&EPA
United States
Environmental Protection
Agency
Office Of Water
(4305)
EPA823-R-95-005
February 1995
Environmental Assessment
Of The Proposed Effluent
Guidelines For The
Pharmaceutical Manufacturing
Industry
-------
-------
ENVIROHMENTAL ASSESSMENT OF THE
PROPOSED EFFLUENT GUIDELINES
FOR THE
PHARMACEUTICAL MANUFACTURING INDUSTRY
Volume I
,, Final Report,,,.,
^ February, 28, 1995
Prepared for:,.. '-•-'
"U.S. Environmental Protection Ageacy
; -;, Office 'of Water-
Office of Science and Technology
Standards; and -Applied Sdence
401 M SteeetT S.W.
Washington, B.C,
Ed Gardetto
Tas^; Manager
-------
-------
TABLE OF CONTENTS
Page No.
EXECUTIVE SUMMARY vii
1. INTRODUCTION 1
2. METHODOLOGY 3
2.1 Projected Water Quality Impacts 3
2.1.1 Comparison of Instream Concentrations with Ambient Water
Quality Criteria 3
2.1.1.1 Direct Discharging Facilities 4
2.1.1.2 Indirect Discharging Facilities 6
2.1.1.3 Assumptions and Caveats 9
2.1.2 Estimation of Human Health Risks and Benefits 11
2.1.2.1 Fish Tissue 11
2.1.2.2 Drinking Water 13
2.1.2.3 Assumptions and Caveats 14
2.2 Projected Air Quality Impacts 15
2.2.1 Estimation of Human Health Risks and Benefits (Fugitive Air
Emissions) 15
2.2.1.1 Preliminary Screening . 16
2.2.1.2 Atmospheric Dispersion Modeling 18
2.2.1.3 Risk Calculations 19
2.2.2 Estimation of POTW Occupational Risks and Benefits 20
2.2.3 Assumptions and Caveats 22
2.3 Documented Environmental Impacts 24
3. DATA SOURCES 25
3.1 Water Quality Impacts 25
3.1.1 Facility-Specific Data 25
3.1.2 Information Used to Evaluate POTW Operations . . 26
3.1.3 Water Quality Criteria (WQC) 27
3.1.3.1 Aquatic Life 27
3.1.3.2 Human Health 28
3.1.4 Information Used to Evaluate Human Health Risks and Benefits .. 32
3.2 Air Quality Impacts 32
3.2.1 Facility-Specific Data 33
3.2.2 Population and Climatologic Data 33
3.2.3 Information Used to Evaluate Human Health Risks and Benefits .. 34
3.3 Literature Review 35
-------
TABLE OF CONTENTS (continued)
Page No.
4. SUMMARY OF RESULTS 36
4.1 Projected Water Quality Impacts 36
4.1.1 Comparison of Instream Concentrations with Ambient Water
Quality Criteria 36
4.1.1.1 Direct Discharges 36
4.1.1.2 Indirect Discharges 38
4.1.2 Estimation of Human Health Risks and Benefits 41
4.1.2.1 Direct Discharges 41
4.1.2.2 Indirect Discharges 42
4.2 Projected Air Quality Impacts 43
4.2.1 Comparison of Human Health Risks and Benefits (Fugitive Air
Emissions) 44
4.2.1.1 "Low Estimate1' 44
4.2.1.2 "Maximum Estimate" 45
4.2.2 Comparison of POTW Occupational Risks and Benefits 46
4.3 Documented Environmental Impacts 47
5. REFERENCES
R-l
11
-------
VOLUME n
Page No.
Appendix A Pharmaceutical Manufacturing Facility-Specific Data A-l
Appendix B National Oceanic and Atmospheric Administration's (NOAA)
Dissolved Concentration Potentials (DCPs) B-l
Appendix C Water Quality Analysis Data Parameters C-l
Appendix D Risks and Benefits Analysis Information . . D-l
Appendix E Air Quality Analysis E-l
Appendix F Direct Discharger Analysis at Current (Baseline) and
Distillation Treatment Levels F-l
Appendix G Indirect Discharger Analysis of Current (Baseline) and
Distillation Pretreatment Levels G-l
Appendix H POTW Analysis at Current (Baseline) and
Distillation Pretreatment Levels H-l
Appendix I Direct Discharger Risks and Benefits Analysis at Current
(Baseline) and Distillation Treatment Levels 1-1
Appendix J Indirect Discharger Analysis at Current (Baseline) and
Distillation Treatment Levels J-l
Appendix K Air Quality Analysis Results K-l
ui
-------
-------
LIST OF TABLES
Page No.
Table 1. Frequency of Pollutants from 14 AC Direct Pharmaceutical
Manufacturing Facilities Discharging to 14 Receiving Streams 48
Table 2. Summary of Pollutant Loadings for AC Direct and Indirect
Pharmaceutical Manufacturers 49
Table 3. Summary of Projected Criteria Excursions for AC Dkect
Pharmaceutical Dischargers 50
Table 4. Summary of Pollutants Projected to Exceed Criteria for
AC Direct Pharmaceutical Dischargers 51
Table 5. Frequency of Pollutants from 3 BD Direct Pharmaceutical
Manufacturing Facilities Discharging to 3 Receiving Streams . 52
Table 6. Summary of Pollutant Loadings for BD Direct and Indirect
Pharmaceutical Manufacturers 53
Table 7. Summary of Projected Criteria Excursions for BD Direct
Pharmaceutical Dischargers 54
Table 8. Frequency of Pollutants from 61 AC Indirect Pharmaceutical
Manufacturing Facilities Which Discharge to 43 POTWS on 42
Receiving Streams 55
Table 9. Summary of Projected Criteria Excursions for AC Indirect
Pharmaceutical Dischargers 56
Table 10. Summary of Pollutants Projected to Exceed Criteria for AC
Indirect Pharmaceutical Dischargers 57
Table 11. Summary of Projected POTW Inhibition and Sludge Contamination
Problems from AC Indirect Pharmaceutical Dischargers 58
Table 12. Summary of Pollutants From AC Indirect Pharmaceutical Dischargers
Projected to Cause POTW Inhibition and Sludge Contamination
Problems 59
IV
-------
LIST OF TABLES (continued)
Page No.
Table 13.
Table 14.
Table 15.
Table 16.
Table 17.
Table 18.
Table 19.
Table 20.
Table 21.
Table 22.
Table 23.
Table 24.
Frequency of Pollutants from 55 BD Indirect Pharmaceutical
Manufacturing Facilities Which Discharge to 45 POTWS
on 45 Receiving Streams •
Summary of Projected Criteria Excursions for BD Indirect
Pharmaceutical Dischargers
Summary of Pollutants Projected to Exceed Criteria for BD
Indirect Pharmaceutical Dischargers •
Summary of Projected POTW Inhibition and Sludge Contamination
Problems from BD Indirect Pharmaceutical Dischargers
Summary of Potential Human Health Impacts for AC/BD Direct
Pharmaceutical Dischargers (Fish Tissue Consumption)
60
61
62
63
64
Summary of Pollutants Projected to Cause Human Health Impacts for
AC/BD Direct Pharmaeutical Dischargers at Current Discharge Levels
(Fish Tissue Consumption)
Summary of Potential Human Health Impacts for AC/BD Direct
Pharmaceutical Dischargers (Drinking Water Consumption) . . .
Summary of Pollutants Projected to Cause Human Health Impacts
for AC/BD Direct Dischargers (Drinking Water Consumption) . .
Summary of Potential Human Health Impacts for AC/BD Indirect
Pharmaceutical Dischargers (Fish Tissue Consumption)
Summary of Pollutants Projected to Cause Human Health Impacts for
AC/BD Indirect Pharmaceutical Dischargers (Fish Tissue Consumption)
Summary of Potential Human Health Impacts for AC/BD Indirect
Pharmaceutical Dischargers (Drinking Water Consumption)
. . 65
. . 66
. . 67
. . 68
. . 69
Summary of Pollutants Projected to Cause Human Health Impacts for
AC/BD Indirect Pharmaceutical Discharges (Drinking Water
Consumption)
70
71
-------
LIST OF TABLES (continued)
Page No.
Table 25. Summary of Air Quality Modeling Analysis for Pharmaceutical
Fugitive Emissions (Low End Estimate Loading) 72
Table 26. Summary of Air Quality Modeling Analysis for Pharmaceutical
Fugitive Emissions (Maximum Estimate Loading) .73
Table 27. Summary of Potential POTW Occupational Exposure Impacts for
Pharmaceutical Indirect Discharges 74
Table 28. Environmental Impact Case Studies of Pharmaceutical Manufacturing
Wastes 75
VI
-------
-------
EXECUTIVE SUMMARY
The Environmental Assessment of the Pharmaceutical Manufacturing Industry quantifies
water quality-related benefits for AC and BD facilities* based on site specific analyses of
current conditions and the conditions that would be achieved by steam stripping process changes.
Instream pollutant concentrations from direct and indirect discharges are estimated using stream
dilution modeling. The benefits to aquatic life are projected by comparing the modeled instream
pollutant concentrations to EPA aquatic life criteria guidance or to toxic effect values. Human
health benefits are projected by: (1) comparing estimated instream concentrations to health-based
water quality toxic effect values or criteria; and (2) estimating the potential reduction of
carcinogenic risk and non-carcinogenic hazard from consuming contaminated fish or drinking
water. Upper-bound individual cancer risks, population risks, and non-cancer hazards are
estimated using modeled instream pollutant concentrations and standard EPA assumptions.
Modeled pollutant concentrations in fish are used to estimate cancer risk and non-cancer hazards
among the general population, sports fishermen and their families, and subsistence fishermen and
their families. Inhibition of POTW operations and sewage sludge contamination are also
evaluated based on current and steam stripping pretreatment levels. Inhibition of POTW
operations are estimated by comparing modeled POTW influent concentrations to available
inhibition levels.
The modeling for this analysis was orginally performed using the EPA distillation process
change as the proposed treatment option. The results of this initial analysis showed limited
environmental gains beyond current treatment. Subsequently, steam stripping was selected by
EPA as the proposed treatment option. The overall projected loads from these two treatment
technologies are roughly comparable. Therefore, although distillation is a more effective
treatment technology for removing volatile organic pollutants than is steam stripping, the initial
analysis was rerun using the steam stripping pollutant loads only for those facilities for which
facilities use fermentation or chemical synthesis processes and BD facilities use
extraction, mixing, compounding and formulating processes.
Vll
-------
aquatic life or human health criteria excursions, or POTW inhibition problems were projected
at current loadings.
The water quality modeling results for 14 direct AC facilities discharging to 14 receiving
streams indicate that at current discharge levels, instream concentrations of 2 pollutants are
projected to exceed chronic aquatic life criteria or toxic effect levels in 7 percent of the
receiving streams. Instream concentrations of 3 pollutants (using a target risk of KT6 for
carcinogens) are projected to exceed human health criteria (developed for consumption of
water and organisms) in 14 percent of the receiving streams. Steam stripping is projected to
eliminate all aquatic life and human health excursions. No excursions of aguatic life criteria
or toxic effect levels or of human health criteria or toxic effect levels are projected at current
or steam stripping discharge levels for 3 direct BD facilities discharging to 3 receiving streams.
Modeling results for 61 indirect AC facilities which discharge to 43 POTWs on 42
receiving streams indicate that at current or steam stripping discharge levels, instream pollutant
concentrations are not projected to exceed chronic aquatic life criteria or toxic effect levels.
At current discharge levels, instream concentrations of 4 pollutants (using a target risk of KT6
for carcinogens) are projected to exceed human health criteria or toxic effect levels (developed
for consumption of water and organisms) in 10 percent of the receiving streams. Projected
excursions of human health criteria are reduced to 1 pollutant hi 5 percent of the receiving
streams at steam stripping pretreatment discharge levels. Results for 55 indirect BD facilities
which discharge to 45 POTWs on 45 receiving streams indicate instream concentrations of 1
pollutant are projected to exceed chronic aquatic life criteria or toxic effect levels in 2 percent
of the receiving streams at current discharge levels. Instream concentrations of 1 pollutant
(using a target risk of KT6 for the carcinogens) are projected to exceed human health criteria
or toxic effect levels (developed for consumption of water and organisms) hi 2 percent of the
receiving streams. No excursions of aquatic life or human health criteria are projected at
steam stripping discharge levels.
vui
-------
In addition, the excess annual cancer cases at current and, therefore, at steam stripping
treatment levels are projected to be far less than 1 from the ingestion of contaminated fish and
drinking water for AC and BD direct and indirect discharges. No systemic toxic effects are
projected at the current treatment level or at the steam stripping treatment level. Seven
pollutants are projected to contribute to POTW inhibition at 13 percent of the 46 POTWs
receiving the discharge from AC facilities. These POTW impacts are projected to be reduced
to three pollutants causing POTW problems at 11 percent of the POTWs by steam stripping.
No inhibition problems are projected at the 50 POTWs receiving discharges from BD facilities
at current or steam stripping pretreatment levels. No sewage sludge regulatory standards are
available for the pollutants analyzed to evaluate potential sludge contamination problems.
The Environmental Assessment of the Pharmaceutical Manufacturing Industry also
evaluates air quality-related benefits. Air quality-related benefits are quantified based on
potential risks to the general public from on-site fugitive emissions from open-air biological
treatment using air dispersion modeling. Two modeling estimates are made based on reductions
in pollutant loads - one based on 308 questionnaire loading data ("low estimate") and one based
on conservative engineering loading estimates ("maximum estimate"). Potential risks to POTW
maintenance workers from occupational exposure to a mixture of gases partitioning from influent
wastewater were also quantified by comparing modeled vapor-phase pollutant concentrations to
ACGIH Threshold Limit Values (TLVs).
The air dispersion modeling results of EPA's release estimates indicate that, based on the
"low estimate", the estimated number of people exposed to risk levels greater than 10"6 would
be reduced by approximately 190,000. This would result in a reduction of 0.023 annual cancer
cases. Based on estimates of the "maximum benefit" of regulating emissions, there would be
a benefit of 0.35 annual cancer cases. Exposure to pollutants released from pharmaceutical
manufacturing facilities are projected to exceed TLVs for POTW workers at 12 POTWs treating
52 pollutants at current discharge levels. At distillation pretreatment discharge levels, TLVs
are projected to be exceeded at 7 POTWs treating 44 pollutants. At steam stripping
IX
-------
pretreatment discharge levels, which are slightly higher than distillation pretreatment discharge
levels, TLVs are projected to be exceeded at 6 to 12 POTWs treating 52 pollutants.
Documented environmental impacts on aquatic life, human health, and POTW operations
from pollutant discharges from pharmaceutical manufacturing facilities are summarized in the
Environmental Assessment of the Pharmaceutical Manufacturing Industry. The summary data
are based on a review published scientific literature, newspaper articles, and survey data. A
total of 16 studies reported environmental impacts from pharmaceutical manufacturing. These
impacts included: (1) human health effects (general population and occupational exposure) such
as dizziness, nausea, respiratory and dermal problems and endocrine dysfunction (reproductive);
(2) aquatic life effects, such as fish kills; (3) effects on the quality of receiving water,
groundwater, soils, sediments, and drinking water; and (4) impairment of POTW operations.
