-------
TABLE VI-12.
SIX TOXIC POLLUTANTS DETERMINED NOT TO INTERFERE
WITH, INHIBIT, OR PASS-THROUGH POWs, AND EXCLUDED
FROM REGULATION UNDER PSES AND PSNS
Benzo(A)Anthracene
Benzo(A)Pyrene
Chrysene
Chromium
Copper
Nickel
TABLE VI-13.
SIX TOXIC POLLUTANTS THAT DO NOT
VOLATILIZE EXTENSIVELY AND DO NOT HAVE
POTtf PERCENT REMOVAL DATA
Acrylonitrile
Bis(2-Chloroisopropyl)Ether
2,4-Dini trophenol
3,4-Benzofluoranthene
Benzo(K)Fluoran thene
Acenaph thylene
VI-40
-------
'TABLE VI-14.
RESULTS OF PSES ANALYSIS TO DETERMINE IF TOXIC
POLLUTANT REMOVALS WERE "... SUFFICIENTLY
CONTROLLED BY, EXISTING TECHNOLOGIES
Percent of. plants at .which the pollu-
tant is adequately treated or costed
Pollutant
Number
3
42
59
74
75
77
Pollutant due td presence
Name treatable
Acrylonitrile
Bis(2-Chloroisopropyl)Ether
2 , 4-Dini trophenol
3 , 4-Benzof luoranthene
Benzo ( k) Fluoran thene
Acenaphthylene
of: .another similarly
toxic pollutant
39%
50%
;ioo%
*
100%
87%
* Analysis could not be performed
VI-41
-------
TABLE VI-15.
THREE TOXIC POLLUTANTS EXCLUDED FROM PSES AND PSNS
REGULATION UNDER PARAGRAPH 8(a)(iii) OF THE
SETTLEMENT AGREEMENT BECAUSE THEY WERE "... SUFFICIENTLY
CONTROLLED BY EXISTING TECHNOLOGIES ..."
2,4-Dini trophenol
Benzo(K)Fluoranthene
Acenaph thylene
TABLE VI-16.
THREE POLLUTANTS RESERVED FROM REGULATION UNDER
PSES AND PSNS DUE TO LACK OF POTW
PERCENT REMOVAL DATA
Acrylonitrile
Bis(2-Chloroisopropyl)Ether
3,4-Benzofluoran theme
VI-42
-------
will be "...sufficiently controlled by existing technologies." The Agency has
also decided to reserve the three remaining toxic pollutants from regulation
under PSES and PSNS in addition to the seven pollutants shown in Table VI-6
(see Tables VI-15 and VI-16, respectively).
VI-43
-------
SECTION VI
REFERENCES
6-1 U!CTKT™ ^HVPCR' STANDARD METHODS FOR EXAMINATION OF WATER AND
WASTEWATER, 4TH EDITION, WASHINGTON, DC, APHA, 19076, P. 549
6-2 Ibid., p. 94
6-3 Ibid., p. 516, 517, 519, 521.
6-4 Ibid., p. 554
6-5 Ibid., p. 534
VI-44
-------
VII. CONTROL AND TREATMENT TECHNOLOGIES
A. INTRODUCTION
This section identifies and describes the principal Best Management
Practices (BMPs) and in-plant and end-of-pipe wastewater control and treatment
technologies currently used or available for the reduction and removal of
conventional, nonconventional, and priority pollutants discharged by the OCPSF
industry. Many OCPSF plants have implemented programs that combine elements
of BMPs, in-plant wastewater treatment, and end-of-pipe wastewater treatment
to minimize pollutant discharges from their facilities. Due to the diversity
of the OCPSF industry, the configuration of these controls and technologies
differs widely from plant to plant.
BMPs are in-plant source controls and general operation and maintenance
(O&M) practices that prevent or minimize the potential for the release of
toxic pollutants or hazardous substances to surface waters or POTWs (7-1).
The following pages describe these in-plant source controls (i.e., process
modifications; instrumentation; solvent recovery; and water reuse, recycle,
and recovery) and O&M practices that are employed, or could potentially be
employed, at OCPSF plants.
Physical/chemical in-plant treatment technologies are used selectively in
the OCPSF industry on certain process wastewaters to recover products or
process solvents, to reduce loadings that may impair the operation of a
biological treatment system, or to remove certain pollutants that are not
sufficiently removed by biological treatment systems. The in-plant treatment
technologies currently used or available to the OCPSF industry and available
performance data for these technologies are described and presented in Part C
of this section.
End-of-pipe treatment systems in the OCPSF industry employ physical,
biological, and physical/chemical treatment, and often consist of a
combination of primary (neutralization and settling), secondary (biological
high rate aeration and clarification), polishing, and/or tertiary (ponds,
filtration, or activated carbon adsorption) unit operations. The end-of-pipe
VII-1
-------
treatment technologies currently used or available to the OCPSF industry and
available performance data for these technologies are described and presented
in Part D of this section.
The performance of selected BPT and BAT total treatment systems,
including nonbiological treatment systems, are presented in Part E of this
section. tfastewater discharge or disposal methods (other than direct to
surface waters and indirect through POTWs) used by OCPSF plants, frequently
called zero or alternate discharge methods, are presented in Part F. Part G
presents treatment and disposal options for the sludges resulting from certain
wastevater treatment operations. Finally, Part H presents the procedures used
to develop the effluent limitations guidelines and standards for the OCPSF
industry.
The Environmental Protection Agency (EPA) developed three technology
options for promulgating BPT. BPT Option I consists of biological treatment,
which usually involves either activated sludge or aerated lagoons, followed by
clarification (and preceded by appropriate process controls and in-plant
treatment to ensure that the biological system may be operated optimally).
Many of the direct discharge facilities have installed this level of treat-
ment. BPT Option II is based on Option I with the addition of a polishing
pond to follow biological treatment. BPT Option III is based on multimedia
filtration as an alternative basis (in lieu of BPT Option II polishing ponds)
for additional total suspended solids (TSS) control after biological
treatment.
EPA has selected BPT Option I-biological treatment with secondary
clarification-as the technology basis for BPT limitations controlling BOD
and TSS for the OCPSF industry. This option has been previously described'by
EPA as "biological treatment." However, a properly designed biological treat-
ment system includes "secondary clarification" which usually consists of a
clarifier following the biological treatment step of activated sludge, aerated
lagoons, etc. The rationale for the selection of BPT Option I as the basis
for the final BPT effluent limitations is discussed in detail in Section IX.
VII-2
-------
EPA developed three final options for BAT effluent limitations. BAT
Option I would establish concentration-based BAT effluent limitations for
priority pollutants based on using BPT-level biological treatment for tHe
end-of-pipe biological treatment subcategory. Since some plants do not have
sufficient BOD5 in their wastewater to support (or require) biological
treatment, there is a non-end-of-pipe biological treatment subcategory. The
plants in this subcategory db not use end-of-pipe biological treatment.; their
BAT Option I treatment involves in-plant controls that consist of physical/
chemical treatment and in-plant biological treatment to achieve toxic
pollutant limitations, with end-of-pipe TSS control if necessary.
' BAT Option II would establish concentration-based BAT effluent
limitations based on the performance of the end-of-pipe treatment component
(biological treatment for the end-of-pipe biological treatment subcategory and
physical/chemical for the non-end-of-pipe biological treatment subcategory),
plus in-plant control technologies that remove priority pollutants prior to
discharge to the end-of-pipe treatment system. The in-plant technologies
include steam stripping to remove selected volatile and semivolatile (as
defined by the analytical methods) priority pollutants, activated carbon for
various base/neutral priority pollutants, chemical precipitation for metals,
alkaline chlorination for cyanide, and in-plant biological treatment for
removal of selected priority pollutants, including several polynuclear
aromatics (PNA), several phthalate esters, and phenol.
BAT Option III adds activated carbon to the end-of-pipe treatment to
follow biological treatment or physical/chemical treatment in addition to the
BAT Option II level of in-plant controls.
The Agency has selected Option II as the basis for BAT limits for both
subcategories. The rationale for the selection of BAT Option II as the basis
for the final BAT effluent limitations for both subcategories is discussed in
detail in Section X.
The Agency is promulgating PSES for all indirect dischargers*based on the
same technology basis as the BAT non-end-of-pipe biological treatment
subcategory. The rationale for selection of this technology basis for the
final PSES effluent limitations guidelines is discussed in Section XII.
VII-3
-------
A review of waste management practices and well-designed and -operated
wastewater treatment system configurations currently in use by the OCPSF
manufacturing facilities, reveals that there are numerous approaches for
implementing effective pollutant control practices. Since the Agency does not
specify what technology must be used to achieve the promulgated numerical
effluent limitations and standards, the following portions of this section
describe the unit operations and treatment practices that provide the bases of
the selected technical options, as well as alternative unit operations and
treatment systems that may also be utilized to achieve pollutant reduction
goals. As noted in Section VIII, the Agency's methodology for estimating the
engineering costs of compliance for individual facilities is based on costing
one or more of the treatment unit operations included in the selected
technology option, depending on the difference between current effluent pollu-
tant concentrations and target effluent concentrations that would be required
to achieve compliance with regulatory requirements.
B. BEST MANAGEMENT PRACTICES
Best Management Practices (BMPs) consist of a variety of procedures to
prevent or minimize the potential for the release of toxic pollutants or
hazardous substances to surface waters or POTWs (7-1). Specific practices
that limit the volume and/or contaminant concentration of polluted waste
streams, such as solvent recovery, water reuse, and various pretreatment
options, involve applying BMPs to facility design. O&M procedures such as
preventive'maintenance measures, monitoring of key parameters, and equipment
inspections that minimize the potential for unit process failures and
subsequent treatment plant upsets are also considered part of BMPs. The
following discussion is divided into two parts: in-plant source controls
(i.e., process modifications; instrumentation; solvent recovery; and water
reuse, recycle, and recovery) and general O&M practices. Several specific
examples of how wastewater treatment plants improved their performances
through minor modifications are also included.
!• In-Plant Source Controls
In-plant source controls include processes or operations that reduce
pollutant discharges within a plant. Some in-plant controls reduce or
VII-4
-------
eliminate waste streams, while others recover valuable manufacturing
by-products.
In-plant controls provide several advantages: income from the sale of
recovered material, reduction of end-of-pipe treatment costs, and removal of
pollutants that upset or inhibit end-of-pipe treatment processes (7-2).
While many newer chemical manufacturing plants were designed to reduce
water use and pollutant generation, improvements can often be made in older
plants to control pollution from their manufacturing activities. The major
in-plant source controls that are effective in reducing pollutant loads in the
OCPSF industry are described below.
a. Process Modifications
Most manufacturers within the OCPSF industry use one or more toxic prior-
ity pollutants in various stages of production. In some cases, problems per-
taining to a difficult-to-treat pollutant can be solved by finding less toxic
or easier to treat substitutes for that compound. In many cases, a suitable
substitute can be found at no or minor additional cost.
In some situations, plants can improve their effluent quality by shifting
from batch processes to continuous operations, thus eliminating the waste-
waters generated between batches by cleanup with solvents or caustic. Such
modifications increase production yields and reduce wastewater generation.
Effluent quality at a facility can sometimes be improved by taking advan-
tage of'unused equipment or by simply reconfiguring existing equipment and
structures. Some plant-specific approaches are as follows:
• Floor drains likely to receive spills can be designed to flow into a
collection sump instead of directly into an industrial sewer system.
This allows concentrated wastes to be recovered, treated, or
equalized prior to being pumped or transferred to the wastewater
treatment plant.
• Highly acidic or basic waste streams can be neutralized or diluted by
being mixed together upstream of the wastewater treatment plant.
VII-5
-------
Unused tanks at a facility can be fitted to intercept shock loadings
and allow concentrated pollutants to be gradually mixed in with
process wastewater at a high dilution rate. Excess tank or lagoon
capacity can also be used to increase detention times and improve
equalization of wastewaters.
