality environmental impacts associated with
the options, EPA selected the third option which adds effluent
limitations for 30 organic pollutants, ammonia and COD and modifies the
cyanide monitoring requirements. EPA believes that this option is
economically achievable and there are no significant adverse non-water
quality impacts associated with it. In addition, EPA believes the
discharge loadings of ammonia, COD and the organic pollutants are
significant from subcategory A and C facilities, and that limitations
on these discharges are appropriate. EPA has also evaluated the
technology bases of the final BAT limitations in the context of the BAT
statutory factors, i.e., the age of equipment and facilities involved,
the process(s) employed, potential process changes and non-water
quality impacts such as energy requirements. EPA believes the final BAT
limitations are appropriate based on its assessment of these factors in
relation to A and C subcategory facilities.
For facilities with subcategories B and D operations, EPA has
identified only the pollutant COD for control by BAT limitations based
on advanced biological treatment (the technology selected as the basis
for the BPT limitations). As discussed under BPT, cyanide is not a
pollutant of concern for subcategories B and D operations and EPA is
withdrawing the current BAT cyanide limitations for facilities with
subcategories B and D operations. EPA

[[Page 50400]]

also has determined that ammonia is not a pollutant of concern for
these subcategories since ammonia is not found in significant amounts
in wastewaters from these operations.
Thus, for subcategories B and D, EPA considered two final BAT
regulatory options. The first option is a no cost option consisting of
the withdrawal of the existing cyanide limitations, the same as the
final BPT withdrawal of cyanide control and the addition of the BPT
revised COD limitations. The second option includes the withdrawal of
the existing cyanide limitations and the addition of the BPT revised
COD limitations and limitations based only on advanced biological
treatment for 30 of the same organic pollutants selected for regulation
at the subcategories A and C facilities.
The total annualized cost and annual pollutant removal associated
with the second option are $0.410 million ($1997) and 22,300 pounds per
year.
EPA has evaluated the discharge loadings of organic pollutants from
subcategories B and D facilities and has determined that 95 percent of
the discharge of organic pollutants is from two facilities. Most direct
discharging subcategories B and D facilities do not discharge any
organic pollutants. EPA believes these organic pollutant discharges are
not sufficient to justify national regulations for these subcategories.
If permit writers determine the need to further control the organic
pollutants from the two facilities, the appropriate limits contained in
the subcategories A and C BAT regulations may be used. For this final
rule, EPA has selected the first option, which is to only add the BPT
revised COD limitations to BAT for subcategories B and D facilities,
and to withdraw the existing cyanide limitations.