Four pharmaceutical manufacturing facilities are identified by States as being point sources
causing water quality problems and are included on their CWA Section 304(1) short list for
discharges of toxics.
-------
1. INTRODUCTION
The purpose of this report is to present (1) an assessment of the water quality benefits
of controlling the discharge from pharmaceutical manufacturing facilities to surface waters and
publicly-owned treatment works (POTWs), and (2) an assessment of the air quality benefits of
controlling on-site fugitive emissions to ambient air from open-air wastewater treatment at
pharmaceutical manufacturing facilities and volatization of chemical discharges at POTWs.
Potential aquatic life and human health impacts of direct discharges on receiving stream water
quality and of indirect discharges on POTWs and their receiving streams are projected at current
and steam stripping treatment levels by quantifying pollutant releases and by using stream
modeling techniques. The potential benefits to human health are evaluated by estimating the
potential reduction of carcinogenic risk and non-carcinogenic hazard from consuming
contaminated fish or drinking water. Potential human health impacts from fugitive air emissions
are projected at low and maximum potential benefit levels using an air dispersion model. Risks
to POTW maintenance workers, who may be exposed to pollutants volatilizing from influent
wastewaters, are also estimated. In addition, recent literature is reviewed for evidence of
documented environmental impacts (e.g., case studies) on aquatic life, human health, and POTW
operations and for impacts on the quality of receiving water and ambient air.
The modeling for this analysis was originally performed using the EPA distillation
process change as the proposed treatment option. The results of this initial analysis showed
limited environmental gains beyond current treatment. Subsequently, steam stripping was
selected by EPA as the proposed treatment option. Although distillation is a more effective
treatment technology for removing volatile organic compounds than is steam stripping^ the
overall projected loads from these two treatment technologies are roughly comparable.
Therefore, the initial analysis was rerun using the steam stripping pollutant loads only for those
facilities for which aquatic life or human health criteria excursions, or POTW inhibition
problems were projected at current loadings.
-------
The following sections of this report describe: 1) the methodology used in the evaluation
of projected water and air quality impacts for direct and indirect discharging facilities and
potential human health risks and benefits (including assumptions and caveats) and in the
evaluation of documented environmental impacts; 2) data sources used for evaluating water and
air quality impacts such as plant-specific data, information used to evaluate POTW operations,
water quality criteria, population and clirnatologic data, and information used to evaluate human
health risks and benefits; 3) a summary of the results of this analysis; and 4) a complete list of
references cited in this report. The various appendices presented in Volume H provide
additional detail on the specific information addressed in the main report.
-------
2. METHODOLOGY
2.1 Projected Water Quality Imparts
The water quality impacts of pharmaceutical manufacturing discharges are evaluated by
comparing projected instream concentrations with ambient water quality criteria,2 and by
estimating the human health risks associated with the consumption of fish and drinking water
from waterbodies impacted by the pharmaceutical industry at various treatment options. The
methodologies used in this evaluation are described in detail below.
2.1.1 Comparison of Instream Concentrations with Ambient Water Quality Criteria
Current and steam stripping pollutant releases3 are quantified and compared and
potential aquatic life and human health impacts resulting from current and steam stripping
pollutant releases are evaluated using stream modeling techniques. Projected instream
concentrations for each pollutant are compared to EPA water quality criteria guidance or to toxic
effect levels (i.e., lowest reported or estimated toxic concentration) for pollutants for which no
EPA water quality criteria guidance has been published. Inhibition of POTW operation and
2In performing this analysis, EPA used guidance documents published by EPA that
recommend numeric human health and aquatic life water quality criteria for numerous pollutants.
States often consult these guidance documents when adopting water quality criteria as part of
their water-quality standards. However, because those State adopted criteria may vary, EPA
used the nationwide criteria guidance as the most representative values.
3Only pollutants that are proposed for regulation are evaluated (See Section 4.1); however,
three pollutants have been removed as candidates since this analysis was completed (glycol
ethers, bis(chloromethyl) ether and dimethylcarbamyl chloride). In addition, pollutant loadings
at steam stripping treatment levels have been received and modeled since this analysis was
initially completed using distillation pollutant loadings. Because the projected loads from these
two treatment technologies are roughly comparable, the initial analysis was rerun using the steam
stripping pollutant loads only for those facilities for which aquatic life or human health criteria
excursions or POTW inhibition problems were projected at current loadings.
-------
sludge contamination are also evaluated. The following three sections describe the methodology
and assumptions used for evaluating the impact of direct and indirect discharging facilities.
2.1.1.1 Direct Discharging Facilities
Using a stream dilution model that does not account for fate processes, projected instream
concentrations are calculated at current and steam stripping treatment levels for stream segments
with AC and BD4 direct discharging facilities. For stream segment!} with multiple
pharmaceutical facilities, pollutant loadings are summed before concentrations are calculated.
The dilution model used for estimating instream concentrations is as follows.
FF + SF
xCF
(Eq. 1)
where:
Cis
L
OD
FF
SF
CF
instream pollutant concentration 0*g/L)
facility pollutant loading (Ibs/year)
facility operation (days/year)
facility flow (million gal/day)
receiving stream flow (million gal/day)
conversion factors for units
The facility-specific data (i.e., pollutant loading, operating days, and facility flow) used
in Eq. 1 are derived from various sources as described in Section 3.1.1 of this report. Three
receiving stream flow conditions (1Q10 low flow, 7Q10 low flow, and harmonic mean flow) are
used for the two treatment technology options. The 1Q10 and 7Q10 flows are the lowest 1-day
and the lowest consecutive 7-day average flow during any 10-year period, respectively, and are
used to estimate potential acute and chronic aquatic life impacts, respectively, as recommended
4AC faculties use fermentation or chemical synthesis processes and BD facilities use
extraction, mixing, compounding and formulating processes.
-------
in the Technical Support Document for Water Quality-based Toxics Control (U.S. EPA, 199 la).
The harmonic mean flow is defined as the reciprocal of the mean value of the reciprocal of
individual values and is used to estimate potential human health impacts. EPA recommends the
long-term harmonic mean flow as the design flow for assessing potential human health impacts
because it provides a more conservative estimate than the arithmetic mean flow. 7Q10 flows
are also not appropriate for assessing potential human health impacts because they have no
consistent relationship with the long-term mean dilution.
Because stream flows are not available for hydrologically complex waters such as bays,
estuaries, and oceans, site-specific critical dilution factors (CDFs) or estuarine dissolved
concentration potentials (DCPs) are used to predict pollutant concentrations for facilities
discharging to estuaries and bays as follows.
c« =
(Eq.2)
where:
Ces
L
OD
FF
CDF
CF
estuary pollutant concentration 0*g/L)
facility pollutant loading (Ibs/year)
facility operation (days/year)
facility flow (million gal/day)
critical dilution factor
conversion factors for units
e: = L x DCP x CF
(Eq.3)
where:
Ces
L
DCP
CF
estuary pollutant concentration (/tg/L)
facility pollutant loading (Ibs/year)
dissolved concentration potential (mg/L)
conversion factor for units
-------
Site-specific critical dilution factors are obtained from a survey of States and Regions recently
conducted by EPA's Office of Pollution Prevention and Toxics (OPPT) (Mixing Zone Dilution
Fac^forNewChemicalExposureAssessments,^^!^^^^^ 1992a). TheStrategic
Assessment Branch of the National Oceanic and Atmospheric Administration's (NOAA) Ocean
Assessments Division has developed DCPs based on freshwater inflow and salinity gradients to
predict pollutant concentrations in each estuary in the National Estuarine Inventory (NED Data
Atlas. These DCPs are applied to predict concentrations. They do not consider pollutant fate
and are designed strictly to simulate concentrations of nonreactive dissolved substances. In
addition, the DCPs reflect the predicted estuary-wide response and may not be indicative of site-
specific locations.
Water quality excursions are determined by dividing the projected instream (Eq. 1) or
estuary (Eq. 2 and Eq. 3) pollutant concentrations by EPA ambient water quality criteria or toxic
effect levels. A value greater than 1.0 indicates an excursion.
2.1.1.2 Indirect Discharging Facilities
(a) Water Quality Impacts
A stream dilution model is used to project receiving stream impacts resulting from
releases by AC and BD indirect discharging facilities as shown in Eq. 4. For stream segments
with multiple pharmaceutical facilities, pollutant loadings are summed before concentrations are
calculated. The facility-specific data used in Eq. 4 are derived from various sources as
described in Section 3.1.1 of this report. Three receiving stream flow conditions (1Q10 low
flow, 7Q10 low flow, and harmonic mean flow) are used for the two treatment technology
options. Pollutant concentrations are predicted for POTWs located on bays and estuaries using
site-specific CDFs or NOAA's DCP calculations (Eq. 5 and Eq. 6).
-------
, (Z/OD)
PF + SF
(Eq. 4)
where:
L
OD
TMT
PF
SF
CF
instream pollutant concentration
facility pollutant loading (Ibs/year)
facility operation (days/year)
POTW treatment removal efficiency
POTW flow (million gal/day)
receiving stream flow (million gal/day)
conversion factors for units
'LjOD x (\-TMTj\
FF
J
x CF \f CDF
(Eq. 5)
where:
'"es
L
OD
TMT
PF
CDF
CF
estuary pollutant concentration 0*g/L)
facility pollutant loading (Ibs/year)
facility operation (days/year)
POTW treatment removal efficiency
POTW flow (million gal/day)
critical dilution factor
conversion factors for units
C7 = L x (1-7M7) x DCP x CF
(Eq. 6)
where:
ces
L
TMT
DCP
CF
estuary pollutant concentration 0*g/L)
facility pollutant loading (Ibs/year)
POTW treatment removal efficiency
dissolved concentration potential (mg/L)
conversion factors for units
-------
Impacts are determined by comparing projected instream pollutant concentrations (Eq. 4)
at reported POTW flows and at 1Q10 low, 7Q10 low, and harmonic mean receiving stream
flows with EPA water quality criteria or toxic effect levels for the protection of aquatic life and
human health (see Section 2.1.1.1 for discussion on receiving stream flows); projected estuary
pollutant concentrations (Eq. 5 and Eq. 6), based on CDFs or DCPs, are compared to EPA
water quality criteria or toxic effect levels to determine impacts. Water quality criteria
excursions are determined by dividing the projected instream or estuary pollutant concentration
by the EPA water quality criteria or toxic effect levels. A value greater than 1.0 indicates an
excursion.
(b) Impacts on POTWs
Impacts on POTW operations are calculated in terms of inhibition of POTW processes
(i.e., inhibition of microbial degradation) and contamination of POTW sludges. Inhibition of
POTW operations is determined by dividing calculated POTW influent levels (Eq. 7) with
chemical-specific inhibition threshold levels. Excursions are indicated by a value greater than
1.0.
=
pF
(Eq. 7)
where:
S"
OD
PF
CF
POTW influent concentration
facility pollutant loading (Ibs/year)
facility operation (days)
POTW flow (million gal/day)
conversion factors for units
8
-------
Contamination of sludge (thereby limiting its use for land application, etc.) is evaluated by
dividing projected pollutant concentrations in sludge (Eq. 8) by available EPA-developed criteria
values for sludge. A value greater than 1.0 indicates an excursion.
Csp = (Z/OZ>) x TMT x PART x SGF x CF
(Eq. 8)
where:
T*
J-.
OD
TMT
PART
SGF
CF
sludge pollutant concentration (mg/kg)
facility pollutant loading (Ibs/year)
facility operation (days/year)
POTW treatment removal efficiency
chemical-specific sludge partition factor
sludge generation factor (5.96 ppm)
conversion factors for units.
Facility-specific data and information used to evaluate POTWs are derived from the
sources described in Sections 3.1.1 and 3.1.2. For facilities that discharge to the same POTW,
their individual loadings are summed before the POTW influent and sludge concentrations are
calculated.
The partition factor is a measure of the tendency for the pollutant to partition in sludge
when it is removed from wastewater. For predicting sludge generation, the model assumes that
1,400 pounds of sludge are generated for each million gallons of wastewater processed (Metcalf
& Eddy, 1972). This results in a sludge generation factor of 5.96 (that is, for every 1 ppb of
pollutant removed from wastewater and partitioned to sludge, the concentration in sludge is 5.96
ppm dry weight).
2.1.1.3 Assumptions and Caveats
The following assumptions are used in this analysis:
-------
Background concentrations of each pollutant, both in the receiving stream and in
the POTW influent, are equal to zero; therefore, only the impacts of discharging
facilities are evaluated. (This assumption may result in an overly conservative
assessment of environmental benefits regarding the frequency of water quality
standards exceedance.)
An exposure duration of 365 days is used to determine the likelihood of actual
excursions of human health criteria or toxic effect levels.
Complete mixing of discharge flow and stream flow occurs across the stream at
the discharge point. This mixing results in the calculation of an. "average stream"
concentration even though the actual concentration may vary across the width and
depth of the stream.
The process water at each facility and the water discharged to a POTW are
obtained from a source other than the receiving stream.
The pollutant load to the receiving stream is assumed to be continuous and is
assumed to be representative of long-term facility operations. This assumption
may overestimate risks to human health and aquatic life.
1Q10 and 7Q10 receiving stream flow rates are used to estimate aquatic life
impacts, and harmonic mean flow rates are used to estimate human health
impacts. 1Q10 low flows are estimated using the results of a regression analysis
conducted by Versar for EPA's Office of Pollution Prevention and Toxics (OPPT)
of 1Q10 and 7Q10 flows from representative U.S. rivers and streams (Upgrade
of Flow Statistics Used to Estimate Surface Water Chemical Concentrations for
Aquatic and Human Exposure Assessment, Versar, 1992). Harmonic mean flows
are estimated from the mean and 7Q10 flows as recommended in the Technical
Support Document for Water-Quality-based Toxics Control (U.S. EPA, 1991a).
These flows may not be the same as those used by specific states to assess
impacts.
Pollutant fate processes such as sediment adsorption, volatilization, and hydrolysis
are not considered. This may result in estimated instream concentrations that are
environmentally conservative (higher).
Pollutants without a specific POTW treatment removal efficiency, provided by
EPA or found in the literature are assigned a removal efficiency of zero;
pollutants without a specific partition factor are assigned a value of zero.
Water quality criteria or toxic effect levels developed for freshwater organisms
are used in the analysis of facilities discharging to estuaries or bays.
10
-------
Of those facilities reporting a wastewater discharge, the number of facilities
modeled was limited by available data on receiving streams and POTWs.