An abandoned steam stripper from a closed process line can be con-
Preheating or cooling waste streams designated for biological treat-
ment can also be a great asset as activated sludge systems generally
Two examples of process modifications from other industries may be appli-
cable to the OCPSF industry. The first involves biological degradation.
Although anaerobic digestion is common at the mesophilic temperature of 30°C,
use of thermophilic digestion has gained popularity of late because of poten-
tially increased solids destruction. New York City, in its wastewater treat-
ment operation, conducted thermophilic digestion directly after mesophilic
digestion. This has led to increased sludge solids destruction, and when
employed with increased decanting, has led to a reduction in sludge volume and
more efficient operation (7-3).
Another modification involves the use of a step-feed operating program.
Having a variety of feed points enables the protection of effluent quality
while steps are taken to correct malfunctions in the biological treatment
process.
b. Instrumentation
Process upsets resulting in the discharge of products, raw materials, or
by-products are important sources of pollution in the OCPSF industry. Well-
designed monitoring, sensor, and alarm systems can enable compensatory action
to be taken before an unstable condition results in such process upsets.
Some common parameters that can be monitored and controlled using various
types of sensors and equipment include flow (both open channel and closed
conduit), pump speed, valve position, and tank level. Analytical measurements
such as PH, dissolved oxygen (DO), suspended solids, and chemical residuals
VII-6
-------
can also be monitored and regulated using feedback control equipment. At many
facilities, the overpressurization of reaction kettles, the bursting of
rupture-disks, and the discharge of chemical pollutants could be controlled
with a proper early warning system.
c. Solvent Recovery
The recovery of waste solvents has become a common practice among plants
using solvents in their manufacturing processes. In some cases> solvents can
be recovered in a sufficiently pure form to be used in the same manner as new
solvents. Solvents of lesser quality may still be usable in other areas of
manufacturing or be sold to another facility for use in applications not
requiring a high level of purity. Also, many private companies exist that
collect and reclaim spent solvents which are then sold back to the same or
other OCPSF facilities.
Solvents that cannot be recovered or reused can be destroyed through
incineration. Incineration may also be the best disposal method for used
solvents that cannot be economically recovered and for wastes such as bottoms
from solvent recovery units.
Solvent recovery, off-site reclaiming, reuse, and incineration are
methods of removing solvents from waste streams before they arrive at an end-
of-pipe treatment system or a POTW. Therefore, they contribute to protecting
biological treatment units from toxic shocks which could cause poor effluent
quality. In addition, as the cost for disposal of hazardous liquid waste
increases, solvent recovery becomes more economical.
d. Water Reuse, Recycle, and Recovery
Water conservation through reuse, recycle, and recovery can result in
more efficient manufacturing operations and a significant reduction in indus-
trial effluent requiring treatment. Recycling cooling water through the use
of cooling towers is a common industrial practice that dramatically decreases
total discharge volume. While noncontact cooling water may require little or
no treatment prior to recycling (other than reducing the water temperature in
cooling towers and adding corrosion inhibitors), treatment of the wastewater
VII-7
-------
prior to reuse is usually necessary to ensure a return stream of sufficient
quality for use in the process. In some cases, the treatment required is
simple, and facilities may already exist on-site (e.g., sedimentation).
fiy reducing the volume of wastewater discharged, recycling often allows
the use of abatement practices that are uneconomical on the full waste stream.
Further, by allowing concentrations to increase, the opportunities for recov-
ery of waste components to offset treatment cost (or even achieve profitabil-
ity) are substantially improved. In addition, pretreatment costs of process
water (and in some cases, reagent use) may be reduced. For example, removal
efficiencies for metals in chemical precipitation units are increased at
higher raw waste concentrations and proper chemical coagulant dosage. More
economical recovery of solvents is obtained from a properly designed steam
stripper at elevated solvent feed levels. Recycling also enables many plants
to achieve zero discharge, eliminating the need for ultimate disposal or
surface discharge.
Recycling systems can achieve significant pollutant load reductions or
zero discharge at relatively low cost. The systems are easily controlled by
simple instrumentation, and relatively little operator attention is required.
The most important design-parameter is the recycle rate (rate of return) to
the process stream or blowdown rate from closed loop recycle systems to avoid
build-up of dissolved solids.
Recycling limitations include the potential for plugging and scaling of
the process lines and excessive heat build-up in the recycled water which may
require cooling prior to reuse. Chemical aids are often used in the recycle
loops to inhibit scaling or corrosion.
Other approaches to reducing industrial discharge volumes include equip-
ment modifications and separation of stormwater and process wastewater. The
use of barometric condensers can result in significant water contamination,
depending upon the nature of the materials entering the discharge water
streams. As an alternative, several plants use surface condensers to reduce
hydraulic or organic loads. Water-sealed vacuum pumps can also create water
pollution problems. These problems can be minimized by using a water recircu-
lation system to reduce the amount of water being discharged.
VII-8
-------
Separation of stormwater and process wastewater enables each waste stream
to receive only the treatment required, and prevents problems caused by large
volumes of stormwater being contaminated by process wastewater, which sub-
sequently requires specialized treatment. If stormwater contains polluted
runoff from contaminated areas of a site, it may be possible to collect the
stormwater in retention basins and then gradually blend it in with process
wastewater in an equalization basin at the beginning of the wastewater treat-
ment cycle.
2. Operation and Maintenance (O&M) Practices
Many O&M practices minimize the potential for unit process failures and
subsequent treatment plant upsets. Inspections of those aspects of site
operation that have the highest potential for uncontrolled chemical releases
should be conducted by qualified maintenance or environmental engineering
staff members. Construction records should be reviewed to assure that under-
ground tanks and pipes have coatings or cathodic protection to inhibit
corrosion. Storage tanks and pipelines should be regularly inspected for
leaks, corrosion, deterioration of foundation or supports, pitting, cracks,
deformation, or any other abnormalities. Seams, rivets, nozzle connections,
valve function and position, and any associated ancillary equipment should.
also be inspected regularly to check for deterioration as well as potential
leaks from human error (e.g., valve not closed, loose pipe connections).
Training is important to assure that an operator reacts properly to upset
conditions. Treatment plant personnel should receive on-the-job and classroom
training covering the fundamental theories of wastewater treatment, specific
information about the equipment in use at that facility, the nature of
manufacturing processes and potential for upset, and prearranged procedures
for responding to upset conditions. Plants with operational flexibility may
be able to compensate to some degree for sudden changes in weather conditions
or inflow volume and quality by adjusting factors such as hydraulic retention
times and clarifier overflow rates through altering recycling rates, putting
backup units on-line, or directing excess wastewater to a holding basin until
flow rates return to normal. In addition, manufacturing personnel upstream of
a treatment plant should be trained in the proper disposal of waste chemicals
VII-9
-------
and the restrictions associated with disposal of wastes in industrial sewers
or storm drains.
Facilities handling a wide range of chemicals should be particularly
sensitive to potential problems arising from incompatible materials mixing in
tanks or pipelines. Monitoring storm sewers and industrial sewers on a
regular basis for toxic and hazardous pollutants is useful in identifying
potential misuse of sewers or evidence of infiltration of industrial wastes.
This type of internal housekeeping helps to reduce the potential for uncon-
trolled releases from a facility or shock loadings to an on-site treatment
plant.
At some facilities, waste treatment operations can be improved by
bringing in private contractors to handle some or all facets of operations.
Contractors experienced in treatment plant operations may have greater avail-
able technical resources to draw from than typical plant personnel in the
event of an operational problem. For example, a company specializing in
sludge handling may be able to improve that aspect of treatment plant
operations with a higher level of expertise and a lower cost than plant
personnel. In addition, a contractor operating several treatment plants may
be able to reduce costs for all facilities through bulk purchasing of
chemicals and pooling parts inventories.
If properly applied, certain O&M practices can compensate for cold
weather temperatures. Plants operating in cold weather conditions must
recognize that unnecessary storage of wastewater prior to treatment may reduce
the temperature of the biotreatment system. Cold weather operation may
require insulation of treatment units, covering of open tanks, and/or tracing
of chemical feed lines. Maintenance of higher mixed liquor suspended solids
(MLSS) concentrations and a reduced food-to-microorganism (F/M) ratio may be
necessary. Plant-specific techniques are presented in the summer/winter
discussion in the secondary treatment technology section.
VII-10
-------
C. IN-PLANT TREATMENT TECHNOLOGIES
1. Introduction
In-plant treatment is directed toward removing certain pollutants from
segregated product/process waste streams before these waste streams are com-
bined with the plant's remaining wastewaters. In-plant technologies, usually
designed to treat toxic or priority pollutants, could often be used for
end-of-pipe treatment of the plant's combined waste streams. Using these
technologies on segregated internal waste streams is usually more cost-
effective, since treatment of low volume, concentrated, and homogenous waste
streams generated by specific product/processes is more efficient.
In-plant treatment is frequently employed to protect the plant's end-
of-pipe treatment by removing the following types of pollutants (7-2):
• Pollutants toxic or inhibitory to biological treatment systems
• Biologically refractive pollutants
• High concentrations of specific pollutants
• Pollutants that may offer an economic recovery potential (e.g., sol-
vent recovery)
• Pollutants that are hazardous if combined with other chemicals down-
stream
• Pollutants generated in small volumes in remote areas of the plant
• Corrosive pollutants that are difficult to transport.
>• Many technologies have proven effective in removing specific pollutants
from the wastewaters produced by OCPSF plants. The selection of a specific
in-plant treatment scheme depends on the nature of the pollutant to be
removed, and on engineering and cost considerations.
The frequency of in-plant treatment technologies in the OCPSF industry is
presented in Table VII-1. This information was compiled from the 546 OCPSF
manufacturers that responded to all three parts of the Section 308 Question-
naire and the 394 Part A plants that responded to only Part A of the Section
308 Questionnaire. OCPSF manufacturers are defined as "full-response" if
VII-11
-------
"SB
=«=
I
=**= CM
4-J
!J
o PJ
-------
over 50 percent of their total plant production includes OCPSF products; if
they treat their OCPSF was testers in a separate treatment system; or if only
one treatment system is employed, the non-OCPSF wastewaters contribute less
than 25 percent of the total process flow. Part A plants are those that meet
the definition of being zero dischargers or do not meet the full-response
requirements stated above as direct or indirect dischargers. The 1983 Section
308 Questionnaire requested information on the plant's general profile
(Part I); detailed production information (Part II); and wastewater treatment
technology, disposal techniques, and analytical data summaries (Part III).
In-plant controls frequently used by OCPSF plants for the treatment of
individual waste streams include steam stripping (82 plants), distillation
(72), filtration (54), chemical precipitation (50), solvent extraction (29),
and carbon adsorption (18).
This section presents a general description and performance data for
selected in-plant treatment processes that are currently used or that may be
applicable to treat wastewaters from the OCPSF industry. General descriptions
of the treatment technologies are based largely upon material found in the EPA,
Treatability Manual, most recently revised in February 1983 (EPA-600/2-82-
OOla). Performance data specific to various technologies are derived from
four sources. The first source is OCPSF data compiled from responses to the
1983 OCPSF Section 308 Questionnaire, responses to the Supplemental Question-
naire sent to 84 facilities, and data collected by EPA in several sampling
studies previously detailed in Section V. The second source is data obtained
from other point source categories found in EPA technical development
documents and the Treatability Manual. The third source is data submitted as
part of public comments, on the proposal and NOAs. Technical literature serves
as the final source of performance data.
2. Chemical Oxidation (Cyanide Destruction)
Oxidation is a chemical reaction process in which one or more electrons
are transferred from the chemical being oxidized to.the chemical initiating
the transfer (the oxidizing agent). The primary function performed by oxida-
tion is detoxification. For instance, oxidants are used to convert cyanide to
VII-13
-------
the less toxic cyanate or completely to carbon dioxide and nitrogen. Oxida-
tion has also been used for the removal of phenol and organic residues in
wastewaters and potable water. Oxidation can also be used to assure complete
precipitation, as in the oxidation of iron from the ferrous (Fe*2) to the
ferric (Fe+3) form where the more oxidized material has a lower solubility
under the reaction conditions. Cyanide destruction (the oxidation of cyanide
to carbon dioxide and nitrogen) is a form of chemical oxidation and will be
used to illustrate the oxidation process, which is discussed in detail below.