E. Pretreatment Standards for Existing Sources (PSES)

EPA proposed pretreatment standards for 45 organic pollutants
(including 37 VOCs), with in-plant monitoring for 12 VOCs and end-of-
pipe monitoring for the remaining 33 organics (25 of which are VOCs)
under coproposal A; and in-plant monitoring only for the 12 VOCs under
coproposal B. EPA received considerable comment on its proposal pass
through analysis which indicated that the 45 organic pollutants passed
through POTW treatment works. Thirty-seven of the organic pollutants,
including 13 alcohols and related compounds had Henry's Law Constants
greater than 10 -6 atm m3/gmole, which was the
physical property used to consider a pollutant to be too volatile to be
treated properly at POTWs. The other eight organic pollutants were
determined to pass through based on the BAT technology percent removal
exceeding that of well operated activated sludge treatment represented
by EPA's 50 POTW data base.
Many commenters objected to the assumption that pollutants with
Henry's Law constants greater than 10-6 atm m3/
gmole would be considered to pass through based on their volatility.
The pollutants commenters identified as being insufficiently volatile
and highly biodegradable included: methanol, ethanol and other
pollutants with Henry's Law constants lower than 1 x 10-5
atm m3/gmole. Commenters indicated that many of the alcohols
and related compounds were easily biodegraded by POTWs and did not pass
through.
EPA also received a number of comments concerning the proposed in-
plant monitoring point for the 12 VOCs. Commenters indicated that CAA
MACT standards not CWA pretreatment standards should control in-plant
emissions of these pollutants from internal wastestreams.
In order to address these and other comments related to controlling
the alcohols and related compounds, EPA conducted a sampling study in
August 1996 at a POTW in Barceloneta, Puerto Rico. This POTW treats
pharmaceutical industry wastewaters containing measurable amounts of
the predominant alcohols and related compounds, such as methanol,
ethanol and isopropanol. The purpose of the sampling study was to
determine the extent to which methanol and other compounds with similar
Henry's Law Constants volatilize in the primary treatment works
(aerated grit chambers and primary clarifiers) prior to the
biodegradation unit process. Amounts volatilized prior to the
biodegradation unit are not considered to be treated.
In the NOA, EPA published the preliminary results of the study
along with those of a separate bench-scale study of anaerobic
degradation in the Barceloneta primary clarifiers conducted by
industry. EPA indicated in the NOA that it was considering a finding of
no pass through for 13 of the organic pollutants (methanol and other
alcohols and related compounds) based on the belief that the
volatization of these pollutants in the primary works of POTWs is
roughly equivalent to that observed in the primary works of direct
discharging BAT level facilities. Thus, the treatment of these
pollutants by a well operated POTW is roughly equivalent to that
achieved by industrial facilities meeting BAT. As noted earlier in
section III.D. EPA proposed PSES for 45 organic pollutants, and
subsequently removed eight pollutants based on no pass through at the
POTWs, thus making a total of 21 (with the alcohols and related
compounds) not passing through POTWs.
In addition to discussing results of its pass through analyses in
the NOA, EPA presented two revised pretreatment options for all four
subcategories, with end-of-pipe monitoring for all VOCs including the
12 volatile pollutants for which in-plant monitoring for PSES/PSNS had
been proposed. In the NOA, EPA indicated that PSES for these 12
pollutants were unnecessary because they would be controlled by the
MACT wastewater standards which require an in-plant compliance
demonstration for 10 of the 12 VOCs which are HAPs. The remaining 12
VOCs, in addition to the two non-HAPs that are part of the 12 VOCs
discussed above, are controlled by end-of-pipe limits based on steam
stripping, with removals incidental to controlling HAPs either directly
by the MACT standards or separately from the MACT standards at smaller
facilities not covered by the MACT rule but controlled by this CWA
final rule.
In finalizing the methodology for the pass through analysis
discussed above, EPA relied on three criteria that had to be met before
a pollutant was deemed to pass through. These criteria included
volatility, solubility in water, and the BAT and POTW technologies
percent removal comparison. With regard to volatility, EPA raised its
Henry's Law Constant threshold for volatility from 1 x 10-
\6\ atm/gmole/m \3\ to 1 x 10-
\5\ atm/gmole/m\3\ based on comments that the Henry's
Law Constant used at proposal was not consistent with what was used for
the OCPSF final rule. Pollutants with Henry Law Constants greater than
1 x 10-\5\ atm/gmole/m \3\ were believed to
volatilize significantly before reaching treatment at a POTW. In
connection with volatility, in order to be consistent with the MACT
standards approved for controlling water soluble HAPs, EPA also
considered whether a pollutant was water soluble because water soluble
compounds are less likely to volatilize than compounds that are
partially soluble. Finally, EPA considered differences in removal
percentages for organic pollutants obtained by comparing the BAT model
treatment system percentage removal to the average pollutant removal
percentage achieved by well-operated POTWs achieving secondary
treatment performance standards.
In developing BAT pollutant removal percentages, EPA only used
pollutant data pairs where the influent

[[Page 50401]]