2.1.2 Estimation of Human Health Risks and Benefits
The potential benefits to human health are evaluated by estimating the risks (carcinogenic
and systemic effects) associated with reducing pollutant levels in fish tissue and drinking water
from current to steam stripping treatment levels.^ The following three sections describe the
methodology and assumptions used for evaluating the human health risks from the consumption
of fish tissue and drinking water derived from water bodies impacted by AC and BD direct and
indirect discharging facilities.
2.1.2.1 Fish Tissue
To determine the potential benefits, in terms of reduced cancer cases, associated with
reducing levels hi fish tissue, lifetime average daily doses (LADD) and individual risk levels are
estimated for each pollutant discharged from a facility based on the instream pollutant
concentrations calculated at current and steam stripping treatment levels in the site-specific
stream dilution analysis (see Section 2.1.1). Estimates are presented for sport fishermen,
subsistence fishermen and the general population. LADDs are calculated as follows.
LADD = (C xIRx BCF xFxD)l(BWxLT)
(Eq.9)
where:
LADD
C
= potential lifetime average daily dose (mg/kg/day)
= exposure concentration (mg/L)
5Since no substantial risks are projected at current treatment levels, i.e., the highest
wasteload, the analysis was not rerun for steam stripping. Distillation treatment levels are used
hi the analysis.
11
-------
IR
BCF
F
D
BW
LT
ingestion rate (see Section 2.1.2.3 - Assumptions)
bioconcentration factor, I/kg (whole body x 0.5)
frequency duration (365 days/year)
exposure duration (70 years)
body weight (70 kg)
lifetime (70 years x 365 days/year)
Individual risks are calculated as follows:
R = LADD x SF
(Eq. 10)
where:
R
SF
individual risk level
potency slope factor (mg/kg-day)"1
The estimated individual pollutant risk levels are then applied to the potentially exposed
populations of sport fishermen, subsistence fishermen, and the general population to estimate the
potential number of excess annual cancer cases occurring over the life of the population. The
number of excess cancer cases is then summed on a pollutant, facility, and overall industry
basis. The number of reduced cancer cases are assumed to be the difference between the
estimated risks at current and steam stripping/distillation treatment levels.
Potential reductions in risks due to reproductive, developmental, or other chronic and
subchronic toxic effects are estimated by comparing the estimated average daily dose and the
oral reference dose (RfD) for a given chemical pollutant as follows:
HQ = ORH%D
(Eq. 11)
where:
HQ
hazard quotient
12
-------
OKI
RfD
oral intake (LADD x BW, mg/day)
reference dose (mg/day assuming a body weight of 70 kg)
A hazard index (i.e., sum of individual pollutant hazard quotients) is then calculated for
each facility or receiving stream. A hazard index greater than 1.0 indicates that toxic effects
may occur in exposed populations. The size of the subpopulations affected are summed and
compared at the various treatment levels to assess benefits in terms of reduced systemic toxicity.
2.1.2.2 Drinking Water
Potential benefits associated with reducing levels in drinking water are determined in a
similar manner. LADDs for drinking water consumption are calculated as follows:
LADD = (C x IR x F x D ) / ( BW x LT )
(Eq. 12)
where:
LADD
C
IR
F
D
BW
LT
potential lifetime average daily dose (mg/kg/day)
exposure concentration (mg/L)
ingestion rate (2L/day)
frequency duration (365 days/year)
exposure duration (70 years)
body weight (70 kg)
lifetime (70 years x 365 days/year)
Estimated individual pollutant risk levels greater than 10"6 are applied to the population served
downstream by any drinking water utilities within 50 miles from each discharge site to determine
the number of excess annual cancer cases that may occur during the life of the population.
Systemic toxicant effects are evaluated by estimating the sizes of populations exposed to
pollutants from a given facility the sum of whose individual hazard quotients yields a hazard
index (HI) greater than 1.0.
13
-------
2.1.2.3 Assumptions and Caveats
The following assumptions are used in the human health risks and benefits analysis.
• A linear relationship is assumed between pollutant loading reductions and benefits
attributed to the clean-up of surface waters.
• Synergistic effects of multiple chemicals on aquatic ecosystems are not assessed.
Therefore, the total benefit of reducing toxics may be underestimated.
• The total number of persons who might consume recreationally caught fish and
the number that rely upon fish on a subsistence basis in each State is estimated,
in part, by assuming that these fishermen regularly share their catch with family
members. Therefore, the number of fishermen in each State is multiplied by the
average household size in each State. The remainder of the population of these
states is assumed to be the "general population" consuming commercially caught
fish.
• Five percent of the resident fishermen in a given State are assumed to be
subsistence fishermen. The other 95 percent are assumed to be sport fishermen.
• Commercially or recreationally valuable species are assumed to occur or be taken
in the vicinity of the discharges included hi the evaluation.
• Ingestion rates of 6.5 grams per day for the general population, 30 grams per day
(30 years) + 6.5 grams per day (40 years) for sport fishermen, and 140 grams
per day for subsistence fishermen are used in the analysis of fish tissue (Exposure
Factors Handbook, U.S. EPA, 1989a)
• All rivers or estuaries within a State are equally fished by any of that state's
resident fishermen and the fish consumed only by the population within that State.
• Populations potentially exposed to discharges to rivers or estuaries that border
more than one State are estimated based only on populations within the State hi
which the facility is located.
• The size of the population potentially exposed to fish caught in. an impacted water
body in a given State is estimated based on the ratio of impacted river miles to
total river miles in that State or impacted estuary square miles to total estuary
square miles in that State. The number of miles potentially impacted by a
facility's discharge is assumed to be 50 miles for rivers and the total surface area
of the various estuarine zones for estuaries.
14
-------
Pollutant fate processes (e.g., sediment adsorption, volatilization, hydrolysis) are
not considered in estimating the concentration in drinking water or fish;
consequently, estimated concentrations are environmentally conservative (higher).
2.2 Projected Air Quality Imparts
Many of the chemicals released by pharmaceutical manufacturing facilities can exhibit
human health toxicity via the inhalation exposure route. A two-part approach is used to assess
environmental impacts from air emissions associated with pharmaceutical treatment options,
particularly those which include steam stripping. The first part assesses potential risks to the
general public from on-site fugitive emissions from open-air biological treatment using EPA's
Graphical Exposure Modeling System (GEMS) Atmospheric Modeling Subsystem (GAMS).
GAMS includes the Industrial Source Complex Long Term (ISCLT) air dispersion model linked
to site-specific weather and population data. The second part assesses potential risks to POTW
maintenance workers from occupational exposures to a toxic mixture of gases partitioning from
influent wastewater. The POTW occupational exposure analysis is based on the procedure
presented in Guidance to Protect POTW Workers from Toxic and Reactive Gases and Vapors
(U.S. EPA, 1992b).
2.2.1 Estimation of Human Health Risks and Benefits (Fugitive Air Emissions)
Pharmaceutical manufacturers use and release several volatile organic compounds that
exhibit carcinogenic and/or systemic health effects on humans and/or laboratory animals. In the
near-ground atmosphere, these chemicals may pose a threat to human health via .inhalation.
Inhalation exposures can be quantitatively assessed using air dispersion models, information on
the location and source of release, mass release amounts, and population density. For the
purposes of this analysis, the exposed population is assumed to be the general public living in
the vicinity of the point of release.
15
-------
Two sets of fugitive emissions from on-site treatment are examined:
• Data provided by industry that represent the "low estimate" of releases; and
• Data generated by EPA OST/EAD that represent the "maximum estimate" of
releases.
The "low estimate" data are compiled from responses to the CWA Section 308
Pharmaceutical Questionnaire. The "low estimate" benefits of steam stripping are estimated by
assuming that this process will remove all volatile pollutants from air emissions.
The "maximum estimate" data are compiled from engineering estimates (December 1994
loading data) based on the WATER? wastewater treatment model and a mass balance approach
to examining wastewater levels prior to treatment and current end of pipe releases. The
"maximum estimate" represents the upper boundary of potential air emissions of volatile organic
compounds from pharmaceutical manufacturing processes.
2.2.1.1 Preliminary Screening
In GAMS, site-specific air modeling is an iterative process implemented on a
facility-specific, pollutant-specific basis. A screening procedure is used to eliminate facility-
pollutant release combinations which result in potential exposures that are small compared to
their toxic effect level. The screening procedure involves calculating a hazard ratio (HAZ)
based on maximum potential concentration exposure. HAZ is the maximum potential downwind
concentration (MAX) divided by the lowest level of concern concentration (LOG) as follows:
HAZ =
( MAX \
~(wc)
(Eq. 13)
16
-------
Facility-pollutant release combinations where HAZ < 1.0 are dropped from further
analysis because the chemical concentration is highly unlikely to reach a level of concern at any
location in the vicinity of the facility. The remaining facility-pollutant release combinations
proceed to the next level of air modeling. The greater the value for HAZ the greater the
likelihood of harmful human health exposure.
The maximum potential downwind concentration is calculated using a Gaussian plume
dispersion equation for annual average concentration presented in the Workbook of Atmospheric
Dispersion Estimates (Turner, 1970). The maximum average annual downwind concentration
equation is:
MAX = f[Q) x
2.03 x Q x CF1 x CF2 [ -0.5 * (|
X x s x u
(Bq. 14)
where,
MAX = Maximum average annual downwind concentration (mg/nr*)
2.03 = Aggregation of constants (unitless)
flff) = Fraction of the year the wind is from direction 0 (unitless) =0.15
Q = Annual loading (Ibs/year)
CFi = Conversion factor of 4.53E+5 mg/lb
CF2 = Conversion factor of 3.17E-8 yr/sec
X = Downwind distance where maximum concentration occurs (m) = 40.55
s = Vertical dispersion coefficient (m) = 2.12
u = Mean wind speed (m/sec) = 5.5
H = Release height (m) = 3
The parameter input values are selected to achieve a maximum potential concentration
using reasonable assumptions for release height, fraction of the year wind blows from one
direction, and wind speed; arid further assuming stable atmospheric conditions, which will
dominate in the long-term. Use of this equation is conservative because it is intended to be
17
-------
applied to stack releases, which tend to have more concentrated plumes than the area source
releases considered here.
The LOG concentrations are compiled from the following sources:
• EPA unit risks (UR) for cancer at a 1CT6 risk level, or EPA reference
concentrations (RfC) from the Integrated Risk Information System (IRIS) or the
Health Effects Assessment Summary Tables (HEAST),
• American Council of Governmental and Industrial Hygienists (ACGffl) Threshold
Limit Values (TLV), and
• Occupational Safety and Health Administration (OSHA) permissible exposure
limits (PEL).
2.2.1.2 Atmospheric Dispersion Modeling
More complex atmospheric modeling analysis is performed on those facility-pollutant
releases identified in the screening procedure using HAZ scores. Site-specific modeling analysis
is used to predict potential atmospheric concentrations from fugitive releases and assess the
potential impact to the surrounding population.
The Industrial Source Complex Long Term (ISCLT) model is used in modeling
atmospheric dispersion. ISCLT is an EPA-supported gaussian plume air dispersion model that
is incorporated within the Office of Pollution Prevention and Toxics (OPPT) Graphical Exposure
Modeling System (GEMS) Atmospheric Modeling Subsystem (GAMS). In GAMS, the ISCLT
algorithms can run with site-specific atmospheric profile and U.S. Census population data inputs.
GAMS requires location identifiers such as latitude and longitude or ZIP code, and locates the
nearest STability ARray (STAR) weather data (usually airports). The STA.R data are used to
predict the pollutant concentration in 16 sectors around concentric rings surrounding the point
of release. These concentrations are then linked with U.S Census data at the block
group/enumeration district (BG/ED) level to estimate exposure levels and excess annual cancer
18
-------
cases. Additional information concerning the ISCLT model and the GAMS system may be
found in Industrial Source Complex (ISC) Dispersion Model User's Guide - Second Edition
(Revised) (U.S. EPA, 1987a), and GAMS Version 3,0 - User's Guide (U.S. EPA, 1990a).
Input data for ISCLT are obtained through user input (site- and pollutant-specific data)
or from GAMS via STAR stations (based on the user-supplied information). The GAMS menu
driven system allows the user to select EPA regulatory defaults or input-specific parameters for
detailed analysis. Latitude and longitude coordinates are obtained from either 308 Questionnaire
responses or the Toxic Release Inventory System (TRIS), and a standard polar receptor grid is
used. The receptor (breathing individual) is assumed to be at ground level, inhale 20 nrVday,
and weigh 70 kilograms (standard adult exposure factors).
Volatilization (fugitive emissions) from water treatment is modeled as an area source
release based on ISCLT equations (U.S. EPA, 1987a). The area of release is determined by
selecting parameters provided in the 308 Questionnaire based on the following hierarchy: (1)
the smallest equalization tank, (2) the smallest other tank (including stabilization and
neutralization), or (3) the average equalization tank reported in the 308 Questionnaire (4973 ft2).
GAMS requires input of side dimension only (in meters). Therefore, all treatment units are
assumed to be square.
2.2.1.3
Risk Calculations
ISCLT model results for each facility are run though the GAMS Exposure and Risk
Estimation (GAMSERE) procedure (U.S. EPA, 1990a) to generate Lifetime Average Daily
Doses (LADDs) associated with BG/ED populations and the number of excess cancer cases over
background levels. The LADD for inhalation used in the GAMSERE procedure is given as:
LADD = (CONC x CFx IK) f BW
(Eq. 15)
19
-------
where,
LADD
CONG
IR
CF
BW
potential lifetime average daily dose (mg/kg/day)
annual average concentration estimate (ug/m )
daily inhalation rate (20 m3/day)
conversion factor (0.001 mg//*g)
body weight (70 kg)
The LADD value is used to evaluate exposure for both systemic and carcinogenic
pollutants. Systemic pollutant LADD values are compared in GAMSERE to reference dose
values to estimate the population exposed to levels exceeding the reference dose. The
cumulative population exposed to greater than the reference dose are reported by BG/ED units.
Excess cancer risk over background is calculated using the a potency slope factor or a unit risk
factor. To estimate excess annual cancer risk, GAMSERE uses the following equation:
RISK = LADD x SF
(Eq. 16)
where,
RISK = lifetime excess risk over background
LADD = potential lifetime average daily dop (mg/kg/day)
SF = potency slope factor (mg/kg/day)"1
2.2.2
Estimation of POTW Occupational Risks and Benefits
Occupational exposure at POTWs is modeled based on the mixture of toxic vapors that
may partition out of the influent water into the surrounding air. POTW workers constitute the
potentially exposed population for this analysis. This analysis does not consider sewer workers
because they presumably will not be exposed to toxic vapors for long periods without utilizing
protective gear.
EPA developed guidance presented in Guidance to Protect POTW Workers from Toxic
and Reactive Gases and Vapors (U.S. EPA, 1992b) to screen industrial discharges for potential
20
-------
adverse effects on POTW workers. The general procedure for predicting the potential vapor
HAZARD associated with the discharge of a mixture of volatile organic compounds
(U.S. EPA, 1992b) includes the following steps:
1. Determine pollutant concentrations (mg/L) in wastewater.
2. Obtain 8 hour/day, 40 hours/week time weighted average ACGIH Threshold
Limit Values (TLV) in units of mg/m3 for pollutants.