Cyanide Destruction. Chlorine in elemental or hypochlorite salt form is
a strong oxidizing agent in aqueous solution, and is used in industrial waste
treatment facilities primarily to oxidize cyanide. Chemical oxidation equip-
ment often consists of an equalization tank followed by two reaction tanks,
although the reaction can be carried out in a single tank. The cyanide alka-
line chlorination process uses chlorine and a caustic to oxidize cyanides to
cyanates and ultimately to carbon dioxide and nitrogen. The oxidation
reaction between chlorine and cyanide is believed to proceed in two steps, as
follows :
(1) CN" + .01, « CNC1 + Cl
(2) CNC1 + 20H~
CNO"
H20
The cyanates can be further decomposed into nitrogen and carbon dioxide by
excess chlorination:
(3) 2CNO" + 40H~ + 3C1,
ecr
2C0
N
2H20
According to the Section 308 Questionnaire data base, 30 OCPSF plants use
chemical oxidation as an in-plant treatment technology; of these, 11 plants
use chemical oxidation for cyanide destruction. Performance data for chemical
oxidation are not available for the OCPSF industry. However, data for cyanide
destruction from the metal finishing industry are available, and can be
applied to the OCPSF industry as discussed in detail later in this section and
in Tables VII-2 and VII-3.
VII-14
-------
CO
CO
tx*
Co
Q
rH CJ
03 G
O -H /-^
B 0 &-S
CU -H ^
03 MH
<+H
w
4_j
3
O
G
CN
1
rH CJ
cd G
O -H <"«•.
B 0 &•?
a) -H •— '
PH HH
w
4-1
3
o
CJ
I—I
|
1
i — 1 O
^ cu
O -H ^N
a o s^
CU -H -~^
<4H
w
3
o
a
M
i-J
4J
CU
e
03
U
03
PM
VO ^O
ON ON
ON ON
O 0
* *
O^
CO
sf CO
CM CM
r~~ r-
co co
ON
CO CO
o o
* • •
0 0
CO CO
• *
CM CM
ON ON
CO O
rH rH
rH rH
• •
sf sf
rH rH
0)
rH
i-H 03
o3 a
4-> CU
EH
•H -~
ij O
bo
4-> CU
C 4->
fi
G -H
CU O
3
CJ ho
°l
4-> W
C -H
CU G
Cub
O
rH rH
CU 03
cu cu
Q s
.•
CU
O
i-j
3
O
CO
VII-15
-------
TABLE VII-3.
PERFORMANCE DATA FOR TOTAL CYANIDE OXIDATION USING CHLORINATION
Plant ID
Adjusted Average Total CN
Effluent Concentration (mg/1)
12065
21051
38051
06075
36623
19050
20079
05021
20078
20080
15070
33073
09026
31021
33024
0.14
0.0
0.0
0.039
0.103
0.031
17.54
0.035
0.083
0.949
0.323
0.707
0.119
0.708
0.204
Source: Development Document for Effluent Limitations Guidelines
New Source Performance Standards for the Metal Finishing
Point Source Category, June 1983.
VII-16
-------
As shown in Table VII-2, removal efficiency for plant #30022 using ozone
as an oxidant varies between 87 and 96 percent. The oxidation of cyanide
using ozone results in high capital and energy costs, and its efficiency is
limited when treating wastewaters containing more than one pollutant. Cyanide
can also be destroyed using hydrogen peroxide, but this results in high energy
costs because the wastewater must be heated prior to treatment. Furthermore,
peroxide only partially oxidizes cyanide to cyanate, and the addition of a
formaldehyde catalyst results in a higher strength (BOD5 level) wastewater.
Results of cyanide oxidation using chlorination from a number of metal
finishing plants can be seen in Table VII-3. Average effluent cyanide
concentrations range from 0.0 (plant #2.1051) to 17.54 mg/1 (plant #20079).
EPA indicated in its December 8, 1986, Notice that it was considering
using the performance data for cyanide destruction from the metal finishing
industry to develop cyanide limitations and standards. These data are based
on alkaline chlorination (a type of chemical oxidation). Public comments on
this notice suggested that EPA should transfer cyanide destruction performance
data from the pharmaceutical manufacturing industry rather than from the metal
finishing industry because of the similarity in wasfew'ater characteristics
shared by the OCPSF and pharmaceutical categories. EPA has evaluated the
pharmaceutical cyanide destruction performance data and has rejected transfer
of these.data for use in the development of OCPSF cyanide limitations because
the cyanide(destruction performance data from the pharmaceutical industry are
from a cyanide hydrolysis system that utilizes high temperatures and pressures
to hydrolyze free cyanide; this particular type of cyanide destruction tech-
nology has not yet been demonstrated to be effective on OCPSF cyanide-bearing
wastewater. EPA believes that the cyanide destruction by alkaline chlor-
ination data from the metal finishing industry are more appropriate for
transfer to the OCPSF industry since this technology is used on cyanide waste
streams in the OCPSF industry.
Another significant issue raised concerning the use of alkaline
chlorination technology in the OCPSF industry was the contention that while
this technology may effectively reduce concentrations of free cyanide in OCPSF
wastewaters, it cannot reduce concentrations of metal-complexed cyanides.
VII-17
-------
Industry comraenters have stated that the limitations and standards should be
for amenable cyanide only. EPA has evaluated the expected amount of cyanide
complexing resulting from the presence of certain transition metals (i.e.,
nickel, copper, silver, and cobalt in OCPSF cyanide-bearing waste streams),
and has concluded that only cyanide complexed by copper, silver, or nickel
could present a problem for treatment by alkaline chlorination. However,
silver is found at such low levels in the process wastewater of so few
product/processes that cyanide complexing would not present a problem, and
only a limited number of product/process waste streams would contain combina-
tions of either copper and cyanide (four sources), or nickel and cyanide (two
sources). For these six product/process sources, a potential for cyanide
complexing is present. However, no data have been submitted to demonstrate
that the actual levels of complexing interfere with the ability of the plant
to meet the total cyanide limitations. Thus, EPA believes that limitations and
standards controlling total cyanide are appropriate for all dischargers
subject to this regulation. A discussion identifying the sources of cyanide
and the product/processes with a potential for complex formation with nickel
and copper are contained in Section V of this document.
3. Chemical Precipitation
Chemical precipitation is a principal technology used to remove metals
from OCPSF wastewaters. Most metals are relatively insoluble as hydroxides,
sulfides, or carbonates, and can be precipitated in one of these forms. The
sludge formed is then separated from solution by physical means such as sedi-
mentation or filtration. Hydroxide precipitation is the conventional method
of removing metals from wastewater. Most commonly, caustic soda (NaOH) or
lime (Ca(OH)2) is added to the wastewater to adjust the pH to the point where
metal hydroxides exhibit minimum solubilities and are thus precipitated.
Sulfide precipitation has also been demonstrated to be an alternative to
hydroxide precipitation for removing metals from certain wastewaters.
Sulfide, in the form of hydrogen sulfide, sodium sulfide, or ferrous sulfide,
is added to the wastewater to precipitate metal ions as insoluble metal
sulfides. Carbonate precipitation, while not used as frequently as hydroxide
VII-18
-------
or sulfide precipitation, is another method of removing metals from waste-
water. A carbonate reagent such as calcium carbonate is added to the waste-
water to precipitate metal carbonates. The solubility of metal hydroxides and
sulfides as a function of pH is shown in Figure VII-1. The solubility of most
metal carbonates is between hydroxide and sulfide solubilities.
Chemical precipitation has proven to be an effective technique for
removing many industrial wastewater pollutants. It operates at ambient
conditions and is well suited to automatic control. Hydroxide precipitation
has been used to remove metal ions such as antimony, arsenic, chromium,
copper, lead, mercury, nickel, and zinc; Sulfide precipitation has mainly
been used to remove mercury, lead, and silver from wastewater, with less
frequent use to remove other metal ions. Carbonate precipitation has been
used to remove antimony and lead from wastewater. To achieve maximum
pollutant removals, chemical precipitation should be carried out in four
phases: 1) addition of the chemical to the wastewater; 2) rapid (flash)
mixing to distribute the chemical homogeneously into the wastewater; 3) slow
stirring to promote particle growth by various coagulation mechanisms
(flocculation); apd 4) clarification (or sedimentation or filtration) to
remove the flocculated solid particles.
The use of chemical precipitation technology as well as the availability
of performance data may be limited for several reasons. First, treatable raw
waste concentrations of product/process sources of priority pollutant metals
are not prevalent throughout the industry. Furthermore, plants that generate
process sources of metals and plants that utilize in-plant chemical precipi-
tation unit operations also tend to rely on co-dilution of metal-bearing
wastestreams by non-metal-bearing process wastewater as well as incidental
metal removals in end-of-pipe treatment systems. Fifty OCPSF plants in the
Section 308 Questionnaire data base report using chemical precipitation as an
in-plant treatment technology; however, very few facilities reported in-plant
chemical precipitation performance data.
Second, sulfide precipitation technology may generate toxic hydrogen
sulfide and may result in discharges of wastewaters containing residual levels
of sulfide. The generation of toxic hydrogen sulfide can be controlled by
VII-19
-------
ci
s
o
1
+•»
1
j
102..
10° -•
10-2 . .
Pb (CH)2
Cr (OH)3
Zn (CH)2
2 34 5 6 7 8 9 10 11 12 13 14
Figure VII-1: Solubility of Metal Hydroxides and Sulfides
as a Function of pH
Source: Treatability Manual. 1981.
VII 20
-------
maintaining the pH of the solution between 8 and 9.5. The discharge of waste-
waters containing sulfide can be controlled by carefully monitoring the amount
of sulfide added.
Third, in some instances, chemical precipitation may be limited by inter-
ference of chelating agents and complexed metal ions. Because of the varying
stabilities,of metal complexes and the wide variety of organic ligands in
OCPSF wastewaters, each plant with highly stable complexes has adapted or
should adapt its treatment system to control the concentrations of the metals
present in its process wastewater. Thus, control options for complexed
metals, and the degree to which control is necessary or cost-effective, are
unique to individual plants.
Several of the strategies employed by the OCPSF industry for treating
complexed metals in process wastewater are as follows:
• Destabilize the complex by chemically reducing the metal's valence to
zero. The released non-rionic metal is insoluble and can be captured
via agglomeration with other solids that are being separated from the
wastewater. Reductive destabilization is also effected by electro-
plating, in which case the metal is captured on the cathode.
• Destabilize the complex by degrading the organic ligand. The released
metal is then captured a's an insoluble salt by subsequent addition of
a reagent (e.g., lime, caustic, or sodium sulfide). In special cases,
ion exchange could be used to capture the metal ion.
• Capture the metal directly from the complex through the addition of a
reagent (e.g., sodium sulfide to a copper complex) that forms an
exceedingly insoluble salt of the metal.
• Concentrate the wastewater (e.g., in an evaporator) beyond the typi-
cally limited solubility of the metal-dye complex, so that it and
other solids separate as a sludge.
• Use carbon adsorption technology to capture the complexed metal from
the wastewater.via the organic ligand, which will adsorb on the carbon
as if it were not complexed.
Specific examples of the abovementioned precipitation technologies are
detailed below:
• Plant 1647. Complexed copper (cuprous, +2) in a dyestuff process
wastewater could not be precipitated effectively in a plant's combined
VII-21
-------
wastewater by lime addition. The segregated wastewater from the
dyestuff process was pretreated with sodium borohydride. Although
relatively expensive, the pretreatment destabilized the complex by
reducing the metal ion to copper (0), which was no longer amenable to
complexation by the organic ligand. Since copper (0) is insoluble,
the plant was then able to effectively remove the metal from the
combined wastewater via agglomeration with other solids precipitated
by the lime addition.
• Plant 1593. Copper (+2) and trivalent chromium (+3) are complexed
with organic ligands in metallized dyes manufactured at the plant.