concentrations were greater than ten times the pollutant method
detection limits which was the approach used in developing the
supporting information for the NOA. In developing the final POTW
pollutant removal percentages, EPA utilized the acclimated data from
the same sources used to develop these percentages for the NOA. These
removal percentages are the POTW removal percentages used in the final
comparison. Thus, in order for a pollutant to be deemed to pass
through, it had to have a Henry's Law Constant greater than
1 x 10-\5\ atm/gmole/m \3\, be less than totally
soluble in water, and have a BAT removal percentage greater than its
POTW removal percentage. Based on this analysis, EPA has determined
that 23 organic pollutants in subcategories A and C and 5 organic
pollutants in subcategories B and D, that pass through POTWs are
regulated by pretreatment standards in today's rule. A more detailed
description of this analysis may be found in section 17 of the final
TDD.
In addition to pretreatment standards for VOCs, EPA proposed
ammonia standards based on either steam stripping or two-stage
nitrification. In May 1995 EPA proposed ammonia pretreatment standards
based only on steam stripping technology. The Agency received a number
of comments concerning the proposed ammonia pretreatment standards.
Some commenters indicated that steam stripping may not be a reliable
treatment technology. Others questioned the need for national ammonia
standards because many POTWs have imposed local limits for ammonia and
others have nitrification capability. EPA discussed both of these
concerns in the NOA. EPA suggested in the NOA that ammonia does not
pass through POTWs with nitrification, and requested comments on the
preliminary discussion not to set pretreatment standards for industrial
users which discharge to POTWs with this technology. Comments from POTW
control authorities and industry supported this approach to developing
PSES ammonia standards. The final rule contains ammonia pretreatment
standards only for subcategories A and C, based on the BAT technology
of nitrification and is applicable to those facilities discharging to
POTWs without nitrification capability.
EPA determined that cyanide passes through POTWs based on the
percent removal comparison with the hydrogen peroxide (BAT) technology.
Thus, EPA proposed revised cyanide pretreatment standards based on
hydrogen peroxide technology but maintaining that the standards based
on in-plant monitoring for the requirements. EPA received comments
raising safety concerns using this technology for high organic strength
wastes. Based on these comments and additional data submitted by
facilities, in the NOA, EPA proposed establishing two sets of cyanide
standards. One standard would be identical to the proposed standards
based on hydrogen peroxide technology, while the other standard would
be based on alkaline chlorination technology and applicable only to
those facilities that could demonstrate, due to safety concerns, that
hydrogen peroxide technology was not an appropriate technology to use
with their wastewater. EPA estimated compliance costs and loadings
removals to be the same for both sets of standards because it was
assumed that the vast majority of facilities would meet these standards
based on the use of the more expensive and efficient hydrogen peroxide
technology.
In developing the final PSES for subcategories A and C, EPA
considered three options. The first option was not to develop
pretreatment standards for ammonia or any of the VOC pollutants, and to
modify the monitoring requirements for the existing cyanide standards.
The second option would build on compliance with the MACT standard with
additional pretreatment standards for 23 VOCS based on steam stripping
technology and ammonia based on steam stripping or nitrification and
modify the cyanide monitoring requirements. The third option would be
the same as the second option, with the addition of revised
pretreatment standards for cyanide.
The annualized compliance costs (1997 dollars) and pollutant
removals for the second and third options (the only ones incurring
costs) are shown below in Table IV.E.1. EPA did not consider additional
options involving small facility exclusions because results of the
economic analyses for the small facilities using the costs for both
options described above showed that both options are economically
achievable (see section V of this preamble for more discussion).

Table IV.E.1--PSES Pretax Options Costs and Pollutant Removals for
Subcategories A and C Indirect Dischargers
------------------------------------------------------------------------
Total
annualized Pollutant
Treatment option cost ($ removals
million 1996) (million lbs)
------------------------------------------------------------------------
Add organics and ammonia and modify
cyanide monitoring requirements........ $44.5 10.653
Add organics and ammonia and revise
cyanide limits......................... 44.8 10.654
------------------------------------------------------------------------

Due to the low pollutant removals achievable by the revised cyanide
standards (approximately 1000 lbs per year with 97 percent of the
removals coming from one facility) in relation to the compliance costs,
EPA has decided not to revise the existing cyanide standards, and has
selected the option to add organics and ammonia only and modify the
current cyanide monitoring requirements. The selected option adds
standards for ammonia and the 23 organic pollutants determined to pass
through (see previous discussion in this section), and modifies the
monitoring point for the current cyanide pretreatment standards for
subcategories A and C.
EPA is setting pretreatment standards for ammonia for subcategories
A and C because of the high loads of ammonia currently being discharged
by a number of pharmaceutical facilities to POTWs that do not have
nitrification capability and receive wastewaters from subcategories A
and C facilities. However, EPA is aware that some POTWs treating
pharmaceutical wastewaters from these subcategories have nitrification
capability, and EPA has made a determination of no passthrough for
ammonia at these POTWS. Thus, PSES ammonia limitations will not apply
to subcategory A and C facilities discharging to POTWs with
nitrification capability. POTWs with nitrification capability oxidize
ammonium salts to nitrites (via Nitrosomonas bacteria) and the further
oxidize nitrites to nitrates via Nitrobacter bacteria and achieve
greater removals of ammonia than POTWs