3. Convert aqueous phase pollutant concentration (mg/L) to vapor phase pollutant
concentrations (mg/m3) in surrounding air using chemical-specific Henry's Law
Constants in the appropriate units as follows.
(Eq. 17)
where:
H
Vapor phase pollutant concentration (mg/m3)
Henry's Law Constant (mg/m3)/(mg/L)
Aqueous phase pollutant concentration (mg/L)
4. Calculate the hazard ratio for a given pollutant by dividing the threshold
concentration (i.e., TLV) identified in Step 2 by the predicted concentration from
Steps.
5. Sum the hazard ratios for all pollutants at the POTW.
6. Identify those POTW facilities with sum hazard ratios > 1, indicating potential
adverse health impacts.
This methodology assumes that equilibrium conditions exist, that Henry's Law Constant
is a good indicator of air-wastewater partitioning, and that the toxic effects indicated by TLVs
are additive across pollutants. In addition, there are a number of general assumptions based on
the definition of an "average worker". The "average worker" is assumed to weigh
70 kilograms, work 40 hours per week, and be in good health. POTW effluent flow, obtained
21
-------
from the NEEDS Survey, is used as a surrogate for influent flow to dilute pollutant load
estimates for the wastewater concentration calculation.
2.2.3 Assumptions and Caveats
The following assumptions are made in the air quality analysis. Many are included in
the previous text of Section 2.2 for necessary clarification and are listed in this section as a
summary.
• The maximum average annual downwind concentration equation (i.e., Eq. 14) is
assumed to be conservative as it is intended to be applied to stack releases, which
tend to have more concentrated plumes than the area source releases considered
in this analysis.
• If a treatment option that includes steam stripping is selected, significant fugitive
air emissions will not occur.
• The screening procedure equation to calculate maximum average annual
downwind concentration assumes the following parameter default values:
Fraction of the year the wind is from direction f(0) = 0.15;
Downwind distance for maximum concentration (X) == 40.55 m;
Vertical dispersion coefficient (s) = 2.12 m;
Mean wind speed (u) = 5.5 m/sec;
Release height (H) = 3 m.
• The exposed population is the general public living in the vicinity of the point of
release and encompassed by the standard polar grid generated in the GAMS
analysis.
• Atmospheric conditions used in the ISCLT model are based on long-term average
values and are, therefore, assumed to be stable under the assumptions used in the
model development.
• No chemical decay rate is employed in the GAMS model. It is assumed that at
these release amounts, dispersion will likely dilute chemical concentrations to
levels far below concern prior to any significant photooxidation or scavenging.
• The receptor (breathing individual) is assumed to be at ground level, inhale 20
m3/day, and weigh 70 kilograms (standard adult exposure factors).
22
-------
The screening procedure is assumed to be a reasonable method limiting detailed
analysis to those releases most likely to adversely effect the surrounding
population.
Open area for equalization tanks and stabilization ponds are provided by EPA's
308 questionnaire. Facilities without this data are assigned the average
equalization tank area for those facilities that did report tank areas.
In the POTW occupational exposure analysis, it is assumed that equilibrium
conditions exist, that Henry's Law Constant is a good indicator of air-wastewater
partitioning, and that the toxic effects indicated by TLVs are additive across
pollutants.
POTW effluent flow, obtained from the NEEDS Survey, is used as a surrogate
for influent flow to dilute pollutant load estimates for the wastewater
concentration calculation.
The POTW occupational exposure analysis (U.S. EPA, 1992b) assumes that an
"average worker":
is exposed to the pollutant throughout their occupational lifetime - ages 18
to 65;
works a 40-hour week;
weighs 70 kilograms;
is healthy, with no prior physical or health deficiencies; and
has a normal respiration rate.
The POTW occupational exposure analysis requires the following simplifying
assumptions for the use of Henry's Law Constant to conditions in the sewer line:
The wastewater and air temperatures at the POTW are approximately 25
degrees Celsius.
The presence of other constituents within the wastewater have no
synergistic or antagonistic effect on the volatilization of any given
pollutant.
The screening approach assumes that the air flow above the POTW unit
operations is negligible, thereby, allowing equilibrium conditions to be
approximated in the headspace above the unit operations.
The screening approach assumes instantaneous attainment of vapor-liquid
equilibrium and does not consider volatilization rates.
23
-------
The POTW occupational exposure analysis assumes that the toxic effects of the
pollutants in the mixture are additive. Therefore, hazard quotients are also
additive when calculating the POTW exposure.
2.3 Documented Environmental Impacts
Published literature and survey data have been reviewed for evidence of documented
environmental impacts on aquatic life, human health, POTW operations, and the quality of
receiving water due to discharges of pollutants from pharmaceutical manufacturing. Reported
impacts on the environment and biota/effect are compiled and summarized for various studies
and are presented by study site and facility.
24
-------
3. DATA SOURCES
3.1 Water Quality Impacts
Readily available EPA and other agency databases, models and reports are used in the
evaluation of water quality impacts. The following four sections describe the various data
sources used in the analysis.
3.1.1 Facility-Specific Data
Projected pharmaceutical facility effluent process flows and distillation pollutant loadings
(Appendix A) are obtained from the Engineering and Analysis Division (BAD) (March/April
1994). Pollutant loadings based on steam stripping were received from BAD in December 1994.
The locations of pharmaceutical manufacturing facilities on receiving streams are
identified using USGS cataloging and stream segment (reach) numbers contained in EPA's
Industrial Facilities Discharge (IFD) data base. Latitude/longitude coordinates, if available, are
used to locate those facilities and POTWs that have not been assigned a reach number in IFD.
The names, locations, and the flow data for the POTWs to which the indirect facilities discharge
are obtained from the Pharmaceutical 308 Questionnaire, EPA's 1992 NEEDS Survey, IFD, and
EPA's Permit Compliance System (PCS).
The receiving stream flow data are obtained from either the W.E. Gates study data or
from measured streamflow data, both of which are contained in EPA's GAGE file. The W.E.
Gates study contains calculated average and low flow statistics based on the best available flow
data and on drainage areas for reaches throughout the United States. The GAGE file also
includes average and low flow statistics based on measured data from USGS gaging stations.
,-
"Dissolved Concentration Potentials (DCPs)" for estuaries and bays are obtained from the
Strategic Assessment Branch of NOAA's Ocean Assessments Division (Appendix B). Critical
25
-------
Dilution Factors are obtained from the Mixing Zone Dilution Factors for New Chemical Exposure
Assessments (U.S. EPA, 1992a).
3.1.2 Information Used to Evaluate POTW Operations
POTW treatment efficiency removal rates are developed from POTW removal data and
pilot-plant studies or by using the removal rate of a similar pollutant when data are not available
(Appendix C). Use of the selected removal rates assumes that the evaluated POTWs are well-
operated and have at least secondary treatment in place. Sources of data include POTW Pass-
Through Analysis for the Pharmaceutical Industry (U.S. EPA, 1993a), EPA's Risk Reduction
Engineering Laboratory (RREL) Treatability Database, and the Development Document for
Effluent Limitations Guidelines and Standards for the Organic Chemicals, Plastics and Synthetic
Fibers Point Source Category (U.S. EPA, 1987b).
Inhibition values are obtained from Guidance Manual for Preventing Interference at
POTWs (U.S. EPA, 1987c) and from CERCLA Site Discharges to POTWs: Guidance Manual
(U.S. EPA, 1990b). The most conservative values for activated sludge are usied. For pollutants
with no specific inhibition value, a value based on compound type (e.g., aromatics) is used.
Sewage sludge regulatory levels, if available for the pollutants of concern, are obtained
from the Federal Register 40 CFR Part 503, Standards for the Use or Disposal of Sewage
Sludge, Final Rules (February 19, 1993). Pollutant limits established for the final use or
disposal of sewage sludge when the sewage sludge is applied to agricultural and non-agricultural
land are used (Appendix C). Sludge partition factors are obtained from the Report to Congress
on the Discharge of Hazardous Wastes to Publicly-Owned Treatment Works (Domestic Sewage
Study) (U.S. EPA, 1986).
26
-------
3.1.3 Water Quality Criteria (WQC)
The ambient criteria (or toxic effect levels) for the protection of aquatic life and human
health are obtained from a variety of sources including EPA criteria guidance documents, EPA's
Assessment Tools for the Evaluation of Risk (ASTER), and EPA's Integrated Risk Information
System (IRIS) (Appendix C). Ecological toxicity estimations are used when published values
are not available. The hierarchies used to select the appropriate aquatic life and human health
values are described in the following sections.
3.1.3.1 Aquatic Life
Water quality criteria guidance documents for many pollutants have been published by
EPA for the protection of freshwater aquatic life (acute and chronic criteria). States often
consult these guidance documents when adopting numeric water quality criteria as part of their
water quality standards. The acute value represents a maximum allowable 1-hour average
concentration of a pollutant at any time and can be related to acute toxic effects on aquatic life.
The chronic value represents the average allowable concentration of a toxic pollutant over a 4-
day period at which a diverse genera of aquatic organisms and their uses should not be
unacceptably affected, provided that these levels are not exceeded more than once every 3 years.
EPA has used EPA criteria guidance documents as one source for projecting potential
water quality impacts of the proposed rule. For pollutants for which no numeric water quality
criteria guidance has been published, specific toxicity values (acute and chronic effect
concentrations reported in published literature or estimated using various application techniques)
are used. In selecting values from the literature, measured concentrations from flow-through
studies under typical pH and temperature conditions are preferred. The test organism must be
a North American resident species of fish or invertebrate. The hierarchies used to select the
appropriate acute and chronic values are listed below in descending order of priority.
27
-------
Acute Aquatic Life Values:
• National acute freshwater quality criteria;
• Lowest reported acute test values (96-hour LC50 for fish and 48-hour
EC50/LC50 for daphnids);
• Lowest reported LC5Q test value of shorter duration, adjusted to estimate
a 96-hour exposure period;
• Lowest reported LC50 test value of longer duration, up to a maximum of
two weeks exposure; and
• Estimated 96-hour LC50 from the ASTER QSAR model.
Chronic Aquatic Life Values:
• National chronic freshwater quality criteria;
• Lowest reported maximum allowable toxic concentration (MATC), lowest
observable effect concentration (LOEC), or no observable effect
concentration (NOEC);
• Lowest reported chronic growth or reproductive toxicity test
concentration;
• Estimated chronic toxicity concentration from a measured acute chronic
ratio for a less sensitive species, quantitative structure activity relationship
(QSAR) model, or default acute:chronic ratio of 10:1.
3.1.3.2 Human Health
Water quality criteria for the protection of human health are established in terms of a
pollutant's toxic effects, including carcinogenic potential. These human health criteria values
are developed for two exposure routes: (1) ingesting the pollutant via contaminated aquatic
organisms only, and (2) ingesting the pollutant via both water and contaminated aquatic
organisms as follows.
28
-------
For Toxicitv Protection (ingestion of organisms only)
HH00 =
xCF
(Eq. 18)
where:
RfD
1%
BCF
CF
human health value 0*g/L)
reference dose for a 70-kg individual (mg/day)
fish ingestion rate (0.0065 kg/day)
bioconcentration factor (liters/kg)
conversion factor for units (1,000 pg/mg)
For Carcinogenic Protection (ingestion of organisms onlv)
„„ BWxRLxCF
tin =
00 SFxIRx BCF
(Eq. 19)
where:
BW
RL
SF
BCF
CF
human health value G*g/L)
body weight (70 kg)
risk level (10"6)
cancer slope factor (mg/kg/day)"1
fish ingestion rate (0.0065 kg/day)
bioconcentration factor (liters/kg)
conversion factor for units (1,000
For Toxicitv Protection (ingestion of water and organisms')
KfDxCF
(01.20)
where:
29
-------
HHWO
RfD
BCF
CF
human health value
reference dose for a 70-kg individual (mg/day)
water ingestion rate (2 liters/day)
fish ingestion rate (0.0065 kg/day)
bioconcentration factor (liters/kg)
conversion factor for units
For Carcinogenic Protection ftngestion of water and organisms)
=
BWxRLx CF
SF
IRf
x BCF
(Eq.21)
where:
EGETWO
BW
RL
SF
BCF
CF
human health value 0*g/L)
body weight (70 kg)
risk level (10-6)
cancer slope factor (mg/kg/day)"1
water ingestion rate (2 liters/day)
fish ingestion rate (0.0065 kg/day)
bioconcentration factor (liters/kg)
conversion factor for units (1,000 /tg/mg)
The values for ingesting water and organisms are derived by assuming an average daily ingestion
of 2 liters of water, an average daily fish consumption rate of 6.5 grams of potentially
contaminated fish products, and an average adult body weight of 70 kilograms (Technical
Support Document for Water Quality-Based Toxics Controls (U.S. EPA, 1991a). Values
protective of carcinogenicity are used to assess the potential effects on human health, if EPA has
established a slope factor.
Protective concentration levels for carcinogens are developed in terms of non-threshold
lifetime risk level. Criteria at a risk level of lO"6 are chosen for this analysis. This risk level
indicates a probability of one additional case of cancer for every 1,000,000 persons exposed.
30
-------
Toxic effects criteria for noncarcinogens include systemic effects (e.g., reproductive,
immunological, neurological, circulatory, or respiratory toxicity), organ-specific toxicity,
developmental toxicity, mutagenesis, and lethality.
The hierarchy used to select the most appropriate human health criteria values is listed
below in descending order of priority:
Calculated human health criteria values using EPA's Integrated Risk Information
System (IRIS) reference doses (RfDs) or slope factors (SFs) used in conjunction
with adjusted 3 percent lipid BCF values derived from Ambient Water Quality
Criteria Documents (U.S. EPA, 1980); three percent is the mean lipid content of
fish tissue reported in the study from which the average daily fish consumption
rate of 6.5g/day was derived;
Calculated human health criteria values using current IRIS RfDs or SFs and
representative BCF values for common North American species of fish or
invertebrates or estimated BCF values;
Calculated human health criteria values using RfDs or SFs from EPA's Health
Effects Assessment Summary Tables (HEAST) used in conjunction with adjusted
3 percent lipid BCF values derived from Ambient Water Quality Criteria
Documents (U.S. EPA, 1980);
Calculated human health criteria values using current RfDs or SFs from HEAST
and representative BCF values for common North American species of fish or
invertebrates or estimated BCF values;
Criteria from th& Ambient Water Quality Criteria Documents (U.S. EPA, 1980);
and
Calculated human health values using RfDs or SFs from data sources other than
IRIS.
This hierarchy is based on Section 2.4.6 of the Technical Support Document for Water
Quality-based Toxics Control (U.S. EPA, 1991A), which recommends using the most current
risk information from IRIS when estimating human health risks. In cases where chemicals have
both RfDs and SFs from the same level of the hierarchy, human health values are calculated
31
-------
using the formulas for carcinogenicity, which always results in the more stringent value of the
two given the risk levels employed.