The product is captured as a presscake on a plate-and-frame filter.
The filtrate, together with wastewater from floor drains and other
processes, is segregated into dilute and concentrated wastewater.
Concentrated wastewater is concentrated still further in an
evaporator, where most of the complexed metals separate as a residue
which is sent to a surface impoundment. Condensed overhead from the
evaporator and the dilute wastewater from a surge lagoon (flow
equalization), neither of which now contains concentrations of
complexed metals above their toxic thresholds, are combined as
influent to a powdered activated carbon (PAC) biological treatment
system.
Prior to segregating the dilute and concentrated wastewaters, the
combined process wastewater flow had to be pretreated with activated
carbon columns to protect the biota from the toxic effects of metals
released after complexing organic ligands had been biodegraded. Since
most of the combined flow was dilute wastewater that did not contain
complexed metals at toxic levels, the treatment system was modified to
segregate the concentrated wastewater for pretreatment to eliminate
the carbon column. Substantial operating cost savings were achieved
by these modifications.
• Plant 1572. Cadmium (+2) chelated with an unknown organic ligand is
used as a catalyst in a reactor. Reactor washout is treated with
sodium hydrosulfide to form a cadmium sulfide precipitate directly
from the complexed cadmium. The solids are captured by centri-
fugation, and the centrifugate is passed - through a rapid sand filter
to capture any fines. The solids from the centrifuge are saved and
are1 available to the plant as a cadmium reclaiming option with the
catalyst supplier.
• Plant 1769. Two organometallic products, tetraethyl lead (TEL) and
tetramethyl lead (TML), are produced at this plant. Although the
chemical bonding in organometallies differs from the metallized dye
complexes discussed previously, the treatment technology is the same
in principle. After adjusting the wastewater to a pH of 8 to 10 with
dilute sulfurie acid, sodium borohydride is added to reduce the ethyl
groups to ethane by hydride transfer. The released lead (+4) then
reacts with water to precipitate lead dioxide, which is captured in a
clarifier. The lead dioxide is recycled to refiners, which regenerate
the lead for sale to the market.
VII-22
-------
• Plant 2447. This plant manufactures oil-soluble dialkyl dithio-
carbamates and water-soluble dithiocarbamates of antimony, cadmium,
nickel, lead, and zinc. The metals in this plant's wasteyater are not
present as stable complexes but as salts of organic acids. This
example is given only to illustrate the wide variety of treatment
strategies used by the OCPSF industry to control metals.
Since metal dithiocarbamates have low solubility in water, a
precipitating reagent is readily available that is effective for con-
trolling these metals in the wastewater. The wastewater .is generated
in batches as washout from mixing tanks and reactors, and is collected
in a storage tank. Depending on the characteristics; of the batch, the
plant will either incinerate the waste, or route it to the wastewater
treatment system. Treatment consists of adding sodium dithiocarbamate
to precipitate the metals, and a coagulant (ferrous sulfate) to aid
settling of the solids in a clarifier.-
Wastewaters from the OCPSF industries generally do not contain high con-
centrations of metal ions. Rayon and certain acrylic fibers manufacturing,
however, generate elevated levels of zinc in wastewaters. Other industrial
processes may also have metals in their wastewaters due to use of metals in
chemical processing and as trace contaminants from raw materials and
equipment.
In the December 8, 1986, Federal Register Notice of Availability, the
Agency proposed to establish limitations for metals, from OCPSF plants with and
without end-of-pipe biological treatment in-place for BAT and PSES based upon
the use of hydroxide precipitation data from several metals industries. For
OCPSF waste streams with complexed metals, EPA proposed the use of sulfide
precipitation to achieve the same limitations. ,
Industry commenters strongly criticized several aspects of EPAfs proposed
approach. First, they argued that most priority pollutant metals are not
present in significant quantities in OCPSF wastewaters. They criticized^the
data base upon which EPA had estimated loadings for these pollutants. They
argued that these pollutants resulted not from OCPSF,processes, many,of which
do not use metals, but rather from non-process wastewaters (e.g., zinc and
chromium used as corrosion inhibitors and .often contained in cooling water
blowdown) or due to their presence in intake waters. The commenters concluded
that EPA should regulate only those metals present in OCPSF process waste-
waters as a result of the process use of the metals, applying the limits to
those wastewaters only.
VII-23
-------
To address these comments, EPA has conducted a detailed analysis of the
process wastewater sources of metals in the OCPSF industry. In response to
criticism that EPA has relied too heavily on limited Master Process File
metals data, EPA reviewed the responses to the 1983 Section 308 Questionnaire
to examine which metals were used as catalysts in particular OCPSF product/
processes, or were for other reasons likely to be present in the effluent from
these processes. When necessary, EPA contacted plant personnel for additional
information. The results of EPA's analysis, together with supporting documen-
tation, are set forth in Section V of this document.
Based upon this analysis, EPA has concluded that chromium, copper, lead,
nickel, and zinc are discharged from OCPSF process wastewaters at frequencies
and levels that warrant national control. However, EPA agrees that many OCPSF
wastewaters do not contain these pollutants or contain them only at insignif-
icant levels. At most plants, process wastewater flows containing these
metals constitute only a small percentage of the total plant OCPSF process
wastewater flow. As a result, end-of-pipe data obtained by EPA often do not
reflect treatment but rather reflect the dilution of metal-bearing process
wastewater by nonmetal-bearing wastewater. Thus, these data are unreliable
for the purpose of setting effluent limitations reflecting the use of best
available technology. Consistent with the comments, EPA has decided to focus
its regulations on metal-bearing process wastewaters only.
The concentration limitations are based upon the use of hydroxide
precipitation technology, which is the standard metals technology that forms
the basis for virtually all of EPA's BAT metals limitations for metal-bearing
wastewaters. Because very little OCPSF data on the effectiveness of hydroxide
precipitation technology are available, EPA has decided to transfer data for
this technology from the metal finishing industry point source category. A
comparison of the metals raw waste data from the metal finishing industry
data base with the validated product/process OCPSF raw waste data indicates
that the concentrations of the metals of concern are generally within an
acceptable range of concentrations found at metal finishing plants, except for
lead. Table VII-4 presents this comparison of available OCPSF and metal
finishing raw waste metals concentrations. With respect to lead, some OCPSF
plants' raw waste concentrations exceed the range of metal finishing raw waste
VII-24
-------
TABLE VII-4.
COMPARISON OF OGPSF AND METAL FINISHING
RAW WASTE METALS AND CYANIDE CONCENTRATIONS
Parameter
Range of OCPSF
Raw Waste Concen-
trations (mg/1)
Metal Finishing
Range of Effluent Long-
Metal Finishing Term Average
RaV Waste Concentration
Concentrations (mg/1) (mg/1)
Total Chromium (119)
Total Copper (120)
Total Cyanide (121)
Total Lead (122)
Total Nickel (124)
Total Zinc (L28)3
0.200-0.799
0.100-14.500
0.140- 5200.000
50.060-218.9002
0.270-4.000
0.400-20.000 ,
p.650-393.000
0.880-108.000
0.045-1680.000
0.052-9.701
!
1,070-167.000
0.630-175.000
0.572
0.815
0.180
0.197
0.942
0.549
1OCPSF raw, waste concentration data are limited to data from the Master
Process File for only product/processes that are validated process sources of
metals. ._
2OCPSF raw waste concentration data for lead are from two validated product/
processes that occur at the same plant. These values compare to the raw
waste concentrations for a lead battery manufacturing facility (identified as
plant #672 in the battery manufacturing industry study). The lead battery
plant raw waste concentration range was 2.21 to 295 mg/1 for lead; its
effluent long-term average concentration (after lime/hydroxide precipitation)
was 0.131 mg/1. The effluent data ranged from 0.01 to 0.81 mg/1.
3Excludes raw waste zinc concentrations from rayon and acrylic fiber
manufacturers. , ;,
VII-25
-------
concentrations. A comparison was made between the available OCPSF raw waste
concentrations and the data from the lead battery subcategory of the battery
.manufacturing point source category.. This comparison, as noted in Table
VII-4, shows that the battery manufacturing lead raw waste concentrations
encompass the range of OCPSF raw waste concentrations. Since hydroxide
precipitation achieves lead effluent concentrations at battery manufacturing
facilities that are as good as or better than those demonstrated by metal
finishing plants, EPA.believes that transfer of metal finishing lead data is
appropriate.
In addition, the metal finishing wastewater matrices contain organic
compounds that are used as cleaning solvents and plating bath additives. Some
of these compounds serve as complexing agents, and their presence is reflected
in the metal finishing industry data base. This data base contains hydroxide
precipitation performance results from plants with waste streams from certain
operations (electroless plating, immersion plating, or printed circuit board
manufacturing) containing complexing agents. This is important because the
data base reflects both treatment of waste streams containing complexing
agents and segregation of these waste streams prior to treatment.
The transfer of technology and limitations from the metal finishing
industry is further supported by the theory of precipitation. Given suffi-
cient retention time and the proper pH (which is frequently achieved by the
addition of a lime hydroxide), and barring the binding up of meta.ls in unusual
organic complexes (see discussion below), a metal exceeding its solubility
level in water can be removed to a particular concentration (i.e., the
effluent can be treated to a level approaching solubility for each constituent
metal). This is a physical/chemical phenomenon that is relatively independent
of the type of wastewater, barring the presence of complexing agents.
Some product/processes do have wastewaters that contain organic compounds
that bind up the metals in stable complexes that are not amenable to optimal
settling through the use of lime. EPA asked for comments in the December 1986
Notice on the use of sulfide precipitation in these situations. Industry
commenters argued that the effectiveness of this technology has not been
demonstrated for highly stable, metallo-organic chemicals. EPA agrees.
VII-26
-------
Strongly complexed priority pollutant metals are used or created, for
instance, in the manufacture of metal complexed dyestuffs (metallized dyes) or
metallized organic pigments. The most common priority pollutant metals found
in these products are trivalent chromium and copper. The degree of complexing
of these metals may vary among different product/processes. Consequently,
each plant may need to use a different set of unique technologies to remove
these metals. Thus, metals limits are not set by this regulation and must be
established by permit writers on a case-by-case basis for certain product/
processes containing complexed metals. These product/processes are listed in
Appendix B to the regulation and in Table X-5.
The list in Table X-5 has been compiled based upon the analysis
summarized in Section V of this document. EPA has concluded that all other
metal-bearing process wastewaters (whether listed in Table"X^5 or established
as metal-bearing by a permit writer) can be treated using hydroxide
precipitation to the levels set forth in the regulation.
As noted previously, since certain manufacturers of rayon and acrylic
fibers have significantly higher raw waste zinc concentrations than any other
OCPSF process wastewaters, the lime precipitation performance data received
from the subject facilities are only applicable to certain types of processes.
Table VII-5 presents a summary of zinc raw waste concentration data and lime
precipitation performance data from three rayon facilities, as well.as one
acrylic fibers plant that uses a zinc chloride/solvent process. Acrylic
fibers facilities using the zinc chloride/solvent process have been combined
with rayon facilities for the purpose of establishing BAT zinc limitations
because ,of their high raw waste zinc concentrations. By comparing the raw
waste concentrations and resulting effluent concentrations for zinc in Tables
VII-4 and VII-5, the fairly distinct differences in the two data sets are
obvious.
4. Chemical Reduction (Chromium Reduction)
Reduction is a chemical reaction process in which one or more electrons
are transferred to the chemical being reduced from the chemical initiating the
transfer (the reducing agent). The major application of chemical reduction
VII-27
-------
TABLE VII-5.
RAW WASTE AND TREATED EFFLUENT
ZINC CONCENTRATIONS FROM RAYON
AND ACRYLIC FIBERS MANUFACTURING
Plant No. Plant Type
Average
Influent Zinc No. of
Concentration .of Influent
(mg/1) Observations
Average
Effluent Zinc
Concentration
(mg/1)
No. of
Effluent
Observations
63 Rayon 143.471
387 Rayon 135.257
1012 Acrylic Fibers 287.686
1774 Rayon 15.570
365
354
363
346
3.847
2.198
2.291
2.409
253
258
358
346
VII-28
-------
involves the treatment of chromium wastes. To illustrate the reduction
process, the conversion of hexavalent chromium to trivalent chromium (chromium
reduction) is discussed below.