[[Page 50402]]

without nitrification. Nitrification can be accomplished in either a
single or two-stage activated sludge system. In addition, POTWs that
have wetlands which are developed and maintained for the expressed
purpose of removing ammonia with a marsh/pond configuration are also
examples of having nitrification capability. Indicators of
nitrification capability are: (1) biological monitoring for ammonia
oxidizing bacteria (AOB) and nitrite oxidizing bacteria (NOB) to
determine if nitrification is occurring, and (2) analysis of the
nitrogen balance to determine if nitrifying bacteria reduce the amount
of ammonia and increase the amount of nitrite and nitrate.
For subcategories B and D, EPA considered two options. The first
option was not to add regulated pollutants to the existing PSES and,
since cyanide is not present in wastewaters for these subcategories
facilities, to withdraw the existing cyanide standards. Thus,
compliance with the MACT standard would be the only requirement for
controlling VOC pollutants. The second option was to add pretreatment
standards for 5 VOCs (not including the alcohols and related compounds
and 19 pollutants determined not to be present in subcategory B and D
wastewaters) based on steam stripping in addition to withdrawing the
existing cyanide standards. No ammonia standards were considered since
facilities in these subcategories do not generate significant levels of
ammonia in their wastewaters. The pretax annualized compliance cost for
this second option is $8.8 million ($1997) and annual pollutant
removals are 3.35 million pounds.
For PSES for subcategories B and D, EPA has selected the second
option. EPA is basing this selection on the fact that the 5 pollutants
(VOCs) have been determined to passthrough, and the pollutant removals
are relatively high with respect to the compliance costs. The costs are
economically achievable and the nonwater quality environmental impacts
are acceptable.

F. New Source Performance Standards (NSPS)

EPA proposed NSPS for 53 organic pollutants, BOD5, TSS
and COD based on steam stripping or distillation and advanced
biological treatment for subcategories A and C. EPA also proposed NSPS
for ammonia and cyanide based on nitrification and hydrogen peroxide
oxidation technologies, respectively for these two subcategories. EPA
received comments indicating that distillation technology was not a
demonstrated technology for removing soluble VOCs (such as methanol),
and therefore, should not be part of the technology basis of NSPS. EPA
has reevaluated its steam stripping and distillation database and has
concluded that distillation technology is sufficiently demonstrated to
be considered BADT (Best Available Demonstrated Technology). However,
after taking into account the high removal of these pollutants
achievable by steam stripping and advanced biological treatment, the
addition of distillation technology is unnecessary. Consequently EPA
did not consider distillation technology as part of final NSPS model
technology.
EPA evaluated technology options capable of achieving greater
pollutant removal of conventional pollutants (BOD5 and TSS),
COD, Organics, Cyanide and Ammonia than those selected as the basis for
existing source limitations (BPT, BCT and BAT). The only option
potentially capable of achieving additional removals involves the use
of granular activated carbon (GAC) adsorption technology. This
technology is capable of reducing the COD from some direct discharging
A and C subcategory facilities. However, there is only limited GAC
performance data available, from one pilot study.
For subcategories B and D, EPA proposed NSPS for 53 organic
pollutants, BOD5, TSS and COD based on in-plant steam
stripping with distillation and end-of-pipe advanced biological
treatment. As was the case with the proposed NSPS for subcategories A
and C, EPA received comments stating that use of distillation
technology as BADT for new sources is inappropriate because its ability
to remove methanol and other water soluble organic pollutants has not
been demonstrated with respect to representative wastestreams.
For subcategories A and C, EPA is promulgating NSPS equal to the
final BAT effluent limitations for 30 organic pollutants, cyanide and
ammonia. For subcategories B and D, EPA is promulgating NSPS equal to
BAT (including withdrawal of the existing cyanide standards). EPA is
also promulgating revised NSPS for BOD5, COD and TSS for all
four subcategories at a level equal to the discharge characteristics of
the best performing BPT plants which for COD is also the BAT/BPT level
of control. These final standards are based on the best available
demonstrated control technologies, which include advanced biological
treatment, cyanide destruct and nitrification. In developing these
final standards, the Agency considered factors including the cost of
achieving effluent reductions, non-water quality environmental impacts,
and energy requirements. EPA finds that the final standards represent
the best available demonstrated control technologies, are economically
achievable and do not present a barrier to entry and have acceptable
non-water quality environmental impacts.