3.1.4 Information Used to Evaluate Human Health Risks and Benefits
Fish ingestion rates for sport fishermen, subsistence fishermen and the general population
are obtained from the Exposure Factors Handbook (U.S. EPA, 1989a). State population data
and average household size are obtained from the 1992 Statistical Abstract of the United States
(U.S. Bureau of the Census, 1992). Data concerning the number of fishermen in each state
(i.e., resident fishermen) are obtained from the 1985 National Survey of Fishing, Hunting, and
Wildlife Associated Recreation (U.S. FWS, 1985). The total number of river miles or estuary
square miles within a state is obtained from the 1990 National Water Quality Inventory - Report
to Congress (U.S. EPA, 1990c). Drinking water utilities located within 50 miles downstream
from each discharge site are identified using EPA's PATHSCAN. The population served by a
drinking water utility is obtained from EPA's Drinking Water Supply Files or Federal Reporting
Data System. Information used in the evaluation is presented in Appendix D.
3.2 Air Quality Impacts
The analyses of air quality impacts require information pertaining to the individual
pharmaceutical manufacturing facility, the receiving POTW if indirect dischargers are occurring,
the exposed population and long-term average atmospheric conditions in the vicinity of each
facility, and the chemicals present in the wastestream. Specific information is obtained from
published EPA guidance documents or quality controlled data bases maintained by EPA
headquarters program offices, if available. Other data sources include documents or data bases
produced or maintained by other Federal agencies, peer reviewed literature, and secondary
sources cited in appropriate EPA documents.
32
-------
3.2.1 Facility-Specific Data
Information pertaining to an individual facility includes annual chemical loads, latitude
and longitude coordinates, and side dimensions and elevation of on-site biological treatment units
(Appendix E). For the analysis of fugitive emissions from open air biological treatment, two
sets of annual chemical loads are examined. The Engineering and Analysis Division (BAD)
provided electronic files of the "low estimate" loading data, which is based on CWA Section 308
Questionnaire responses (December 1994). BAD generated the "maximum estimate" loading
data based on conservative assumptions using a wastewater treatment model (WATER?) and a
mass balance (December 1994). The "maximum estimate" loading data are engineering
estimates representing the upper bound for potential air emissions of volatile organic compounds.
Latitude and longitude coordinates are obtained from Section 308 Questionnaire responses or
from the Toxic Release Inventory (TRI) data base maintained by the Office of Pollution
Prevention and Toxics (OPPT). A single side dimension (in meters) of the on-site equalization
tank was taken from surface area values provided in the Section 308 Questionnaire responses.
The elevation (used as the fugitive emission release height) is assumed to be 3 meters in all
cases. OPPT also makes this assumption in performing screening level exposure assessments
conducted under Toxic Substances Control Act (TSCA) authority.
For the analysis of toxic vapor partitioning at POTWs, the concentration of chemical
constituents in wastewater transferred to each POTW and the POTW influent flow is required.
Chemical concentrations are calculated from annual indirect load and effluent flow data provided
by BAD (April 1994). Total POTW influent flow is obtained from the 1992 NEEDS Survey.
3.2.2 Population and Climatologic Data
Factors needed to help determine the potential extent and magnitude of exposure from
fugitive releases include population and long-term average atmospheric conditions, as well as
the assumed characteristics of exposed persons. The spatial population distribution surrounding
a given set of latitude and longitude coordinates is available from 1980 U.S. Census data
33
-------
incorporated in the GEMS mainframe computer modeling system. GEMS also contains
information on long-term average wind speed, wind direction frequency, atmospheric stability,
and temperature needed to run the ISCLT model. An average adult body weight of 70
kilograms, an average adult inhalation rate of 20 m3 per day, and an average lifetime of 70
years are used to represent all exposed persons. These values are reported in the Exposure
Factors Handbook (U.S. EPA, 1989a)
3.2.3 Information Used to Evaluate Human Health Risks and Benefits
Toxicity assessment for the fugitive emission analysis is based on the chemical-specific
ambient reference concentration (RfC) for noncarcinogenic effects, and slope factor (SF) for
carcinogenic effects (Appendix E). RfCs and SFs are obtained first from IRIS, and secondarily
from the 1992 HEAST. The RfC is an estimate of a daily exposure level (adjusted to
concentration units by assuming a 70 kilogram body weight and a 20 m3 per day inhalation rate)
for the human population, including sensitive subpopulations, that is likely to be without an
appreciable risk of deleterious noncarcinogenic health effects over a lifetime as defined in Risk
Assessment Guidance for Superfund (U.S. EPA, 1989b). EPA recommends a threshold level
assessment approach because several protective mechanisms must be overcome prior to the
appearance of an adverse noncarcinogenic effect. EPA assumes that cancer growth can be
initiated from a single cellular event, and, therefore, should not be subject to a threshold level
assessment approach. The SF is an upper bound estimate of the probability of cancer per unit
intake of a chemical over a lifetime (U.S. EPA, 1989b). The Office of Water recommends use
of a 10"6 (one excess cancer case per 1,000,000 individuals) acceptable cancer probability in its
Technical Support Document for Water Quality-Based Toxics Control (U.S. EPA, 1991a).
For the POTW toxic vapor partitioning analysis, chemical-specific data include exposure
guidelines/standards for occupational exposure and air-water partitioning coefficients
(Appendix E). The American Conference of Governmental Industrial Hygienists (ACGffl)
threshold limit values (TLVs), derived for 8-hour day 40-hour week exposure, are used in the
34
-------
hazard screening procedure. TLVs are obtained from the Threshold Limit Values for Chemical
Substances and Physical Agents and Biological Exposure Indices (U.S. EPA, ACGIH, 1990d).
Henry's Law Constant (HLC), the measured or estimated ratio of vapor pressure to solubility,
is used as the air-water partition coefficient. Most HLCs are obtained from the Toxic Chemical
Release Inventory Screening Guide (U.S. EPA, 1989c), the Superjund Chemical Data Matrix
(U.S. EPA, 199Ib), or the quantitative structure activity relationship (QSAR) system maintained
by EPA's Environmental Research Laboratory in Duluth, MN.
3.3 Literature Review
Literature abstracts are obtained through the computerized information system DIALOG
which provides access to Enviroline, Pollution Abstracts, Aquatic Science Abstracts, and Water
Resources Abstracts. Data are also obtained from the 1990/1992 State Water Quality
Assessments (305(b)) Reports, the Pharmaceutical Outreach Questionnaire (U.S. EPA, 1993b),
newspaper articles (Washington Post, Baltimore Evening Sun), and the 1990 State 304(1) short
lists.
35
-------
-------
4. SUMMARY OF RESULTS
4.1 Projected Water Quality Impacts
4.1.1 Comparison of Instream Concentrations with Ambient Water Quality Criteria"
The results of this analysis indicate the water quality benefits of controlling discharges
from pharmaceutical manufacturing facilities to surface waters and POTWs. The following two
sections summarize potential aquatic life and human health impacts on receiving stream water
quality and on POTW operations and their receiving streams for AC and BD direct and indirect
discharges7. All tables referred to in these sections are presented at the end of Section 4.
Appendices F, G, and H present the results of the stream modeling for each type of discharge,
respectively.
4.1.1.1 Direct Discharges
(a) AC Facilities
The effects of direct wastewater discharges on receiving stream water quality are
evaluated at current and steam stripping treatment levels, for 14 facilities discharging 43
pollutants to 14 receiving streams (14 rivers) (Table 1).
performing this analysis, EPA used guidance documents published by EPA that
recommend numeric human health and aquatic life water quality criteria for numerous pollutants.
States often consult these guidance documents when adopting water quality criteria as part of
their water-quality standards. However, because those State adopted criteria may vary, EPA
used the nationwide criteria guidance as the most representative values.
7Three pollutants have been removed as candidates for regulation since this analysis was
completed (glycol ethers, bis(chloromethyl) ether and dimethylcarbamyl chloride). In addition,
pollutant loadings at steam stripping treatment levels have been received and modeled since this
analysis was initially completed using distillation pollutant loadings. Because the projected
loads from these two treatment technologies are roughly comparable, the initial analysis was
rerun using the steam stripping pollutant loads only for those facilities for which aquatic life
or human health criteria excursions or POTW inhibition problems were projected at current
loadings.
36
-------
Pollutant loadings for 14 facilities at current discharge levels are 5.13 million pounds-
per- year (Table 2). These loadings are reduced to 0.24 million pounds-per-year at distillation
levels; a reduction of 95 percent. Since steam stripping is a less effective treatment technology,
a lower reduction in pollutant loadings would occur at steam stripping discharge levels.
Instream pollutant concentrations are projected to exceed human health criteria or toxic
effect levels (developed for water and organisms consumption) in 14 percent (2 of the total 14)
of the receiving streams at current discharge levels (Table 3). A total of 3 pollutants at current
are projected to exceed instream criteria or toxic effect levels using a target risk of KT6 for
carcinogens (Table 4). No excursions of human health criteria or toxic effect levels are
projected at steam stripping discharge levels (Table 3).
Instream pollutant concentrations are projected to exceed chronic aquatic life criteria
or toxic effect levels in 7 percent (1 of the total 14) of the receiving streams at current
discharge levels (Table 3). A total of 2 pollutants at current are projected to exceed instream
criteria or toxic effect levels (Table 4). No excursions of chronic aquatic life criteria or toxic
effect levels are projected at steam stripping discharge levels (Table 3).
Excursions of human health criteria or toxic effect levels (developed for organisms
consumption only) and of acute aquatic life criteria or toxic effect levels are also presented in
Table 3. A similar reduction in the number of pollutants exceeding criteria is noted.
(b) BD Facilities
The effects of direct wastewater discharges on receiving stream water quality are
evaluated at current and steam stripping treatment levels for 3 facilities discharging 6 pollutants
to 3 receiving streams (3 rivers) (Table 5).
37
-------
Pollutant loadings for 3 facilities at current discharge levels are 15,780 pounds-per-year
(Table 6). These loadings are reduced to 620 pounds-per-year at distillation levels; a reduction
of 96 percent. Since steam stripping is a less effective treatment technology, a lower reduction
in pollutant loadings would occur at steam stripping discharge levels.
No excursions of human health criteria or toxic effect levels or of aquatic life criteria
or toxic effect levels are projected at current or steam stripping discharge levels (Table 7).
4.1.1.2 Indirect Discharges
(a) AC Facilities
The effects of POTW wastewater discharges of 54 pollutants on receiving stream water
quality are evaluated at current and steam stripping discharge levels, for 61 facilities, which
discharge to 43 POTWs on 42 receiving streams (35 rivers and 7 estuaries) (Table 8).
Pollutant loadings for 61 facilities at current discharge levels are 29.98 million pounds-
per-year (Table 2). The loadings are reduced to 3.44 million pounds-per-year after distillation
pretreatment: a reduction of 88 percent. Since steam stripping is a less effective treatment
technology, a lower reduction in pollutant loadings would occur after steam stripping
pretreatment.
Instream pollutant concentrations are projected to exceed human health criteria or toxic
effect levels (developed for water and organisms consumption) in 10 percent (4 of the total 42)
of the receiving streams at current discharge levels (Table 9). A total of 4 pollutants at current
are projected to exceed instream criteria or toxic effect levels using a target risk of 10"6 for the
carcinogens (Table 10). Instream concentrations of 1 pollutant are projected to exceed human
health criteria or toxic effect levels in 5 percent (2 of the total 42) of the receiving streams at
steam stripping pretreatment discharge levels.
38
-------
Instream pollutant concentrations are not projected to exceed chronic aquatic life criteria
or toxic effect levels at current or steam stripping pretreatment discharge levels (Table 9).
No excursions of human health criteria or toxic effect levels (developed for organisms
consumption only) and of acute aquatic life criteria or toxic effect levels are projected
(Table 9).
In addition, the potential impacts of 65 facilities, which discharge to 46 POTWs, are
evaluated in terms of inhibition of POTW operation and contamination of sludge. Fifty-one (51)
pollutants are evaluated for potential POTW operation inhibition. No pollutants are evaluated
for potential sludge contamination problems since EPA sludge criteria are not available for any
of the pollutants of concern. At current discharge levels, inhibition problems are projected to
occur at 13 percent (6 of the 46) of the POTWs for 7 pollutants (Tables 11 and 12). Inhibition
problems are reduced after steam stripping pretreatment to 3 pollutants at 11 percent (5 of the
46) of the POTWs.
(b) BD Faculties
The effects of POTW wastewater discharges of 23 pollutants on receiving stream water
quality are evaluated at current and steam stripping pretreatment discharge levels, for 55
facilities, which discharge to 45 POTWs on 45 receiving streams (31 rivers and 14 estuaries)
(Table 13).
Pollutant loadings for 55 facilities at current discharge levels are 1.58 million pounds-
per-year (Table 6). The loadings are reduced to 0.03 million pounds-per-year after distillation
pretreatment: a reduction of 98 percent. Since steam stripping is a less effective treatment
technology, a lower reduction in pollutant loadings would occur after steam stripping
pretreatment.
39
-------
Instream pollutant concentrations are projected to exceed human health criteria or toxic
effect levels (developed for water and organisms consumption) in 2 percent (1 of the total 45)
of the receiving streams at current discharge levels** (Table 14). A total of 1 pollutant at
current is projected to exceed instream criteria or toxic effect levels using a target risk of 10"6
for the carcinogens (Table 15). No excurions of human health criteria or toxic effect levels
are projected at steam stripping pretreatment discharge levels (Table 14).
Instream pollutant concentrations are projected to exceed chronic aquatic life criteria
or toxic effect levels in 2 percent (1 of the total 45) of the receiving streams at current
discharge levels (Table 14). A total of 1 pollutant at current are projected to exceed instream
criteria or toxic effect levels (Table 15). No excursions of chronic aquatic life criteria or toxic
effect levels are projected at steam stripping pretreatment discharge levels.
Excursions of human health criteria or toxic effect levels (developed for organism
consumption only) and of acute aquatic life criteria or toxic effect levels are also presented in
Table 14.
In addition, the potential impacts of 61 facilities, which discharge to 50 POTWs, are
evaluated in terms of inhibition of POTW operation and contamination of sludge. Twenty-three
(23) pollutants are evaluated for potential POTW operation inhibition. No sludge criteria are
available to evaluate potential sludge contamination problems. No inhibition problems are
projected to occur at current or steam stripping pretreatment discharge levels (Table 16).
°One additional human health excursion for bis(chloromethyl) ether is projected at current
and steam stripping pretreatment discharge levels; however, bis(chloromethyl) ether
hydrolyses very rapidly and is not expected to pose a threat to human health.