Chromium Reduction. A common chemical used in industrial plants for the
reduction of chromium is sulfur dioxide. Chemical reduction equipment usually
consists of one reaction tank where gaseous sulfur dioxide is mixed with the
wastewater. The reduction occurs when sulfurous acid, produced through the
reaction of sulfur dioxide and water, reacts with chromic acid as follows:
(1) 3S02 + 3H20 = 3H2S03
(2) 3H2S03 + 2H2Cr04 = Cr2(S04)3 + 5H20
According to the Section 308 Questionnaire data base, 11 OCPSF plants use
chromium reduction as an in-plant treatment technology.
5. Gas Stripping (Air and Steam)
Stripping, in general, refers to the removal of^relatively volatile com-
ponents from a wastewater by the passage of air, steam, or other gas through
the liquid. The stripped volatiles are usually processed further by recovery
or incineration.
Stripping processes differ according to the stripping medium chosen for
the treatment system. Air and steam are the most common media, with inert
gases also used. Air and steam stripping are described below.
Air Stripping. Air stripping is essentially a gas transfer process in
which a liquid containing dissolved gases is brought into contact with air and
an exchange of gases takes place between the air and the solution. In
general* the application of air stripping depends on the environmental impact
of the resulting air emissions. If sufficiently low concentrations are
involved, the gaseous compound can be emitted directly to the air. Otherwise,
air pollution control devices may be necessary.
'VII-29
-------
The exchange of gases takes place in the stripping tower. The tower
consists of a vertical shell filled with packing material to increase the
surfade area for gas-liquid contact, and fans to draw air through the tower.
The towers are of two basic types-countercurrent towers and crossflow towers.
In countercurrent towers, the entire airflow enters at the bottom of the
tower, while the water enters the top of the tower and falls to the bottom.
In crossflow towers, the air is pulled through the sides of the tower along
its entire height, while water flow proceeds down the tower.
The removal of pollutants by air stripping is adversely affected by low
temperatures, because the solubility of gases in water increases as
temperature decreases.
Steam stripping. Steam stripping is essentially a fractional
distillation of volatile components from a wastewater stream. The volatile
component may be a gas or an organic- compound that is soluble in the waste-
water stream. More recently, this unit operation has been applied to the
removal of water immiscible compounds (chlorinated hydrocarbons), which must
be reduced to trace levels because of their toxicity.
Steam stripping is usually conducted as a continuous operation in a
packed tower or conventional fractionating distillation column (bubble cap or
sieve tray) with more than one stage of vapor/liquid contact. The preheated
wastewater from the last exchanger enters near the top of the distillation
column and then flows by gravity countercurrent to superheated steam and
organic vapors (or gas) rising up from the bottom of the column. As the
wastewater passes down through the column, it contacts the vapors rising from
the bottom of the column. This contact progressively reduces the concen-
trations of volatile organic compounds or gases in the wastewater as it
approaches the bottom of the column. At the bottom of the column, the waste-
vater is heated by the incoming steam, which also reduces the concentrations
of volatile components to their final level. Much of the heat in the
wastewater discharged from the bottom of the column can then be recovered by
preheating the feed to the column.
VII-30
-------
Reflux (condensing a portion of the vapors from the top of the column and
returning it to the column) may be practiced if it is desired to alter the
composition of the vapor stream that is derived from the stripping column
(e.g., increase the concentration of the stripped material for recovery ,
purposes). There also may be advantages to introducing the feed to a tray
below the top tray when reflux is used. Introducing the feed at a lower tray
(while still using the same number of trays in the stripper) will have the
effect of either reducing steam requirements, as a result of the need for less
reflux, or yielding a vapor stream richer in the volatile components. The
combination of using reflux and introducing the feed at a lower tray will
increase the concentration of the volatile organic components in the overhead
(vapor phase) beyond that obtainable by reflux alone and increase the poten-
tial for recovery.
Stripping of the organic (volatiles) constituents of the wastewater
stream occurs because the organic volatiles tend to vaporize into the steam
until its concentration in the vapor and liquid phases (within the stripper)
are in equilibrium. The height of the column and the amount of packing
material and/or the number of metal trays along with steam pressure in the
column generally determine the amounts of volatiles that can be removed and
the effluent pollutant levels that can be attained by the stripper. After the
volatile pollutant is extracted from the wastewater into the superheated
steam, the steam is condensed to form two layers of generally immiscible
liquids—the aqueous and volatile layers. The aqueous layer is generally
recycled back to the steam stripper influent feed stream because it may still
contain low levels of the volatile. The volatile layer may be recycled to the
process, incinerated on-site, or contract hauled (for incineration,
reclaiming, or further treatment off-site) depending on the specific plant's
requirements.
Steam stripping is an energy-intensive technology in which heat energy
(boiler capacity) is required to both preheat the wastewater and to generate
the superheated steam needed to extract the volatiles from wastewater. In
addition, some waste streams may require pretreatment such as solids removal
(e.g., filtration) prior to stripping because accumulation of solids within
the column will prevent efficient contact between the steam and wastewater
VII-31
-------
phases. Periodic cleaning of the column and its packing materials or trays is
a necessary part of routine steam stripper maintenance to assure that low
effluent levels are consistently achieved.
Steam strippers are designed to remove individual volatile pollutants
based on a ratio (Henry's Lav Constant) of their aqueous solubility (tendency
to stay in solution) to vapor pressure (tendency to volatilize). The column
height and diameter, amount of packing or number of trays, the operating steam
pressure, and temperature of the heated feed (vastevater) are varied according
to the strippability (using Henry's Lav Constant) of the volatile pollutants
to be stripped. Volatiles vith lover Henry's Lav Constants require greater
column.height, more trays or packing material, greater steam pressure and
temperature, more frequent cleaning, and generally more careful operation than
do volatiles vith higher strippability (7-4). Although the degree to vhich a
compound is stripped can depend to some extent upon the vastevater matrix, the
basis for the design and operation of steam strippers is such that matrix
differences are taken into account for the volatile compounds the Agency has
evaluated.
Since Henry's Lav Constants vere such important design parameters, the
Agency initially proposed that, for consolidation purposes, toxic pollutants
could be grouped into three general ranges of Henry's Lav Constants termed
high, medium, and low; these groups are presented in Table VII-6. The pollu-
tants in the lov Henry's Lav Constant group vere determined to require
treatment other than steam stripping (i.e., carbon adsorption or in-plant
biological treatment). The remaining groups vere then used in the development
of^team stripping cost curves and in the transfer of steam stripping perfor-
mance data to toxic pollutants vithout performance data, depending on vhether
they fell vithin the high or medium grouping. For the purposes of this docu-
ment, these groupings are designated "strippability" groups.
According to the Section 308 Questionnaire data base, eight OCPSF plants
report using air stripping and 82 report using steam stripping as an in-plant
treatment technology. Steam stripping performance data collected during the
EPA 12-Plant Study or submitted by industry for selected volatile organic
compounds are presented in Table VII-7. The data indicate that high removal
VII-32
-------
X
r~
X
-*
'o 'o
X X
x
o
«-l
X
'o 'o
8
in
-*
o
a\
CO
S fi
X"s
*
• -H
03
1
5
s 2
OT
•H
CQ
1
1
S
a
s
O
CM
-s
a s
s
a
a «
»
O
i-f
O
O
•X
• H
u
•rt
T3
1 §
1 1
|
M
1 s
Si
2
v£>
CS
O
CD
N.X
0)
f~(
•H
M
4->
•H
a
,3
o
^-^
Lchloroe thane
Q
o
•
lloroethane (0
r*-4
(U
G
-Trichloroetha
CM
4->
Q)
,2-Tetrachloro
CM
O
"***
Lene Chloride
>,
— •
ichloropropane
Q
^-*
ichloropropenc
Q
i-»
mochloromethar
o
CM
^
omome thane (0,
u
r\
V
j_j
hloromethyl)Ei
o
i— i
n^
-Chloroisopro]
00458)
CM •
i — i
*ff .
orophenyl Phei
00912)
_c o
,_(
>,
s
0)
A
D->
rH
>l
c ^^
0) r~
JS t-H
O.ST
0 0
2 o
m
lichlorobenzen
i— i
O)
N
k-Trichloroben
*•/
r-t
o
o
•^
iH
O
I
0
n
•tn
;
* 3
g
O
t-H
O
CS
1
y^V
*
-H
EC
bO
re
n
(U CO C
C -—, CO (U
(y OS VO 'ri
g:
I 1 1
! a a
i
§
i
-g
§
O
-------
to
§
OS
o
W i
ggl
1-"
' O
bo
•rt
U
O in
CO O I
CO CO
O> V-* r-t 4-1
>,
,57
CS
(0 O
O O
CO
0) , -H
C CO CM
Q BE) CO T-H
o o
o o
O O
w
-------
il
a
•H
|a - coco cooqin^rHOrHco^q^qinqcoqcoq^rqcsrHqcsiq
Eo~
I
S^ii^S'ggga^sagjsas^a^aga^sag^
•^ ^ ^ 5Q ^
CJStn ?3 Ql~1 ^ rHrHrH iS
C-J CO rH ' ^
t)
gf S* 9 8 9 »S3 8"* 3 !
M CO CO
ES
a
Ifi
fj U U
133 -5?
73 rH CSJ
-------
efficiencies (e.g., most plant-pollutant combinations are over 992) can be
achieved for these volatile organic compounds. It should also be recognized
.that most treatment systems consist of several unit processes and that addi-
tional removal of organic compounds will likely occur, especially in systems
with biological treatment units.
Nitrobenzene performance data .from two plants in the OCPSF industry that
employed steam stripping followed by activated carbon are presented in Table
VII-8. The data indicate that a high removal efficiency (e.g., approximately
99*) can be obtained for this semi-volatile organic compound by using these
two processes. However, the data shown in Table VII-9 also indicate that com-
petitive adsorption may be occurring among nitrobenzene, the dinitrotoluenes
(2,4- and 2,6-dinitrotoluene), and the nitrophenols (2- and 4-nitrophenol and
2,4-dinitrophenol) which seem to favor adsorption of nitrophenols over nitro-
benzene because of their more attractive chemical affinity to the carbon. The
nitrotoluene data are not available because matrix interferences prevented
quantitation with the analytical methods that had been used.
6. Solvent Extraction
Solvent extraction, also referred to as liquid-liquid extraction, involves
the separation of the constituents of a liquid solution by contact with
another immiscible liquid for which the impurities have a high affinity. The
separation can be based either on physical differences that affect differen-
tial solubility between solvents or on a definite chemical reaction.
The end result of solvent extraction is to separate the original solution
into two streams-a treated stream and a recovered solute stream (which may
contain small amounts of water and solvent). Solvent extraction may thus be
.considered a recovery process since the solute chemicals are generally
recovered for reuse, resale, or further treatment and disposal. A process for
extracting a solute from solution will typically include three basic steps:
1) the actual extraction, 2) solvent recovery from the treated stream, and
3) solute removal from the extracting solvent. The process may be operated
continuously.
VII-36
-------
VII-37
-------
J
o
to
§
O
O
t* Q o 3
£g*
gg
M
o w"
< &
. w
>• N
H 3
<: «
Q 0
g
M
a
g
Cn
1
(
t
(
i
i
I
4.
•r
s
b
c
•n
rH
a
e
n)
1 CO
O -w
w c
ai a)
l-l S
0 r-l
1 M-l
0 >*-)
cq
o
M
•H C
c
0 3
C rH
0) M-l
£ U-l
0 W
O
4-> 4-1 I
rt C
Z IV
-il?