G. Pretreatment Standards for New Sources (PSNS)

EPA proposed PSNS for 45 organic pollutants, cyanide and ammonia
for subcategories A and C, and the same 45 organic pollutants only, for
subcategories B and D. The technology basis for the proposed organic
pollutant standards was steam stripping with distillation, and the
technology bases for the proposed cyanide and ammonia standards were
hydrogen peroxide oxidation and steam stripping technologies,
respectively.
The proposed pretreatment standards for new sources were more
stringent than the proposed PSES. However, for the final rule, EPA was
unable to identify a technology that would achieve greater removal of
the pollutants to be controlled by the PSES being promulgated today and
is therefore promulgating PSNS equal to PSES for all four
subcategories.

H. Best Conventional Pollutant Control Technology (BCT)

EPA proposed BCT equal to BPT for the conventional pollutants
BOD5 and TSS for all four subcategories. The Agency
indicated that it had not identified technologies that achieve greater
removals of conventional pollutants other than those associated with
the proposed revision of BPT limits, and that these technologies did
not pass the two-part BCT cost reasonable test. EPA has not received
any comments concerning its proposal BCT cost test analysis. The Agency
has repeated the cost test with the postproposal data, with the same
results. Based on the failure to identify any incremental conventional
pollutant removal technology options that pass the BCT cost reasonable
test, EPA is promulgating BCT limitations equal to the existing BPT
limitations for BOD5 and TSS for all subcategories.

V. Assessment of Costs and Impacts for the Final Pharmaceutical
Regulations

A. Introduction

The economic analysis for the final pharmaceutical effluent
limitations guidelines and standards assesses the costs and impacts of
these guidelines. The results of this analysis are contained in the
record for this final

[[Page 50403]]

rule and are summarized in a document entitled Economic Analysis for
Final Effluent Guidelines and Standards for the Pharmaceutical Industry
(EPA-821-B-98-009). Included in the Economic Analysis (EA) and
summarized below are (1) the annualized costs of the rule by
subcategory, separately and together with the costs of the MACT
standards rule discussed previously; (2) the impacts of the rule both
separately and together with the MACT standards on pharmaceutical
facilities, both existing and new sources; (3) the impacts of these
rules on pharmaceutical firms; (4) the impacts of these rules on
employment and communities; and (5) other secondary impacts on trade,
inflation, POTWs, environmental justice, and distributional equity.
Also included in the EA are a Final Regulatory Flexibility Analysis as
required under the Regulatory Flexibility Act and a Cost-Benefit
Analysis, as required under the Unfunded Mandates Reform Act (UMRA) and
Executive Order 12866, which are summarized in Sections V.E and V.F of
this preamble. An additional document, Cost Effectiveness Analysis for
Effluent Limitations Guidelines and Standards for the Pharmaceutical
Industry (EPA-821-B-98-010), assesses the cost-effectiveness of the
rule. The results of this analysis are summarized below in Section V.G.