40
-------
4.1.2 Estimation of Human Health Risks and Benefits
The results of this analysis indicate the potential benefits to human health by estimating
the risks (carcinogenic and systemic effects) associated with current and reduced pollutant levels
in fish tissue and drinking water. The following two sections summarize potential human health
impacts from the consumption of fish tissue and drinking water derived from water bodies
impacted by AC and BD direct and indirect discharges. Appendices I and J present the results
of the modeling for each type of discharge, respectively.
4.1.2.1 Direct Discharges
The effects of direct wastewater discharges on human health from the consumption of fish
tissue and drinking water are evaluated at current and distillation treatment levels9 for 17
AC/BD facilities discharging 44 pollutants to 17 receiving streams (17 rivers) (Tables 1 and 5).
(a) fish Tissue
At current discharge levels, 2 streams, receiving the discharge of 4 carcinogens from
2 facilities, have total estimated individual pollutant cancer risks greater than 10"6 (Tables 17
and 18). Total estimated risks (greater than 10*) range from 2.1E-6 to 7.5E-6 for subsistence
fishermen. Total risks greater than 10"6 are not projected for the general population or snort
fishermen. Total excess annual cancer cases are estimated to be 3.8E-5 for subsistence
fishermen. No systemic toxicant effects (hazard index greater than 1.0) are projected. At
distillation discharge levels, no total estimated individual pollutant cancer risks greater than 1O6
or systemic toxicant effects are projected for the general population, sport fishermen or
ynbsistence fishermen (Tables 17 and 18).
9PoUutant loadings at steam stripping treatment levels have been received since this analysis
was initially completed. Since no substantial risks are projected at current, the highest
wasteload, none are projected for steam stripping.
41
-------
(b) Drinking Water
At current discharge levels, 2 streams, receiving the discharge of 4 carcinogens from
2 facilities, have total estimated individual pollutant cancer risks greater than 10"6 (Tables 19
and 20). Estimated risks range from 2.5E-5 to 4.8E-5. However, there is no drinking water
utility located within 50 miles downstream of either discharge site. No systemic toxicant effects
(hazard index greater than 1.0) are projected (Table 19).
i
At distillation discharge levels, no streams are projected to have a total estimated
individual pollutant cancer risk greater than 10"6 (Tables 19 and 20). No systemic toxicant
effects (hazard index greater than 1.0) are projected (Table 19).
4.1.2.2 Indirect Discharges
The effects of POTW wastewater discharges on human health from the consumption of
fish tissue and drinking water are evaluated at current and distillation discharge levels10 for
116 AC/BD facilities that discharge 55 pollutants to 77 POTWs on 74 receiving streams (59
rivers and 15 estuaries) (Tables 8 and 13).
(a) Fish Tissue
At current discharge levels, 1 stream, receiving the discharge from 4 facilities, has a
total estimated individual pollutant cancer risk greater than 10"6 from 5 carcinogens (Tables 21
and 22). Total estimated risk is 3.8E-6 for subsistence fishermen. Total risks greater than 10"6
are not projected for the general population or sport fishermen. Total excess annual cancer
10Pollutant loadings at steam stripping pretreatment levels have been received since this
analysis was initially completed. Since no substantial risks are projected at current, the highest
wasteload, none are projected for steam stripping.
42
-------
cases are estimated at 2.4E-5 for subsistence fishermen. No systemic toxicant effects (hazard
index greater than 1.0) are projected (Table 21).
At distillation pretreatment discharge levels, no streams are projected to have total
estimated individual pollutant cancer risks greater than 10"6 (Tables 21 and 22). No systemic
toxicant (hazard index greater than 1.0) are projected (Table 21).
(b) Drinking Water
At current discharge levels, 6 streams, receiving the discharge of 6 carcinogens from
10 facilities, have total estimated individual pollutant cancer risks greater than 10"6 (Tables 23
and 24). Estimated risks range from 1.3E-6 to 1.8E-5. Two (2) streams, receiving the
discharge of 3 carcinogens from 2 facilities, have drinking water utilities located within 50 miles
downstream. The total estimated individual pollutant cancer risks range from 1.3E-6 to 1.9E-6.
The total number of estimated annual excess cancer cases is 2.2E-3. No systemic toxicant
effects (hazard index greater than 1.0) are projected (Table 23).
At distillation pretreatment discharge levels, 1 stream, receiving the discharge of 2
carcinogens from 1 facility, has a total estimated individual pollutant cancer risk greater than
10"6 (Tables 23 and 24). The total estimated individual pollutant cancer risk is 1.5E-7.
However, there is no drinking water utility located within 50 miles downstream of discharge
site. Thus, annual excess cancer cases cannot be projected. No systemic toxicant effects
(hazard index greater than 1.0) are projected (Table 23).
4.2 Projected Air Quality Impacts
The results of this analysis indicate the potential air quality risks and benefits of
controlling fugitive air emission discharges from AC direct and AC and BD indirect discharging
pharmaceutical manufacturing facilities. Projected occupational exposure benefits are presented
for the control of discharges to POTWs. The following two sections summarize potential human
43
-------
health risks resulting from fugitive emissions of volatile organics originating from
pharmaceutical manufacturing wastewater at the "low estimate" and "maximum estimate"
loadings. The third section presents occupational exposure benefits of controlling discharges
from pharmaceutical manufacturing facilities to POTWs.
4.2.1 Comparison of Human Health Risks and Benefits (Fugitive Air Emissions)
The potential air quality benefits of controlling fugitive air emissions from AC direct and
AC and BD indirect discharging pharmaceutical manufacturing facilities are presented below.
The "low estimate" and "maximum estimate" loads are presented separately.
The results of the air quality benefits are presented based on modeling a subset of the
overall loading data (Appendix K). The subset is defined using a screening method to rank
facility-pollutant releases based on the maximum potential downwind concentration and pollutant
levels of concern (Appendix K). Facilities with screening hazard ratios above 1.0 are selected
for site-specific analysis. The screening procedure significantly reduces the number of
facility-pollutant release combinations that are modeled.
4.2.1.1 "Low Estimate"
The preliminary screening method evaluates 45 facilities discharging 50 pollutants via
fugitive emissions. There are 288 facility-pollutant loads, releasing 7.08 million pounds of
pollutants per year. Under the assumptions listed in the methodology section, the benefit
associated with the control of fugitive emissions is based on loading values from EPA's 308
questionnaire survey.
The screening procedure identifies 22 facility-pollutant discharges with hazard ratios
greater than 1.0. These releases represent 10 pollutants from 14 facilities at a benefit load of
approximately 2.5 million pounds-per-year. Atmospheric modeling includes 17 facility-pollutant
44
-------
releases of 6 carcinogens. Four additional facility-pollutant releases of 4 non-carcinogenic
pollutants were analyzed.
Based on the "low estimate" load benefits, 190,220 people nationwide are projected to
be exposed to risk levels exceeding W6 (Table 25). Projected "low estimate" load benefits
include 0.023 excess annual cancer occurrences. Chloroform has the largest impact of any
single chemical based on the "low estimate" load benefits. The GAMS air modeling analysis
did not project any population exposed to ambient levels above the threshold for non-
carcinogenic effects.
Three pollutants; bis(chloromethyl) ether, dimethyl carbamyl chloride, and glycol ethers,
are not included in the air quality modeling analysis. These pollutants are no longer considered
for regulation in the pharmaceutical manufacturing industry.
4.2.1.2 "Maximum Estimate"
The preliminary screening method evaluated 84 facilities discharging 52 pollutants via
fugitive emissions. There are 453 facility-pollutant loads, generating a maximum potential
benefit of 38 million pounds of pollutants reduced per year.
The screening procedure identified 43 facility-pollutant scores with hazard ratios greater
than 1.0. These air modeling applications include 23 facilities with a maximum benefit loading
of approximately 11.6 million pounds per year for 12 pollutants. These include 27 carcinogenic
and 16 systemic releases of 6 carcinogenic and 6 non-carcinogenic pollutants, respectively.
The "maximum estimate" air quality modeling analysis projects 2.4 million people (1990
population) at cancer risk levels exceeding 10"6 would benefit from the air load reduction
(Table 26). The load reduction would provide a benefit of 0.35 reduced annual cancer case
occurrences. Chloroform and methylene chloride cause the greatest potential carcinogenic
effects based on the "maximum estimate" loads. In addition, the air modeling analysis projects
45
-------
that approximately 126 thousand people would benefit from the reduced exposure to chemicals
associated with non-carcinogenic effects, principally triethylamine.
Three pollutants (bis(chloromethyl) ether, dimethyl carbomyl chloride, and glycol ethers)
are not included in the summary for the "maximum estimate" analysis. These pollutants are not
currently included in the proposed regulation.
4.2.2 Comparison of POTW Occupational Risks and Benefits
Pollutant loadings at current discharge levels for 129 pharmaceutical manufacturing
facilities discharging a total of 32.6 million pounds-per-year to 85 POTWs are analyzed. The
loadings are reduced to 3.47 million pounds-per-year at the full distillation pretreatment. This
represents a 89 percent decrease. Since $fa«m gripping is a less effective treatment technology,
a lower reduction in pollutant loadings would occur at steam stripping pretreatment.
A total of 12 POTWs treating 52 pollutants are identified with summed hazard ratios
greater than 1.0 at the current discharge levels. Individual pollutant hazard ratios ranged from
2.6E-10 to 2,530 at the current discharge levels, with 23 occurrences of 10 pollutants
exceeding the hazard ratio of 1.0 at the same POTW (Table 27). Benzene is associated with the
greatest risk to POTW workers.
A total of 7 POTWs treating 44 pollutants are identified with summed hazard ratios
greater than 1.0 at full distillation pretreatment discharge levels. Individual pollutant hazard
ratios ranged from 0 to 25.8 with 6 occurrences of 3 pollutants exceeding the hazard ratio of 1.0
(Table 27). At steam stripping pretreatment discharge levels, which are slightly higher than
distillation levels, 6 to 12 POTWs treating 52 pollutants are projected with summed hazard
ratios greater than 1.0.
46
-------
4.3 Documented Environmental Impacts
In a review of literature abstracts, State 305(b) reports, newspaper articles, and the
Pharmaceutical Outreach Questionnaire, 16 studies noted environmental impacts from
pharmaceutical manufacturing (Table 28). Impacts included: (1) human health problems (worker
exposure and population) such as dizziness, nausea, respiratory and dermal problems and
endocrine dysfunction (reproductive); (2) aquatic life effects, such as fish kills; (3) effects on
the quality of receiving waters, groundwater, soils, sediments, and drinking water; and
(4) impairments to POTW operations. In addition, 4 pharmaceutical manufacturing facilities are
identified by States as being point sources causing water quality problems and are included on
their 304® Short List. Section 304® of the Water Quality Act of 1987 requires States to
identify waterbodies impaired by the presence of toxic substances, to identify point source
discharges of these toxics, and to develop Individual Control Strategies (ICSs) for these
discharges. The Short List is a list of waters for which a State does not expect applicable water
quality standards (numeric or narrative) to be achieved after technology-based requirements have
been met due entirely or substantially to point source discharges of Section 307(a) toxics. A list
of facilities included on the 304® Short List are provided in Table 28.
47
-------
Table 1. Frequency of Pollutants from 14 AC Direct Pharmaceutical
Manufacturing Facilities Discharging to 14 Receiving Streams
Pollutant Name
Number of Detections by
Facility
ACETONE
ACETONITRILE
AMMONIUM HYDROXIDE
AMYL ALCOHOL
AMYL ACETATE, n-
BUTANOL, 1-
BUTYL ACETATE, n-
CHLOROFORM
CHLOROMETHANE
CYANIDE
CYCLOHEXANE
DICHLOROETHANE, 1,2-
DIETHYL ETHER
DIMETHYL SULFOXIDE
DIMETHYLACETAMIDE, N,N
DIMETHYLAMINE
DIMETHYLFORMAMIDE, N,N
ETHANOL
ETHYL ACETATE
ETHYLENE GLYCOL
FORMALDEHYDE
FORMAMIDE
FURFURAL
HEPTANE, n-
HEXANE, n-
ISOPROPANOL
ISOPROPYL ACETATE
ISOPROPYL ETHER
METHANOL
METHYL ETHYL KETONE
METHYL FORMATE
METHYL ISOBUTYL KETONE
METHYLAMINE
METHYLENE CHLORIDE
METHYLPYRIDINE, 2-
PHENOL
PROPANOL, 1-
PYRIDINE
TERT-BUTYL ALCOHOL
TETRAHYDROFURAN
TOLUENE
TRIETHYLAMINE
XYLENES
8
3
6
1
2
1
1
4
2
4
2
3
3
1
1
1
2
6
5
2
5
1
2
1
3
8
1
2
7
2
2
1
1
5
1
1
1
1
1
3
6
1
4
Note: Only pollutants proposed for regulation are evaluated. (Three pollutants have been removed
as candidates for regulation since this analysis was completed - glycol ethers, bis(chloromethyl)
ether and dimethylcarbamyl chloride).
Source: Engineering and Analysis Division (BAD), March 1994.
48
-------
Table 2. Summary of Pollutant Loadings for AC Direct and Indirect Pharmaceutical Manufacturers*
Current
Organics
Cyanide
Ammonium Hydroxide
Total
Distillation***
Organics
Cyanide
Ammonium Hydroxide
Total
No. of Pollutants
No. of Facilities fEvaluated)
_:, loadings, pourids-per-yeai'
Direct
4,314,840
45
817.732
5,132,617
167,320
6
71.167
238,493
43
14
Indirect
29,452,484
1,083
526.370
29,979,937
3,349,795
62
88.514
3,438,371
54
61
„! Total
33,767,324
1,128
1T344.102
35,112,554
3,517,115
68
159.681
3,676,864
**58
75
**
***
Only pollutants proposed for regulation are evaluated. (Three Pollutants have been removed as
candidates for regulation since this analysis was completed - glycol ethers, bis(chloromethyl) ether
and dimethylcarbamyl chloride).
The same pollutant may be discharged from a number of facilities; therefore, the total does not
equal the sum of pollutants.
Analysis not conducted for steam stripping.
49
-------
co
3
E?
CIS
•s
CO
5
•3
.H
'"§
o>
ej
CC
1
eu
4— *
o
2
S
3
Ui
<2
s«
g
o
"co
l_l
3
X
w
.5
w
s
s
1
8
o
£
tl_r
0
>,
1
3
CO
CO
'^ -1S
JC
U
o
"'ii ' '
§*«>
^ a
s
O
X>
CT3
H
CO ^> ^5 ^^
o o o o o o
o"
c-i
CO
0
VC5
C^ CO fO C? ^5 ^5
o"
a.-
— < CS M O O O
Co^
CO,
»-< CS P* O O O
^ o ea» '"T- o
-^ !z '£ -S ^ 3 *g
6 S- 3 § 6 » 3
5- § 8 • | S a S
**^ £4 KJ ^^ ^(G CM CO ^^
2§i| i«i|
r3co«£H JcoiStS
vJ 1/2
D
.s
to
' S3
0
CS
a
O)
Uc
5
O ^3
£3 O
2t
i ^
•° §
" a 1
"S ^ -o
•c3 'S «
0) a>
t^t
.« -- "^
2 *§ is "S
jfjl
•o 1 2^
B S s S
S3 *•*
fj »-rt •N **
§) S .2 2
5 ^s 2
ZH *P ^
S V 2*3
1 'S o ^
^ +s ^s o
G,tS | ^
11 It
II If
^ f
s
o
i^r
2
^tn
O
c
"§
&
i
C3
"3
^*
o
o<
o
1
I
•s
§
1
2
oT
«2
•1
co"
s
§
*o
«<-i
O
u,
•|
§
C8
O
"C
-tt
w
1
u
X
I
•J2
2
^
Oi
50
-------
2
S
C3
1
u«
u
S
7
P4
C3
00
CO
O
1
x
w
<4-l
O
t-t
H,
•-Si
o
•x.