M-l
H
4-f
C
OJ
la
^ M-l
"* W
s
S 4J
-i C
J 0)
^ 3
! iH
14-1
H
)(
^ji»
^oooSS10"^^
CO S 3 g § » gj .H CO CM
^ o o o> ,-j oT o « co" rC
^N ^^ ^ *""' ^>*
JO Q vO CO Q Q Q Q /^
^ r^cMSgggg
t— ( ?
^n
S i i § § S § § § I
s a s a R » a g a °
Q >O r~% /^
§ 1 | § S I S g § i
vo sr m sr vo CM vo .H" ,-T ,4"
' ^ ^ ^ S *3 2 fe
aillSlg.sss
•.«v.^.srrv|^-*voco
co CM CM ,H ,H ,4 co" CM" CM" CM"
£°s?co£!2Su-1'-(<-<
co23S^3§2^cQ
sr
II I S g i S § S g
SSsls2Ss2§
SSSSSS32gs
3 § 5 S 1 1 1 I i g
"coc^^^^:^^^^
VII-38
-------
Solvent extraction is presently applied in two main areas: 1) the
recovery of phenol frpm aqueous wastes, and 2) the recovery of halogenated
hydrocarbon solvents from organic solutions containing other water-soluble
components. ,
Although effective in recovering solvents and other organic compounds for
recycle and reuse, solvent extraction is not a widespread wastewater treatment
technology because effluent concentration levels that are acceptable for
recycle and reuse are generally too high for wastewater discharge. According
to the Section 308 Questionnaire data base, 29 OCPSF plants use solvent
extraction as an in-plant control or a raw material reclamation technology.
Performance data are summarized for petroleum refining and organic chemical
manufacturing plants in Volume III of the Treatability Manual. The data show
a wide variation in removal efficiency, varying from 12 to 99 percent. Most
volatile organics are removed with greater than 90 percent efficiency, but
base/neutrals show removal efficiencies generally below 75 percent.
7. Ion Exchange
Ion exchange involves the process of removing anions and cations from
wastewater. Wastewater is brought in contact with a resin that exchanges the
ions in the wastewater with a set of substitute ions. The process has four
operations carried out in a complete cycles service, backwash, regeneration,
and rinse. The wastewater is passed through the resin until the available ex-
change sites are filled and the contaminant appears in the effluent (break-
through point). When this point is reached, the service cycle is stopped and
the resin bed is. backwashed with water in a reverse direction to that of the
service cycle. Next, the exchanger is regenerated (converted to original
form) by contacting the resin with a sufficiently concentrated solution of the
substitute ion. Finally, the bed is rinsed to remove excess regeneration
solution prior to the next service step.
Ion exchange is used in several ways. In industrial wastewaters, ion
exchange may be used to remove ammonia, arsenic, chromium, and nickel. It is
commonly used to recover rinse water and process chemicals, or to reduce salt
concentrations in incoming water sources.
VIIv-39
-------
According to the Section 308 Questionnaire data base,' only seven OCPSF
plants use ion exchange as an in-plant treatment technology. Based on the
limited number of OCPSF plants employing ion exchange and the absence of OCPSF
ion exchange performance data, ion exchange was not considered as a BAT or
PSES -candidate technology. Performance data for ion exchange systems in the
metal finishing industry are presented in Table VII-10. Although removal
efficiencies are greater for the electroplating and printing circuit board
plants (e.g., 91 to greater than 99%) than for plant #11065 (e.g., zero
removal to greater than 99%), the influent pollutant concentrations are also
much greater.
8. Carbon Adsorption
Activated carbon adsorption is a proven technology primarily used for the
removal of organic chemical contaminants from individual process waste
streams. Carbon has a very large surface area per unit mass and removes
pollutants through adsorption and physical separation mechanisms. In addition
to removal of many organic chemicals, activated carbon achieves limited
removal of other pollutants such as BOD5 and metals. Carbon used in a fixed
column, as opposed to being directly applied in granular or powdered form to a
waste stream, may also act as a filtration unit.
Activated carbon can be used as an in-plant treatment technology in order
to protect downstream treatment units such as biological systems from high
concentrations of toxic pollutants that could adversely affect system
performance. In-plant activated carbon treatment also enables removal of
pollutants from low volume waste streams before the waste streams mix with and
contaminate much larger volumes of wastewater, which would be more difficult
and costly to treat.
According to the Section 308 Questionnaire data base, 18 OCPSF plants are
known to use activated carbon as an in-plant treatment technology. Although
performance data for a specific individual in-plant carbon adsorption unit
prior to biological treatment were not available, the Agency collected
performance data from a carbon adsorption unit following steam stripping at an
OCPSF facility for which the carbon adsorption unit treated a separate process
VII-40
-------
TABLE, yn-io.
TYPICAL ICN EXCHANGE PERFORMANCE DATA.
Electroplating Plant ^,
Parameter
Zinc (Zn)
Cadmium (Cd)+3
Chromium (Cr+ )
Chromium (Cr )
Copper (Cu)
Iron (Fe)
Nickel (Ni)
Silver (Ag)
Tin (Sn)
Cyanide (CM)
Manganese (Mn)
Aluminum (A3.)
Sulfate (S04)
Lead (Pb)
Gold (Au) .
Prior To
Purifi-
cation
14.8
5.7
3.1
7.1
4.5
7.4
6.2
1.5
.1.7
9.8
4.4
5.6
After
Purifi-
cation
0.40
0.00
0.01
0.01
0.09
0.01 ,
0.00
0.00
0.00
0.04
0.00
0.20 ..
Removal
Efficiency
.<*>'.
• 97
100
100
100
98
100
100
•100
100,
100
100
96
Printed Circuit Board Plant
Prior To
*' Purifi-
, cation
_
-
• -
43.0
1.60
9.10
. 1.10
3.40
—
210.00
" 1.70
2.30
After
Purifi-
cation
-
—
—
o.io
0.01
o!io
0.09 :
*•"
2.00
0.01
. 0.10
Removal
Efficiency
(%)
100
99 .
*1 /Vv
100
91
97
99
99
% .
Plant #11065, which was visited and sampled, employs an ion exchange unit to remove metals
from rinsewater. The results of the sampling are displayed below.
POLLUTANT CONCHM33ATION (mg/1)
"~ Plant #11065 ~^~~
Parameter
TSS
Cu
Ni
Cr, Total
Cd
Pb
Day.l
Input To
Ion Exchange
6.0
52.080
0.095
0.043
0.005
0.010
Day 2
Effluent From
Ion Exchange
.4.0
0.118
0.003
0.051.
0.005
0.011
Removal
Efficiency
(%)
33
'100 "
97
0
0
0
Input To
Ion Exchange
1.0
189.3
0.017
0.026
0.005
0.010
Effluent From
Ion Exchange
1.0 -
0.20
0.003
0.006
0.005
0.010
Removal
Efficiency
(%)
0
100
82
77
0
0
Sources Development Document for Effluent Limitations Guidelines New Source Performance
Standards for the Metal Finishing Point Source Category, June 1983.
1Concentrations in mg/1.
vn-4i
-------
waste stream prior to discharge. This unit was sampled during the EPA
12-Plant Study. This plant manufactures only interrelated products whose
similar waste streams are combined and sent to a physical/chemical treatment
system consisting of steam stripping followed by activated carbon. The toxic
pollutants associated with these waste streams are removed by either steam
stripping or activated carbon, or a combination of both.
The Agency has decided to use this available performance data from the
end-of-pipe carbon adsorption unit as the basis for establishing BAT limits
for four pollutants (2-nitrophenol, 4-nitrophenol, 2,4-dinitrophenol, and
4,6-dinitro-o-cresol), and the combination of steam stripping and activated
carbon adsorption for nitrobenzene. Table VII-11 presents the performance
data for the carbon adsorption unit at this plant. These data show very good
removals (greater than 99%) for the carbon adsorption unit for 4,6-dinitro-
o-cresol, 2-nitrophenol, 4-nitrophenol, and 2,4-dinitrophenol. However, the
concentration data indicate that for 2,4-dinitrophenol and nitrobenzene the
carbon adsorption unit is experiencing competitive adsorption phenomena. As
shown in Table VII-9, this condition exists when a matrix contains adsorbable
compounds in solution that are being selectively adsorbed and desorbed.
9. Distillation
Distillation is a unit process usually employed to separate volatile
components of a waste stream or to purify liquid organic product streams. ' The
process involves boiling a liquid solution and collecting and condensing the
vapor, thus separating the components of the solution. The vapor is collected
in a vessel where it is condensed, resulting in a separation of materials in
the feed stream into two streams of different composition.
The distillation process is used to recover solvents and chemicals from
industrial wastes that otherwise would be destroyed by waste treatment.
Although effective in recovering solvents and other organic compounds for
recycle and reuse, distillation is not a widespread wastewater treatment tech-
nology because effluent levels that are acceptable for recycle and reuse are
generally too high for wastewater discharge. According to the Section 308
Questionnaire, 72 OCPSF plants use distillation as an in-plant control and/or
secondary product or raw material reclamation technology.
VII-42
-------
. 13
*»-*
s
fi-
A. A A
ON ON ON ON
R
c; o o
a s •
Q Q g
a
p*-.
00
CM
CS1 -*
vn-43
-------
No performance data are available for distillation as a wastewater
control technology.
10. Filtration
Filtration is a proven technology for achieving the removal of suspended
solids from wastewaters. The removal is accomplished by the passage of water
through a physically restrictive medium (e.g., sand, coal, garnet, or diato-
maceous earth) with resulting entrapment of suspended paniculate matter by a
complex process involving one or more removal mechanisms, such as straining,
sedimentation, interception, impaction, and adsorption. In-plant filtration
can serve to remove suspended solids and subsequently improve the performance
of downstream treatment units that may be adversely affected by larger parti-
cles in the waste stream. In addition, filtration units can serve to collect
solids with reclamation value from specific waste streams.
According to the Section 308 Questionnaire data base, 54 OCPSF plants use
filtration as an in-plant treatment technology. Performance data for filtra-
tion as an in-plant technology were not available in the OCPSF industry; how-
ever, performance data for hydroxide precipitation plus in-plant filtration
from the metal finishing point source category for TSS and selected metals are
presented in Table VII-12, along with the hydroxide precipitation performance
data from metal finishing for comparison purposes.
11. Reverse Osmosis
Reverse osmosis is a pressure-driven membrane process that separates a
wastewater stream into a purified "permeate" stream and a residual "concen-
trate" stream by selective permeation of water through a semipermeable
membrane. This occurs by developing a pressure gradient large enough to
overcome the osmotic pressure of the ions within the waste stream. This
process generates a permeate of relatively pure water, which can be recycled
or disposed, and a concentrate stream containing most of the pollutants
originally present, which can be treated further, reprocessed, or recycled.
Reverse osmosis systems generally require extensive pretreatment (pH
adjustment, filtration, chemical precipitation, activated carbon adsorption)
of the wastewater stream to prevent rapid fouling or deterioration of the
membrane surface.
VII-44
-------
TABLE VII-12.
PERFORMANCE DATA FROM HYDROXIDE PRECIPITATION AND
HYDROXIDE PRECIPITATION PLUS FILTRATION FOR
METAL FINISHING FACILITIES
Parameter
Hydroxide Precipitation
only
(mg/1)
Hydroxide Precipitation
Plus Filtration
(mg/1)
Total Suspended Solids
Chromium, Total
Copper
Lead
Nickel
Zinc
16.8
0.572
0.815
0.051
0.942
0.549
12.8
0.319
0.367
0.031
0.459
0.247
Source: Development Document for Effluent Limitations Guidelines New Source
Performance Standards ffcr the Metal Finishing Point Source Category,
June 1983.
VII-45
-------
Reverse osmosis has been used in industry for the recovery and recycle of
chemicals. Metals and other reusable materials can easily be separated from a
waste stream. Although reverse osmosis is slightly more effective than chemi-
cal precipitation for metals removal, it is very expensive and appropriate
only for low volume waste streams high in'dissolved solids.