B. Summary of the Economic Analysis Methodology and Data

EPA determined the annualized costs of compliance in exactly the
same way as was done for proposal, with the exception of the choice of
discount rate (discussed in V.C). Costs are annualized at seven percent
over 16 years (a 1-year installation period a 15-year project life is
assumed). The cost annualization also accounts for tax shields on both
O&M and depreciation (calculated using the modified accelerated cost
recovery system allowed by IRS rules) to develop a posttax estimate of
annual costs (see Section 4 of the Economic Analysis for a detailed
discussion). For analytical consistency, MACT standards costs are also
annualized in the same way, both pretax and posttax. This is slightly
different from the way EPA annualized the MACT standards costs in the
preamble to the MACT standards rule, where costs are annualized at
seven percent over ten years (with no delay for installation) to create
a pretax annual cost (i.e., without accounting for tax shields).
Additionally, the MACT standards costs presented in the preamble to the
MACT standards rule include costs for new sources, which are not
included in this preamble. Despite the differences in annualization
method, the current cost annualization approach in no way conflicts
with the alternative analysis.
To assess impacts on firms and facilities, EPA has set up three
baselines in the analysis. Baseline 1 is the usual baseline analyzed in
all effluent guidelines. It is a scenario that reflects a baseline
condition without additional regulation, that is, no additional
effluent limitations guidelines and standards or MACT standards costs
are considered. This baseline is taken from the current (i.e., 1990
Survey) financial data. Baseline 2 incorporates certain MACT standards
costs pertaining only to wastewater emission controls, and does not
include costs for controlling emissions from process vents, equipment
leaks and storage tanks. This baseline is presented in the EA, but
results of this baseline (which are not appreciably different from
those for Baseline 1) are not discussed at length in this preamble.
Baseline 3 incorporates costs for all components associated with the
MACT standards rule. EPA estimated the capital and operating costs for
MACT standards cost components for emission controls on wastewater
streams (on which Baseline 2 is based), as well as the capital and
operating costs for all MACT components (on which Baseline 3 is based)
as a part of the Agency's MACT standards rulemaking process.
To model Baseline 2, EPA used the capital and operating costs
associated with the wastewater emission controls for all facilities in
the MACT analysis for which costs were developed and matched them to
the facilities that are also in the effluent guidelines analysis.
However, a number of facilities in the effluent guidelines analysis are
not covered by the MACT standards and were not assigned MACT costs.
EPA annualized the costs at seven percent over 16 years in the cost
annualization model and also developed a present value of posttax
compliance costs over this same time frame. EPA subtracted the present
value posttax compliance costs from the Baseline 1 present value
posttax facility earnings (derived from the Survey data) to determine
Baseline 2 posttax earnings for each facility in the effluent
guidelines analysis. EPA used this same approach to derive Baseline 3
posttax earnings (for those facilities without MACT standards costs,
earnings are the same in all three baselines).
A facility whose posttax earnings are zero or negative in Baseline
1 is counted as a Baseline 1 closure; a facility whose posttax earnings
are zero or negative in Baseline 2 is counted as a Baseline 2 closure;
and a facility whose posttax earnings are zero or negative in Baseline
3 is counted as a Baseline 3 closure.
EPA then incorporated the present value posttax costs of the
effluent guidelines into each of the baselines in the same way as MACT
standards costs were incorporated to calculate postcompliance, posttax
earnings. EPA then tallied the closure results (in terms of whether
postcompliance, posttax earnings are zero or negative) by counting
postcompliance closures incrementally from each baseline. In other
words, EPA considered any closures that occurred additional to those
occurring in each of the baselines as postcompliance closures under the
three baseline scenarios. Any facilities that certified that the
effluent guidelines would have no impact on them were assumed not to
close under any baseline or in postcompliance. Note that as in the
proposal Economic Impact Analysis (EIA), impacts on single-facility
firms were assessed at the firm level.
MACT standards costs were also incorporated into firm-level data
under the same three baseline scenarios. In the firm-level analysis,
however, the key data that could change were assets, liabilities, and
earnings before interest and taxes, which were used in an equation
called Altman's Z, a multi-discriminant ratio analysis approach to
identifying relative firm health. This equation is composed of several
common financial ratios that are weighted according to their relative
ability to predict bankruptcy based on empirical industry data. The
result of this equation is called the Altman's Z-score. Scores below a
certain value are considered indicative of poor financial health and a
high likelihood of bankruptcy.
For Baseline 1, EPA used the current survey data in the Altman's Z
model to determine a Baseline 1 Altman's Z-score. For Baseline 2, EPA
took the MACT standards capital costs aggregated at the firm level
(since firms often own more than one facility) and adjusted both assets
and liabilities to reflect the acquisition of capital equipment through
an increase in debt. EPA then adjusted earnings before interest and
taxes by subtracting the annualized amount of operating costs plus
depreciation computed by the cost annualization model, given the
Baseline 2 MACT standards capital and operating costs (also aggregated
at the firm level) and then computed a Baseline 2 Altman's Z-score.
EPA used the same approach using the Baseline 3 MACT standards
operating and capital costs to create the Baseline 3 Altman's Z-score.
If any of

[[Page 50404]]

these three baseline scores were below the cutoff point considered a
sign of poor financial health, EPA considered the firm a baseline
failure.
Compliance costs for the effluent guidelines were then used in the
same manner to further adjust the financial data used in the Altman's Z
model in each of the baselines. Where the Altman's Z-score changed from
one reflecting a healthy firm or one in indeterminate status in any of
the baselines to one of poor financial health, EPA considered the firm
to be a postcompliance firm failure relative to the baseline under
consideration.
EPA's methodology for computing output and employment effects is
discussed in detail in Section V.C. These effects are presented as net
effects in Section V.D.4. To compute net effects, EPA calculated both
losses and gains in output and employment and subtracted losses from
gains (or vice versa). Thus EPA calculated net national-level output
effects, net national-level employment effects, and net direct
employment effects (employment losses in the pharmaceutical industry
driven by output losses in the industry). EPA also estimated the
employment losses estimated to occur as a result of closures and
failures. These types of losses were used to determine whether any
community-level impacts are likely.
Trade impacts were assessed in the same way as in the EIA for the
proposal, except that a profit margin analysis has been added, as
described below in Section V.C. Impacts on inflation were assessed by
comparing the cost of the regulation to gross domestic product (GDP).
The potential for distributional impacts was assessed by identifying
facilities where compliance costs were greater than 10 percent of
operating costs and determining what types of products might be most
affected if costs are passed through to consumers. The users of these
products were then qualitatively identified to determine if these
potential users might be disproportionately represented by economically
disadvantaged groups. Impacts on environmental justice were also
qualitatively addressed.