C3 ^^ C^
-H o O -* Ol cN
S* S^ S*
^i' si' *^^
<~* o o •— < 01 CM
1 •'»'•!
-s
<* 1
•»
II £} 1 ^
II' «_/ 11 ua rM m oo >/T! S
i
I
BJ
I
e
•C
CO
§
3
a>
^
!
g
O)
1
CO
c
.2
12
3
§ S
^1
"3 2
1
CQ
O
«
•S3
II
S
-------
Table 5. Frequency of Pollutants from 3 BD Direct Pharmaceutical
Manufacturing Facilities Discharging to 3 Receiving Streams
Pollutant Name
Number of Detections by Facility
ACETONE
ETHANOL
FORMALDEHYDE
ISOPROPANOL
METHANOL
POLYETHYLENE GLYCOL 600
1
2
1
2
1
1
Note: Only pollutants proposed for regulation are evaluated. (Three pollutants have been removed as
candidates for regulation since this analysis was completed - glycol ethers, bis(chloromethyl) ether
and dimethylcarbamyl chloride).
Source: Engineering and Analysis Division (BAD), March 1994.
52
-------
Table 6. Summary of Pollutant Loadings for BD^Direct and Indirect
Pharmaceutical Manufacturers
Current
Organics
Cyanide
Ammonium Hydroxide
TOTAL
pistillation***
Organics
Cyanide
Ammonium Hydroxide
TOTAL
No. of Pollutants
No. of Facilities (Evaluated)
Loadings, poxmoVper-year <
, Direct
15,780
0
0
15,780
620
0
0
620
6
3
Indirect
1,576,114
0
146
1,576,260
30,348
0
25
30,373
23
55
Total
1,591,894
0
146
1,592,040
30,968
0
25
30,993
**23
58
**
Only pollutants proposed for regulation are evaluated. (Three pollutants have been
removed as candidates for regulation since this analysis was completed - glycol ethers,
bis(chloromethyl) ether and dimethylcarbamyl chloride).
The same pollutant may be discharged from a number of facilities; therefore, the total
does not equal the sum of pollutants.
*** Analysis not conducted for steam stripping.
53
-------
e
(W
Q
1
I
Q
m
CO
§
a
I
u
1
5^
I
s
t/5
3
JO
If
SCO
,
Vjl
<<£
e
o
t*
O O
O O
o o o
o o o
o o o
o o o
o o o
o o o
o o o o o
-
is "o o
55 a, H
o
s
I
2
^
3
S
8
Oi W
55 4)
'S'S «
0-13 S
s -
O co
II 2-
3- -g '
CO *^?
O «5
O -rs
J? co
^ g
S £
.
E >»^*
isi
54
-------
Table 8.
Frequency of Pollutants from 61 AC Indirect Pharmaceutical Manufacturing Facilities
Which Discharge to 43 POTWS on 42 Receiving Streams
Pollutant Name
Number of Detections by Facility
ACETONE
ACETONITRILE
AMMONIUM HYDROXIDE
AMYL ALCOHOL
AMYL ACETATE, n-
ANDLINE
BENZENE
BUTANOL, 1-
BUTYL ACETATE, n-
CHLOROBENZENE
CHLOROFORM
CHLOROMETHANE
CYANIDE
CYCLOHEXANE
DICHLOROBENZENE, 1,2-
DICHLOROETHANE, 1,2-
DIETHYL ETHER
DIETHYLAMINE
DIMETHYL SULFOXIDE
DIMETHYTLACETAMIDE, N,N
DIMETHYLANDLINE, N,N-
DIMETHYLCARBAMYL CHLORIDE
DIMETHYLFORMAMIDE, N,N
DIOXANE, 1,4-
ETHANOL
ETHYL ACETATE
ETHYLENEGLYCOL
ETHYLENE GLYCOL MONOETHYL ETHER
FORMALDEHYDE
FORMAMIDE
HEPTANE, n-
HEXANE, n-
ISOBUTYRALDEHYDE
ISOPROPANOL
ISOPROPYL ACETATE
ISOPROPYL ETHER
METHANOL
METHYL CELLOSOLVE
METHYL ETHYL KETONE
METHYL FORMATE
METHYL ISOBUTYL KETONE
METHYLAMINE
METHYLENE CHLORIDE
PETROLEUM NAPTHA
PHENOL
POLYETHYLENE GLYCOL 600
PROPANOL.l-
PYRIDINE
TERT-BUTYL ALCOHOL
TETRAHYDROFURAN
TOLUENE
TRICHLOROFLUOROMETHANE
TRIETHYLAMINE
XYLENES
Note- Only pollutants proposed for regulation are evaluated. (Three pollutants have been removed1 as candidates for regulation
since this analysis was completed - glycol ethers, bis(chloromethyl) ether and dimethylcarbamyl chloride).
Source: Engineering and Analysis Division (BAD), March 1994.
55
-------
CO
S3
8?
ea
•S
03
S
-a
o
•a
C3
S
3
Wl
i
*""
<
.2
2
*g
I
i
5
"•N -i *g .S ^ 5 "g
6^3 g e S- s
z-m -c ^ i g
- al« asglH
£Sa2 S»5«
fi is ^ .0 s?ii'oe
s ^ Oi H ^^fiH
12
M —
§ IS
tx U
^1
tl
• ?a o
»
11 g "§
^rt ^^ f3
1 'B -o
*c3 ^* S
> o>
O dj *-<
L^ flj
J 1 *S
lf!1
I si 1
llii
'S T3 S *
w> S .2 «T
i* "3 5
S ii 2 -$
^ ]§ <2 -^
S* ^ ~ ^°
"e3 M i
«> > O S
J | 2*4
Ilil
OH c^ cC co
.5^10
lilt
|*0.
.2
*U(
.S
O
*C5
^
33
o
s
I
3
a
s
1
«,
i
1
s
£^
«s
1
• «v
CO
o
S3
I"
03
e
o
"i
Vta
*Ui
0
"S
,
Pollutants ma
*
56
-------
If
O O O Ol O
•s
09
13
o
'•imt
I
£
I
-o
H-^
O
ca
*C
1
I
2
•8
o
O)
1
I
s
CO
o?
I
1
'IP
S>'$
Q
\0
O
o
8
8
a
O
O O O Ol
O O O Ol
Cti Ul
•a
o
8
to
go
— t- CO
8
s
§
cu
.S
<0
1
I
5 °
12
«s 1*
J3 .0
D e3
fe ^
II
O £
OH (3
II
•d-ir
1§
s p
•a g
II «
1§
2 -a
.3 w
s
1|
< as
-ffiO
Jl
isii
iggi
TOTAL
Iii
ll!
s ^
O ca
01 "SL
>, >•
"Si
o §
57
-------
g
s»
CS
•s
CO
73
8
C3
3
04
tJ
.§
•3
£
0
1
1
s
IS
s
04
.2
•4— »
C
's
1
o
U
00
•o
53
1
,o
S
• •M
H"4
>^
j?
3
O
H
1
e
•a
§
S
o
U
6O
"2
S
W5
-
1
•g
•§
1
f
s
O
04
1
t
^"*
O
^•l
CO
3
CO
t)
§
\o t^-
, ,J
93
O
*l*"i
03
En
~T_
o £^
g I
s a
-°
2 °^
O3 4)
^ — "s
II l>"^
S *4i '
t3 <2 "2
•i "s «
^ wa OK
|P
S .SS
*JH ^ ^Q
§"
— " CO
III
ti
§,
i
03
|
a
§
S
1
«S
S
JC8
1
2i
1
.s
'I
_.
•o
_s
1
I
58
-------
•s
•g
I
w
•s
CO
CO
s r3
cu ,f^
^ <§
t Pha
nation
ect
*"* "S
O o
is
ants evaluated.
o
QH
2
W)
"S
59
-------
Table 13. Frequency of Pollutants from 55 BD Indirect Pharmaceutical Manufacturing Facilities
Which Discharge to 45 POTWS on 45 Receiving Streams
Pollutant Name
Nus&ejf of Detections by Facility
ACETONE
AMMONIUM HYDROXIDE
AMYL ACETATE, n-
BIS(CHLOROMETHYL)ETHER
BUTANOL, 1-
CHLOROFORM
DIETHYL ETHER
DIMETHYL SULFOXIDE
ETHANOL
ETHYL ACETATE
ETHYLENE GLYCOL
ETHYLENE GLYCOL MONOETHYL ETHER
FORMALDEHYDE
HEXANE, n-
ISOPROPANOL
ISOPROPYL ACETATE
ISOPROPYL ETHER
METHANOL
METHYLENE CHLORIDE
PHENOL
POLYETHYLENE GLYCOL 600
PYRIDINE
TOLUENE
4
4
1
1
2
1
2
3
32
2
3
3
6
1
22
1
1
16
9
7
4
1
1
Note: Only pollutants proposed for regulation are evaluated. (Three pollutants have been removed as candidates
for regulation since this analysis was completed - glycol ethers, bis(chloromethyl) ether and
dimethylcarbamyl chloride).
Source: Engineering and Analysis Division (BAD), April 1994.
60
-------
3 '
•s
s
I
A
P-.
1-4
<2
CO
O
'52
1
.•a
*t-i
u
•s
a
I
*
*
*
8
*
*
I
f?
en
.
, "&&
- '*i
o •.
:-i
»t»4 u 4L •^~ « 4t
^3* * ;; N**/ H
* ^H JH
to
o o o
O 0 0
o o o
oo
^T o w
° *B e
|| 1
IS 1
r; -^* c*
^> £
Dki E^
03
(No.)
ants (No.)
Excursions
.§
"S
.2
I
o
s
ts magnitude of excursio
45 and number of pollu
lation are evaluated.
ers, bis(chloromethyl
en
=
ca v
IS
ill!
.2 G
I ^
.S M
gS t^
o -o
E g
go
SI' IS
aS 5 &.=«
•° 3 S3
2*3
*
* *
61
-------
wo
If
£d
•s
1
fi
Q
03
I
•s
O)
o
CO
I
£
CO
O
J
I
w
%
I
's
<»
"
§•
e
6
§
-* O O|»-c
•^ o
0 -H OIlH
0 O 010
.
Number of p
Numbers re
oretreat
W
62
-------
•g
.52
Q
73
1
g
4J
g
£
c
o o
o o
*a
s-
«o
1
•i
1
o>
"S
12
« 'G
^
§
€ -S
8<5
n
o o o
000
"•3 . I
2 2 5
fil »
Ǥ
I
1
NO
I
co
^s
ai
"o
o
•c
OQ
S
2
^•a
*2 2
|£
II
PH H
03
'S
1
I
63
-------
••
d,
6
CO
H
03
•s
CO
s
s
Ui
CC
Q
ffl
.2
t
S
03
1
I
00
I
CO
§
_C
"S3
a
: 0
A
§
^
c£
1
sc
S3
§ <8
~*C jg
01 (J
87-
v S
& c
I6
H -
Cft
I
prf
"^vo
|A
I
I'
s
-
•
f / S
^
^^ S?
o o o o o ooooo
•<-<< ~ti "S
S2J5s§2-2o .22S3§o-°
U 55 U O i^5 H 5 55 O O cJa* <5a
•s
03
1
4
Ul
o
<+-
5
03
12
O
eg .
4>
•gs
^ >H
%2
B •*
§ u
t-l — .
S3 ^
i> c3
J3 -O
. O oi
lit
J 3 ^
111
^ S uT
'tt S s
0 fi -S
H H ®
«J ^J' X-V
14
^ ts s
^3 s o
§ 1 ^
•o « -g
S if
J: 2*
O co
«-i|
^s,^
JSs
1 S^
S «»-• DO
CO ^? '
S 5
I S"8 JJ'
g d, S J=
«3 O — «
to <-i O< u
21-11
fill
§a-s^
•a >.* "
£913
H
0
BJ
e
"5
a
1
S
1
09
U,
i
*o
o
o
^
.co
'co
•^
C3
64
-------
*
1/3
"53
5>
•s
s
I
I
E
CO
od
CO Xi
§5
vo oo ID >o
co go rj 5J
\^ So *r
-<' ci Vj
w n a
» -
es
0|
K
o o o
00
^. a ^
\9~ o c-
o »-i o
t> *•§-
A S *T?
i-t &
5i°
I
000
z z
o o
i s
rj cs
S §•§
« .O £3
0 0
4>
Stream No. 2
Chloroform
Dichloroeth
TOTAL
1
"S3
o
1
I
S
1
3
5 £-2
ii!
•g Q. S
P< a> i_T
«
-
'
CQ
-
*a3
O
if
1
•8
s
tp
ii
|
§»
o> S. •« 3 05
U-i ?^ J*| «1 **
2 I tf-a
S5T* LM
U4 fll
a
lt|<|
P O § Z Z
!
65
-------
-?
,9
•4— »
1*
1
U
S
to
£
c1
'S
c
•s
a
OQ
Si
&*
03
A
O
5
1
•4_»
8
o
U4
03
£
MH
1
5
Q
£9
o
i
•s
8
3
S
1
w
«+-
o
>>
13
S
s
c/3
j
»
s
o
tot
A
! §
•5
*5
s»f
1
SC
S *
e
a
i
g-*»
< ?§
i
X
w*J
3
^
.s
^ °
^ A
9
^
-
0000 OOOO
^^ ^^ ^^ ^gt ^gt ^^t ^^ ^^t
Z'Z.'Z'Z, 2ZZZ
— -
^
S
•**
2
w
ts-*oo oooo
1 1
0 0
VI VI
^3 •—;
'•2 *4^
P D
^^* ^KS y^^i , _ ^^N WJ <^^
^? [^£ O (^ *^t O
5- "0^ 5^ 60O
^cSc Is'dS'Jc
irScS SzScS
( 7| r^> t)jQ *r^ ty) *s ^^v t^ «^ tJD
S S §Q § jS g §Q §
mil mil
U&oU^U Qi^U^O
t:
,c
e
0
'ts
5
«j
* Q. a> 3
S i-fl
« eo oo ""^
"g 5 0 o
g "3 ^co ^
— ^'s, ii
If IS
I '
,
C
"S
o
1
I
-------
a
o
u
c«
S
in
§
S
.f
•a
1
£9
u
f
1
I
a
o
*->
•s
8
I
2
CO
*
*
(Thr
l)eth
H
SS:i
£5*
O O3
II '5 3
2
O
O
Q>
67
-------
09
'•£
? 8
•2 .S
a, <*>
05
C
o
a
O5
05
H
•S
fe
05
S
£?