12. Ultrafiltration
Ultrafiltration is a physical unit process, similar to reverse osmosis,
that is used to segregate dissolved or suspended solids from a liquid stream
through the use of semipermeable polymeric membranes. The membrane of an
ultrafilter forms a molecular screen that separates molecular particles based
on their differences in size, shape, and chemical structure. A hydrostatic
pressure is applied to the upstream side of a membrane unit, which acts as a
filter, passing small particles such as salts while blocking larger emulsified
and suspended matter. Ultrafiltration differs from reverse osmosis in the
size of contaminants passed. Ultrafiltration generally retains .participates
and materials with a molecular weight greater than 500, while reverse osmosis
membranes generally pass only materials with a molecular weight below 100.
Ultrafiltration has been used in oil/water separation and for the removal
of macromolecules such as proteins, enzymes, starches, and other organic
polymers. Ultrafiltration is presently not a widely used process but has
potential application to OCPSF wastewater treatment. Summary performance data
are available from EPA's Volume III Treatability Manual for the aluminum
forming, automobile and other laundries, rubber manufacturing, and timber
products processing industries and are presented in Table VII-13. The data
show a wide variation in removal efficiencies and effluent levels. An experi-
mental combined Ultrafiltration and carbon adsorption system does show
promise. This system consists of powdered activated carbon suspended in
wastewater. The mixture is then pumped through 20 ultrafilter modules
arranged in two parallel trains. Heavy metal removal data for this system are
presented in Table VII-13.
VII-46
-------
TABLE VII-13.
ULTRAFILTRATION PERFORMANCE DATA FOR METALS
IN LAUNDRY WASTEVATER-OPA LOCKA, FLORIDA
Parameter (mg/1)
Zinc
Copper
Lead
Chromium (total)
Cadmium
Raw
0.52
0.51
0.4
0.1
0.03
Supernatant
<0.20
0.14
0.1
<0.01
<0.02
Permeate
<0.20
0.06
0.01
' <0.01
<0.02
Source: Van Gils, G. and M. Pirbazari. August 1986. Development of *
Combined Ultrafiltration and Carbon Adsorption System for Industrial
Wastewater Reuse and Priority Pollutant Removal. Environmental
Progress 5(3):167-170.
VII-47
-------
13. Resin Adsorption
Resin adsorption is a process that may be used to extract and, in some
cases, recover dissolved organic solutes from aqueous wastes. Waste treatment
by resin adsorption involves two basic steps: 1) contacting the liquid waste
stream with the resin, allowing the resin to adsorb the solutes from the
solution, and 2) subsequently regenerating the resin by removing the adsorbed
chemicals, often accomplished by simply washing with the proper solvent.
Resin adsorption is similar in nature to activated carbon adsorption; the most
significant difference being that resins are chemically regenerated while
carbon is usually thermally regenerated, eliminating the possibility of mater-
ial recovery. Resins generally have a lower adsorptive capacity than carbon,
and are not likely to be competitive with carbon for the treatment of high
volume waste streams containing moderate or high concentrations of mixed
wastes with no recovery value.
Current applications of resin adsorption include removal of copper and
chromium both as salts and organic chelates, removal of color associated with
metal complexes and organics, and the recovery of phenol from a waste stream.
According to the Section 308 Questionnaire data base, no plants reported using
resin adsorption. No data, are available from other industries.
14. In-Plant Biological Treatment
For certain segregated waste streams and pollutants, in-plant biological
treatment is an effective and less costly alternative to carbon adsorption for
control of toxic organic pollutants, especially those which are effectively
absorbed into the sludge and are relatively biodegradable. In-plant
biological treatment may require longer detention times and certain species of
acclimated biomass to be effective as compared to end-of-pipe biological
treatment that is predominantly designated to treat BOD5. EPA has determined
that in-plant biological treatment with an acclimated biomass is as effective
as activated carbon adsorption for removing priority pollutants such as
polynuclear aromatics (PNAs) like naphthalene, anthracene, and pyrene; phenol;
and 2,4-dimethylphenol as shown in the sampling data collected at plant #1293
of the 12-Plant Sampling Study, which are presented later in this section.
Plant #1293 is a coal tar facility with flows of less than 50,000 gallons per
VII-48
-------
day (gpd), which generates the highest raw waste concentrations of these toxic
pollutants. Its treatment system consists of equalization, extended above-
ground aerated lagoon, and secondary clarification prior to discharge to a
POTW. This treatment system reduces the concentrations of all the above-
mentioned toxic pollutants to their respective analytical minimum levels.
After reviewing the performance data from this plant, the Agency deter-
mined that other relatively biodegradable toxic pollutants could also be
controlled by this type of dedicated biological treatment system (i.e., with a
minimum amount of dilution with other process wastewaters). This determina-
tion was made after review of performance data from selected end-of-pipe
biological treatment systems (plant 1948 and #2536) receiving wastewaters
whose main toxic pollutant constituents included the following: acrylo-
nitrile, bis (2-ethylhexyl) phthalate, di-N-butyl phthalate, diethyl
phthalate, and dimethyl phthalate.
The Agency has determined that these data are appropriate for use in
characterizing the performance of in-plant biological treatment based upon the
waste stream characteristics of the influent to the treatment systems. The
selected plants generate major sources of these pollutants.
According to the Section 308 Questionnaire data base, 33 OCPSF plants
report using some form of biological, treatment prior to discharge to an end-
of-pipe treatment system (direct dischargers) or POTW (indirect dischargers).
Table VII-14 presents the performance data for the three plants chosen by the
Agency to represent the performance of in-plant biological treatment.
D. END-OF-PIPE TREATMENT TECHNOLOGIES
1. Introduction
End-of-pipe treatment systems in the OCPSF industry often consist of
primary, secondary, and polishing or tertiary unit operations. In primary
treatment, physical operations are used to remove floating and settleable
solids found in wastewater. In secondary treatment, biological and chemical
processes are used to remove most of the organic matter. In polishing or
tertiary treatment, additional combinations of unit operations and processes
VII-49
-------
U-l
a
:
I
•AA.AAAAAAAAA
AAAAAAAAA
aa a. aaaaaa-9. 99999999
Q Q
9 a s s a a S-E IQ s s
a s s s
saaaaaaaaaaaaa.9999aa
OcOcMOOOOOOfO
8 6
a
S
-
SR
IN)
O\ r-t
01
* » £> gj g}
9
sj- v£>
§ §§§
rH r-H rH r-l
a
•H Q)
8 «3
rH O
f ^
<5 CM
rri 4^ -C +3
is* f A
»111 f
vn-50
-------
are used to remove other constituents that are not removed by primary or
secondary treatment. Many technologies have proven effective in removing
specific pollutants from the wastewaters produced by OCPSF plants. The selec-
tion of a specific end-of-pipe treatment scheme depends on the nature of the
pollutant to be removed and on engineering and cost considerations,: Data on
the frequency of application of specific primary, secondary, and polishing or
tertiary end-of-pipe treatment technologies are presented in Tables VII-15,
VII-16, and VII-17, respectively. Primary treatment technologies used by the
OCPSF plants to remove floating and settleable solids, to protect the biolog-
ical segment of the system from shock loadings, and to assure the efficiency
of biological treatment include neutralization (365 plants), equalization
(297), primary clarification (144), and nutrient addition (114). Secondary
treatment technologies used by OCPSF plants to remove organic matter include
secondary clarification (174 plants), activated sludge (143), and aerated,
lagoons (89). Polishing or tertiary treatment technologies used to remove
certain constituents not sufficiently removed by the primary and secondary
systems include polishing ponds (64 plants), filtration (41), and carbon
adsorption (21).
2. Primary Treatment Technologies
Although the final BPT, BAT, and PSES effluent limitations guidelines are
not based on these primary treatment technologies, many OCPSF facilities uti-
lize one or some combination of these technologies to enhance the performance
of subsequent treatment steps (e.g., biological). The Agency encourages the
use of any of the primary treatment technologies discussed to improve the
removal efficiency of the overall treatment system. <
a. Equalization
Equalization involves the process of dampening flow and pollutant
concentration variation of wastewater before subsequent downstream treatment.
By reducing the variability of the raw waste loading, equalization can: ;
significantly improve the performance of downstream treatment processes that
are more efficient if operated at or near uniform hydraulic, organic, and
solids loading rates and that reduce effluent variability associated with slug
VII-51
-------
=»*=
'!
;l
4—'
!J
I
o w
=H=
I 4-*
:l
a
CMIO,-),
CM O t-H r-l
•4-lOOOOrHOO CMO
CM
CM
B
II
I
I
•rl
III
g
833
VH-52
-------
1
*
cy> sf cs co
CO CM rH
CS CO
o ,o o ooo
=t4=
<•$
O
o o ooo o
O O CM O (0
2S
"8
"S y
I
O i-l
o •<
13
.H
CQ
'-ffl
I
g
ju 8
•8 S
I
I
s
"8
S
VH-53
-------
£3'
& eg'
"I"
o o o
» *J
;j
O CM O O
!i
CM T-I 1-1 o
!|
=8=
's s ^
=ft= p-i
CM tH r-l O
CM
X O 4J
I
I
S
via:-54
-------
raw waste loadings. Equalization is accomplished in a holding tank manufac-
tured from steel or concrete, or in an unlined or lined pond. The retention
time of the tank or pond should be sufficiently long to dilute the effects of
any highly concentrated continuous flow or batch discharges on treatment plant
performance. -
Equalization is reliable from both equipment and process standpoints, and
is used to increase the reliability of the flow-sensitive treatment processes
that follow by reducing the variability of flow and pollutant concentrations.
Equalization is a common treatment technology to the OCPSF industry. Accor-
ding to the Section 308 Questionnaire data base, 297 OCPSF plants use
equalization as a primary treatment technology.
b. Neutralization
Neutralization involves the process of adjusting either an acidic or a
basic waste stream closer to a neutral pH. Neutralization may be accomplished
in either a collection tank, rapid mix tank, or an equalization tank by mixing
acidic and alkaline wastes, or by the addition of chemicals. Alkaline waste-
waters are typically neutralized by adding sulfuric or hydrochloric acid, or
compressed carbon dioxide. Acidic wastewaters may be neutralized with
limestone or lime slurries, soda ash, or caustic soda. The selection of
neutralizing agents depends upon cost, availability, ease of use, reaction
by-products, reaction rates, and quantities of sludge formed. The most
commonly used chemicals are lime (to raise the PH) and sulfuric acid (to lower
the pH).
Neutralization of an excessively acidic or basic waste stream is
necessary in a variety of situations, including 1) the precipitation-of
dissolved heavy metals; 2) the prevention of metal corrosion and damage to
other construction materials; 3) preliminary treatment allowing effective
operation of the biological treatment process; 4) the providing of neutral pH
water for recycle uses; and 5) the reduction of detrimental effects in the
receiving water.
VII-55
-------
Neutralization is highly reliable with proper monitoring, control, and
proper pretreatment to control interfering substances. Neutralization is a
common treatment technology to the OCPSF industry; according to the Section
308 Questionnaire data base, 365 OCPSF plants neutralize their wastewaters.
c. Screening
Screening is the process of removing coarse and/or gross solids from
wastewater before subsequent downstream treatment, and is usually accomplished
by passing wastewater through drum- or disk-type screens. Typically, coarse
screens are stainless steel or nonferrous wire mesh with openings from 6 to
20 mm. Fine screens have openings that are less than 6 mm. Solids are raised
above the liquid level by rotation of the screen and are backflushed into
receiving troughs by high-pressure jets.
Screening has proven to be a very reliable process when properly designed
and maintained. According to the Section 308 Questionnaire data base,
49 OCPSF plants use screening as a primary treatment technology.
d. Grit Removal
Grit removal is achieved in specially designed chambers. Grit consists
of sand, gravel, cinders, or other heavy solid materials that have subsiding
velocities or specific gravities substantially greater than those of the
organic putrescible solids in wastewater. Grit chambers are used to protect
moving mechanical equipment from abrasion; to reduce formation of heavy de-
posits in pipelines, channels, and conduits; and to reduce the frequency of
digester cleaning that may be required as a result of excessive accumulations
of grit in such units.