C. Changes to the Economic Analysis Since Proposal

The most significant change in the EA since proposal is associated
with the change in costs. The costs of the effluent limitations
guidelines and standards for the pharmaceutical industry point source
category are now substantially lower than those estimated at proposal
because the costs of controlling air emissions are now a part of EPA's
MACT standards. Impacts from the final rule do not change measurably
from proposal, however, mostly because impacts both now and at proposal
were estimated to be very small.
Costs for control of air pollutants, previously assigned to the
effluent guidelines at proposal, are now assigned to the MACT standards
requirements. The economic analyses show the impacts of the effluent
guidelines against three separate regulatory baselines: no MACT
standards requirements in place, wastewater emissions control and
treatment system requirements in place, and all MACT standards
requirements in place (see Section II.E. of this preamble for a
description of MACT standards requirements). In this way, EPA can
present impacts from the effluent guidelines alone and in combination
with impacts from the MACT standards requirements. The methods EPA used
to assess the impact of MACT standards on the baselines against which
the effluent guidelines are measured were discussed in Section V.B.
EPA is now using a seven percent discount rate in all of its
analyses. Previously, the Agency used the seven percent rate only in
determining the pretax cost of the regulation. EPA has chosen to use a
seven percent social discount rate (in real terms) in this analysis,
rather than the 11.4 percent discount rate used in the proposal, for
two reasons. First, the seven percent discount rate is strongly
recommended by the Office of Management and Budget for use in economic
analyses (see the EA for more details). Second, the cost of capital has
generally declined since 1990. This change in discount rate, however,
has little effect on the analysis. A comparison of estimated impacts in
the proposal to impacts as estimated here show that the analyses are
not sensitive to assumptions about discount rates in the ranges used.
In terms of content, the economic analyses are now presented as a
more comprehensive report, in which the EIA and Regulatory Impact
Analysis (RIA) have been combined into one report (the EA). The cost-
benefit portion of the RIA is now contained in Section 10 of the EA
report.
EPA has also made a few methodological changes in its firm and
facility analyses. In the EIA for the proposal, EPA included salvage
value in the calculations for the facility closure analysis for
projection of baseline closures (i.e., before compliance costs are
considered) and postcompliance closures. EPA recognized some potential
difficulties with the salvage value calculations and, in the proposal
EIA, investigated the effects of assuming salvage value does not play a
role in determining facility viability. EPA found that the facility
closure projections were not sensitive to the alternate salvage value
assumption. Furthermore, industry also commented that using salvage
value overstated baseline closures. Thus EPA believes that its current
analysis, which does not consider salvage value but rather uses
negative posttax earnings as the indicator of closure, is the best
methodology to use, given the uncertainty of salvage value data.
An additional difference in the closure analysis addresses the
issue of non-self-supporting facilities (baseline facility closures).
In the current analysis, EPA investigates all baseline closures at the
firm level to determine if a multi-facility firm could install and
operate pollution control equipment at all of its affected facilities,
including those estimated as baseline closures. If the firms can
continue to support a baseline closure facility without risk of
failure, EPA determines that impacts to the firm and its affected
facilities are minimal. EPA performed this analysis under the
assumption that if the facility was not expected to support itself in
the baseline, the firm level is the appropriate level at which to
assess impacts.
EPA also modified the methodology for determining impacts on firms.
In response to comments that baseline firm failures were overstated
because the Agency used benchmarks that identified lowest quartile
firms as baseline failures, EPA reassessed the methodology and turned
to a more sophisticated method for determining firm financial health.
EPA used a multi-discriminant analysis approach for evaluating the
financial health of firms. This analysis, developed by Edward Altman,
is known as Altman'

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A98-21027. Public record. Not legal advice.