03
•S
en
5
*c3
_^J
*•>
1
§
U-i
C3
w
"3
Q
CQ
§
w
s
*s
K
SC:
'
*ctt
3
< IS
Cft ^^
ll
g
»?2
^v?
: t-t
O
iQ^
rs A
I
O
' -••
••"
o o
ooooo ooooo
•^ ^>
f^ nj M ^
^ ^ ^< ^ ^i" "^ ^ ^3 ^J ^ "^
^
w
oo
— <»/1OO— OOOOO
x-^ ^^
6 6
S S
| ^ | | e
Ills! Uls
O c? o* fli flJ C © c? o* ^ 3J
^Tv ^1^ too ^^ .52 ^ ^ *••! ^^ tj) ^^ ,°^ (5
4^ E C3 ^4 cfl ^^ ••• S ^ 2 °^
h s 'o 2 ti 'S H s S 'o g tJ 'S5
32j3§0-°o .a^jgS0-0
U 35 U O ^5 H 5 5a O O c8"t»
*3
03
1
t-c
«*-
1
12
"§
^^
if
o ®
g J3
S «
<-i — .
IS §"
. "« "I
*ri > «J
751
*2 § S
si's
•3 §< §
£ § y
;e!
*§ ^ ?j»
g "2 "S
" ta S
CQ a o
^ls
C? 4) s <~^
5 w 8
5 O »S*
4) *+* 6J)
el'
ca o S JJ
J3 O ••« co
ffjl
"i M * £
SO en ^
°-"33 ii
I III
g
O
oi
c
1
•s
1
§
^*H
-2
«>
t
8
S
c
1
jta
•!"
c
*
68
-------
S3
3 a
O I
•S
V3
•a
t~<
If
*" E
18
to
to
co
O
I
oo
•** ra
i/> ,0
ca
fO
S «>
If 2 rf , l
•g-.'S . s
69
-------
•-
I
a
6
£?
CQ
-s
CO
5
Q
CQ
ce
1
3
a
.«
' o
*"*
A
1
1
*
O
e3
o
1
1 8
"**« S
•B4 Q
><
IS
0
H
a»
S:
fe
o
«3\p
IA
*> -
1
*••*
H
H
,
-
oooo oooo
co m
<«3^«3^ ««
ZZts^m Z Z £ £
w w
OO ON
^H ^-H
22 ^
W W fTl
co co v^
C* O ^
NOVOCNco ^CSOO
op 02
55
1 1
0 0
VI VI
5 P
£ ^ £ £ ^ >?'
fe apo o gp^
Irli ilili
^.j r~*^ ^f) *!^ ^n ^ »2 s^ wO "S OO
e — o /it o H «_iO/s;o
«ScHs3< sS^^c
fc « '§ -S '§ ft « § '§ -S "§
3is«s:«O .S2i-a:r:«a
U 55 U ^ U H Q 55 U & U
«
4>
"S
a
1
("H
S
es
3
T3
C
O3
U
OQ .
^3 ^^v
*o *o
Is
12
C3 "I"
Js ^
. fli S
tn > £
^3
^ w s
•S3 S fi
5 *s "B
Q W4 CQ
Ci< fli "*
"+-<§«>
o b .2
sEx
list
3 J 8
2 > ^
*t3 * "S
^ 1 "&
^ 0 *
" '^ ^
"S "§i ®
151
S <« 60
c"S J,
1 o ^ 2,
-------
71
-------
09
.2
I
J
.ts
§*
Pt,
13
1
^
g
(M)
•e-g
.2 .3
« j
^» a>
sa
il
a?
ti
S3
O
<
«4-l
O
oo
in
CM
o>
s
r/\
•s 1
« a;
11
Z {*•
1^
PQ »y"*
§1
kO •w
•S! «
ilf^"*'
(S
*•*
(TJ d
S 0
•<< »5
i u
a »
W K
u
*»
a> a>
s g
31
t* O
^J ^.
JK O
w s
u
§
CQ
§
^
t
O
0,
'
i
J2j
1
1
•"
o o o o o
2t «s ^ co ir> en
9999 9 9 f>
tXJ LQ QQ tQ [XJ QQ ^^
.•* o\ o \r> o\ m *
m oo o en o o> o
•* -H in i-< en ci
Ills 1 1
22 fN S ^ S 55 *^
en ^^ en o\ c^ ^~^
en in o oo o oo o
r- o\ oo en 'C
2 g 2 « -o JS «
§ g g -o^J ^ J
*O C^IH G S ^* *^3 ^J 3> *-^ ^J If
| | | J ^ | J.J | ^ g
« S 8 Q 'Q d s s S 5 S
'i
1
o
i
a
«
e
S
X)
^
"o
a.
i
>>
jg
"^N
^f^
•4-»
1
J
•V
&o
^^
G
O
(J
6
2
§
"° d
•O O
•g rg
cd !3
O
W
S
^
72
-------
.2
a
I
I
,0
1
tn
.52
s
w
fl>
*•>
'5b
1
IS
•3-
&«>
|g
< -c
v
ri
in
~bj
?? g05
•* O co '— '
9999
w w w w
en ts -H ir>
\o r4 ri O
W
oo
85
<
•s
£
CO
vo
cs
<1J
I
K
ea
.1
1
, "-,
OS
<»> 12
> S
35
s-s
i -3 2
^ ° s <
^M
°is
!
I
D J J
1
•8
I
i
O
a,
I
I
ca
g 1 o § § g
'•= 1 o S 3 g
'= g 3 3 .8 5
as u a Q O
<
« i/ a* -^
1 ll'l'Sl'
S 3 « * » ®
S J 1 « g £
[t, K S S H H
1
z
73
-------
Table 27. Summary of Potential POTW Occupational Exposure Impacts
for Pharmaceutical Indirect Discharges
Current
POTW
w/ Hazard Ratio > 1
Pollutants
w/Hazard Ratio > 1
Acetonitrile
Benzene
Chloroform
Diethylamine
Dimethylamine
Heptane, n-
Hexane, n-
Methylene Chloride
Toluene
Triethylamine
Distillation*
POTW
w/Hazard Ratio > 1
Pollutants
w/ Hazard Ratio > 1
Acetonitrile
Benzene
Hexane, n-
Total Number
85
12
52
10
4
1
2
1
1
5
2
2
4'
1
85
7
44
3
4
1
1
Hazard Score Range
1.5-2,561
1.1-2,530
1.2-25.8
2,530
6.0-16.1
5.0
9.3
1.1-6.0
3.0-62.9
3.4-6.7
1.1-2.2
1.9
1.2-31.1
1.1-25.8
1.2-25.8
5.3
1.1
* Analysis not completed for steam stripping.
74
-------
to
-------
1
I
8
•c
I
00
5
t3
I
1
I
1
00
cs
•1 •
§
g
w
'
C^r
CLJ
1
<
H
o
A
P3
1
W
Z
1
z
pa
%s
^
?J
^^
3
*
" ,
••
Since 1986 plant improperly
discharging wastes to groundwater
(leaks) and discharging excessive
levels to POTW. Overload of
POTW resulted in pollution of Lake
Onondaga. The company agreed to
pay a fine of $3.5 million and to
build a waste-water treatment plant
for the NY facility by 1996 at an
estimated cost of $30 million. Odor
control study and volatile compounds
study currently underway by POTW.
•o i
** Si
5£ ,
1 1
2 "2 .1 S
4) ^"^ U S
*w *3 ^ S
o m 3^ o
^ § ° ^
O *>,O §
0. J3 A- t3
^
|
0)
35
•O
§
U|
•a
s
1
00
•5
V
!
•3
ua j3
° §
ll
>,
w S
o5 C*
i-l
If
CO CO*
1
*1
la
s ^
CO i-J
Plant records confirmed that on-site
groundwater contamination resulted
from historical and accidental spills
of organic compounds.
1
"S"
1
fA
0
1
"9
"I
.3
to
1
*« &0
•3 -S
1 ||
111
i
1
4f
1
O)
Impacts on the Ramapo River due to
discharges from Orange County
Sewer District No. 1 POTW. Fish
kill also occurred. Nepera Chemical
was subject of an enforcement
hearing concerning alleged violations
of its discharge permit limitations for
ammonia. Problem has been
resolved.
|
2
„
s*
*T3
§
tf, v*
vl i— i
8 >
g JJ
0 §
1 S
H 5
.
1
1
J?
.g
1
5
=3
•g
2
1
"s
5
u
z
I
1"^
1 1
K P<
1
W
1
.a
•8
1
(D
•o
IB .
>» 05
11
1
1
1
J
.
I
i &
t|
Hj ^3
- 1
O 60
f43
ti
§
•§ S
S §
I
CO
i
ts'
1
i
CO
Direct contact with and ingestion of
contaminated soil by workers and
area residents who access the site.
Impacts can also occur from
ingestion and direct contact with
groundwater.
1
I
X!
..
i**?,^ | ^
.2 a> '? ^> >
S S ^^ q "Q uj
•2 . S ° 'C «
I § ^ ^ .2 •£
l||l|l
•o ° 8 -S - f
g 8 J « g |
•g g .g "3 ">. §
•S ^ ""§,00
3 ^ a s- M °
llllll
l-c
"a go § S eo
•2-5 o w -S
llltll
OH 2 O-. C/3 S (X
8
S
1
1
f
o
(O
-------
1 -1 5 S -s
"832
-------
1
525
te
§.a
*Jl.s-*
iits-j|.«i
<3 «i 2 "•3 "S S JS
"'Jlilfl
ill-*
till
1111111.1
It
si
J
a
8
,
ii
•s
I
•o
S
CO
s
O
-2 a
00
-------
en
-------
Metcalf & Eddy, Inc.
York.
5. REFERENCES
(1972) Wastewater Engineering. McGraw-Hill Book Company, New
NIOSH. (1985) Pocket Guide to Chemical Hazards. U.S. Department of Health and Human
Services.
Turner, D.B. (1970) Workbook of Atmospheric Dispersion Estimates. Research Triangle
Park, NC: U.S. Environmental Protection Agency, Office of Air Programs.
U.S. Bureau of the Census. (1992) Statistical Abstract of the United States:
Washington, DC: U.S. Bureau of the Census.
1992.
U.S. EPA. (1980) Ambient Water Quality Criteria Documents. Washington, DC: U.S.
Environmental Protection Agency, Office of Water. EPA 440/5-80 Series. [Also refer to any
updated criteria documents (EPA 440/5-85 and EPA 440/5-87 Series)].
U.S. EPA. (1986) Report to Congress on the Discharge of Hazardous Wastes to Publicly-Owned
Treatment Works (Domestic Sewage Study). Washington, DC: U.S. Environmental Protection
Agency, Office of Water Regulations and Standards.
U.S. EPA. (1987a) Industrial Source Complex (ISC) Dispersion Model User's Guide. Vol. I.
Second Edition (Revised). Washington, DC: U.S. Environmental Protection Agency, Office
of Air Quality Planning and Standards. EPA-450/4-88-0029.
U.S. EPA. (1987b) Development Document for Effluent Limitations Guidelines and Standards
for the Organic Chemicals, Plastics and Synthetic Fibers Point Source Category. Volumes I and
n. Washington, DC: U.S. Environmental Protection Agency. Available from NTTS,
Springfield, VA. PB88-171335.
U.S. EPA. (1987c) Guidance Manual for Preventing Interference at POTWs. Washington, DC:
U.S. Environmental Protection Agency.
U.S. EPA. (1989a) Exposure Factors Handbook. Washington, DC: U.S. Environmental
Protection Agency, Office of Health and Environmental Assessment. EPA/600/8-89/043.
U.S. EPA. (1989b) Risk Assessment Guidance for Superfund. Washington, DC: U.S.
Environmental Protection Agency, Office of Solid Waste and Emergency Response. Available
from NTIS, Springfield, VA. PB-90-155581.
R-l
-------
S EPA (19890 Toxic Chemical Release Inventory - Risk Screening Guide Washington,
US. Environmental Protection Agency, Office of Pesticides and Toxic Substances.
EPA/560/2-89-002.
US EPA (1990a) GAMS Version 3.0-User's Guide. Washington, DC: U.S Envkonmentel
Sotection Agency, Office of Pesticides and Toxic Substances. Exposure Evaluation Division.
EPA Contract No. 68-02-4281. GSC TR-32-90-010.
U S EPA (1990b) CERCLA Site Discharges to POTWs: Guidance ManuaL Washington,
DCV U.S.' Envkonmental Protection Agency, Office of Emergency and Remedial Response.
EPA/540/G-90/005.
U.S. EPA. (1990c) National Water Quality Inventory - Report to Congress. Washington, DC:
U*.S*. Environmental Protection Agency, Office of Water.
TT S EPA ACGffi (1990d) Threshold Limit Values for Chemical Substances and Physical
Age'ntfati B^cal Exposure Indices. Washington, DC: U.S. Environmental Protection
Agency.
U.S. EPA. (1991) Technical Support Document for Water
Washington, DC: U.S. Environmental Protection Agency, Office of Water. EPA/505/2-90-001.
Available from NTTS, Springfield, VA. PB91-127415.
U S EPA (1992a) Mixing Zone Dilution Factors for New Chemical Exposure Assessments,
SraVSpoi TcSober 1992. Washington, DC: U.S. Environmental Protection Agency.
Contract No. 68-D9-0166. Task No. 3-35.
U S EPA (1992b) Guidance to Protect POTW Workers from Toxic and Reactive Gases and
Va^r?wihington5DC: U.S. Environmental Protectior Agency Office of Water. EPA812-
B-92-001. Available from NTIS, Springfield, VA. PB92-173-236.
U.S. EPA. (1993a) POTW Pass-Through Analysis for the
Wellington, DC: U.S. Environmental Protection Agency. EPA Contract No.
U S EPA. (1993b) Pharmaceutical Outreach Questionnaire. Washington, DC: U.S.
Environmental Protection Agency. EPA Contract No. 68-CO-0032.
U.S. EPA. (1994) Superfund Chemical Data Matrix. Washington, DC: U.S. Environmental
Protection Agency, Office of Solid Waste.
US FWS. (1985) National Survey of Fishing, Hunting and Wildlife Associated Recreation.
u!s*. Department of the Interior Fish and Wildlife Service.
R-2
-------
Versar, Inc. (1992) Upgrade of Flow Statistics Used to Estimate Surface Water Chemical
Concentrations for Aquatic and Human Exposure Assessment. Report prepared by Versar Inc.
for the EPA's Office of Pollution Prevention and Toxics.
R-3
-------
-------
-------
-------
-------
------- |