Normally, grit chambers are designed to remove all grit particles with a
0.21 mm diameter, although many chambers have been designed to remove grit
particles with a 0.15 mm diameter. According to the Section 308 Questionnaire
data base, 41 OCPSF plants use grit removal as a primary treatment process.
e. Oil Separation (Oil Skimming, API Separation)
Oil separation techniques are used to remove oils and grease from waste-
water. Oil may exist as free or emulsified oil. The separation of free oils
VII-56
-------
and grease is accomplished by gravity, and normally involves retaining the
oily waste in a holding tank and allowing oils .and other materials less dense
than water to float to the surface. This oily top layer is skimmed off the
wastewater surface by.a mechanism such as a rotating drum-type or a belt-type
skimmer. Emulsified oil, after it has gone through a "breaking" step
involving chemical or thermal processes to generate free oil, can also be
separated using a skimming system.
Oil separation is used throughout the OCPSF industry to recover oil for
use as a fuel supplement or for recycle, or to reduce the concentration of
oils, which reduces any deleterious effects on subsequent treatment or
receiving waters. In the OCPSF industry, oil separation also removes many
toxic organic chemicals (typically :large non-polar molecules) that tend to
concentrate in oils and grease. However, since the removal of these toxic
pollutants is incidental to oil separation/removal, this treatment process was
not used as the technology basis for this final regulation. Still, the Agency
encourages its use to improve the performance of the overall treatment system
for removing unwanted floating oils and greases.
According to the Section 308 Questionnaire data base, 86 OCPSF plants use
oil separation; 58 use API separation (a common gravity oil separation based
upon design standards published by the American Petroleum Institute); and
111 practice oil skimming as a preliminary treatment technology. No OCPSF
performance data are available; however, data from the iron and steel manufac-
turing and electrical and electronic components industries are presented in
Volume III of the EPA Treatability Manual. The data show generally high
removal efficiencies for metals and toxic organics.
f. Flotation
Flotation is a process by which suspended solids, free and emulsified
oils, and grease are separated from wastewater by releasing gas bubbles into
the wastewater. The gas bubbles attach to the solids, increasing their
buoyancy and causing them to float. A surface layer of sludge forms, and is
usually continuously skimmed for disposal. Flotation may be performed in
several ways, including foam (froth), dispersed air, dissolved air, vacuum
VII-57
-------
flotation, and flotation with chemical addition. The principal difference
between these variations is the method of gas bubbles generation.
Flotation is used primarily in the treatment of wastewater streams that
carry heavy loads of finely divided suspended solids or oil. Solids having a
specific gravity only slightly greater than water, which would require abnor-
mally long sedimentation times, may be removed in much less time by flotation.
Thus, it is often an integral part of standard clarification.
According to the Section 308 Questionnaire data base, 31 OCPSF plants
used dissolved air flotation as a primary treatment technology. No OCPSF
performance data are available. The Volume III EPA Treatability Manual
presents performance data from textile mills, pulp and paper mills, auto and
other laundries, and petroleum refineries. The data show a median removal
efficiency of 61 percent for BOD5 and a median effluent concentration of
250 rag/1. Toxic removal efficiencies show large variations.
8> Clarification (settling, sedimentation)
Qlarification is a physical process used to remove suspended solids from
wastewater by gravity settling. Settling tanks, clarifiers, and sedimentation
ponds or basins are designed to let wastewater flow slowly and quiescently,
providing an adequate retention time to permit most solids more dense than
water to settle to the bottom. The settling solids form a sludge at the
bottom of the tank or basin. This sludge is usually pumped out continuously
or intermittently from settling tanks or clarifiers, or scraped out period-
ically from sedimentation ponds or basins.
Settling is used alone or as part of a more complex treatment process.
It is usually the first process applied to wastewaters containing high
concentrations of settleable suspended solids. Settling is also often used in
conjunction with other treatment processes such as removal of biomass after
biological treatment or removal of metal precipitates after chemical
precipitation. Clarifiers, in conjunction with chemical addition, are used to
remove materials such as dissolved solids that are not removed by simple
sedimentation (chemically assisted clarifiers are discussed later in this
section under polishing and tertiary treatment).
VII-58
-------
Clarification (or sedimentation or settling) is a common primary
treatment technology in the OCPSF industry; according to the Section 308 -
Questionnaire data base, 144 OCPSF plants use primary clarification.
h. Coagulation and Flocculation
Chemical coagulation and flocculation are terms often used interchange-
ably to describe the physiochemical process of suspended particle aggregation
resulting from chemical additions to wastewater. Technically, coagulation
involves the reduction of electrostatic surface charges and the formation of
complex hydrous oxides. Coagulation is essentially instantaneous in that the
only time required is that necessary for dispersing the chemicals in solution.
Flocculation is the time-dependent physical process of the aggregation of
wastewater solids into particles large enough to be separated by sedimenta-
tion. ' ' . ...... , .-., •... .. . '< •••• ,.
The purpose of coagulation is to overcome electrostatic repulsive surface
forces and cause small particles to agglomerate into larger particles, so that
gravitational and inertia! forces will predominate and affect the settling of
the, particles. The process can be grouped into two sequential mechanisms:
• Chemically induced destabilization of the repulsive surface-related
forces, thus allowing particles to stick together when contact, .between
particles is made.
• Chemical bridging and physical enmeshment between the non-repelling
particles, thus allowing for the formation of large particles.
'. " ~ . -.'''-
There are three different types of coagulants: inorganic electrolytes,
natural organic polymers, and synthetic polyelectrolytes.
Inorganic electrolytes are salts or multivalent ions such as alum
(aluminum sulfate), lime, ferric chloride, and ferrous sulfate.' The
inorganic coagulants act by neutralizing the charged double layer of colloidal
particles and by precipitation reactions. Alum is typically added to the
waste stream as a solution. At an alkaline pH and upon mixing, the alum
hydrolyzes and forms fluffy gelatinous precipitates of aluminum hydroxide.
These precipitates, partially as a result of their large surface area, act to
VII-59
-------
enmesh small particles and thereby create large particles. Lime and ion
salts, as well as alum, are used as flocculants primarily because of this
tendency to form large fluffy precipitates of "floe" particles.
Natural organic polymers derived from starch, vegetable materials, or
monogalactose act to agglomerate colloidal particles through hydrogen bonding
and electrostatic forces. These are often used as coagulant aids to enhance
the efficiency of inorganic coagulants.
Synthetic polyelectrolytes are polymers that incorporate ionic or other
functional groups along the carbon chain in the molecule. The functional
groups can be either anionic (attract positively charged species), cationic
(attract negatively charged species), or neutral. .Polyelectrolytes function
by electrostatic bonding and the formation,of physical bridges between
particles, thereby causing them to agglomerate. These are also most often
used as coagulant aids to improve floe formation.
The coagulation/flocculation and sedimentation process entails the
following steps:
• Addition of the coagulating agent to the liquid
• Rapid mixing to dispense the coagulating agent throughout the liquid
ml?lng ,t0 allov for cont*ct between small particles
and agglomeration into larger particles.
Coagulation and flocculation are used for the clarification of industrial
wastes containing colloidal and suspended solids. Coagulants are most
commonly added upstream of sedimentation ponds, clarifiers, or filter units to
increase the efficiency of solids separation. This practice has also been
shown to improve dissolved metal removal as a result of the formation of
denser, rapidly settling floes, which appear to be more effective in absorbing
and adsorbing fine metal hydroxide precipitates. Coagulation may also be used
to remove emulsified oil from industrial wastewaters. Emulsified oil and
grease is aggregated by chemical addition through the processes of coagulation
and/or acidification in conjunction with flocculation. Performance data for
VII-60
-------
coagulation/flocculation units are presented in the context of TSS and metals
removal in the section on chemical precipitation.
According to the Section 308 Questionnaire data base, 42 OCPSF plants
utilize coagulation and 66 OCPSF plants utilize flocculation as part of their
preliminary treatment systems.
3. Secondary Treatment Technologies
a. Activated Sludge ,
The activated sludge process is a biological treatment process primarily
used for the removal of organic material from wastewater. It is characterized
by a suspension of aerobic and-facultative microorganisms maintained in a
relatively homogenous state by mixing or by the turbulence induced by aera-
tion. These microorganisms oxidize soluble organics and agglomerate colloidal
and particulate solids in the presence of dissolved molecular oxygen. The
process can be preceded by sedimentation to remove larger and heavier solid
particles if needed. The mixture of microorganisms, agglomerated particles,
and wastewaters (referred to as mixed liquor) is aerated in an aeration basin.
The aeration step is followed by sedimentation to separate biological sludge
from treated wastewater. The major portion of the microorganisms and solids
removed by sedimentation are recycled to the aeration basins to be recombined
with incoming wastewater, while the excess, which constitutes the waste
sludge, is sent to sludge disposal facilities.
The activated sludge biomass is made up of bacteria, fungi, protozoa, and
rotifers. The bacteria are the most important group of microorganisms as they
are responsible for stabilization of the organic matter and formation of the
biological floe. The function of the biomass is to convert the soluble
organic compounds to cellular material. This conversion consists of transfer
of organic matter (also referred to as substrate or food) through the cell
wall into the cytoplasm, oxidation of substrate to produce energy, and
synthesis of protein and other cellular components from the substrate. Some
of the cellular material undergoes auto-oxidation (self-oxidation or
endogenous respiration) in the aeration basin, the remainder forming net
growth or excess sludge. In addition to the direct removal of dissolved
VII-61
-------
organics by biosorption, the biomass can also remove suspended matter and
colloidal matter. The suspended matter is removed by enmeshment in the
biological floe. The colloidal material is removed by physiochemical
adsorption on the biological floe. Volatile compounds may be driven off to a
certain extent in the aeration process. Metals are also partially removed,
and accumulate in the sludge.
The effectiveness of the activated sludge process is governed by several
design and operation variables. The key variables are organic loading, sludge
retention time, hydraulic or aeration detention time, oxygen requirements, and
the biokinetic rate constant (K). The organic loading is described as the
food-to-microorganism (F/M) ratio, or the kilograms of BOD5 applied daily to
the system per kilogram of mixed liquor suspended solids (MLSS). The MLSS in
the aeration tank is determined by the rate and concentration of activated
sludge returned to the tank. The organic loading (F/M ratio) affects the BOD5
removal, oxygen requirements, biomass production, and the settleability of the
biomass. The sludge retention time (SRT) or sludge age is a measure of the
average retention time of solids in the activated sludge system. Sludge
retention time is important in the operation of an activated sludge system as
it must be maintained at a level that is greater than the maximum generation
time of microorganisms in the system. If adequate sludge retention time is
not maintained, the bacteria are washed from the system faster than they can
reproduce themselves and the process fails. The SRT also affects the degree
of treatment and production of waste sludge. A high SRT results in carrying a
high quantity of solids in the system and obtaining a higher degree of treat-
ment and also results in the production of less waste sludge. The hydraulic
detention time is used to determine the size of the aeration tank and should
be determined by use of F/M ratio, SRT, and MLSS. The biokinetic rate
constant (or K-rate) determines the speed of the biochemical oxygen demand
reaction and generally ranges from 0.1 to 0.5 days'1 for municipal waste-
waters. The value of K for any given organic compound is temperature-
dependent; because microorganisms are more active at higher temperatures, the
value of K increases with increasing temperatures (7-5). Oxygen requirements
are based on the amount required for BOD5 synthesis and the amount required
for endogenous respiration. The design parameters will vary with the type of
vastewater to be treated and are usually determined in a treatability study.
VII-62
-------
The oxygen requirement to satisfy BOD5 synthesis is established by the
characteristics of the wastewater. the oxygen requirement to satisfy
endogenous respiration is established by'the total solids maintained in the
system and their characteristics. A detailed discussion of typical design
parameters used in the OCPSF industry and how these parameters are used in the
Agency's compliance cost estimates are presented in Section VIII.
Modifications of the activated sludge process are common, as the process
is extremely versatile and can be adapted for a wide variety of organically
contaminated wastewaters. The typical modification may represent a variation
in one or more of the key design parameters, including the F/M loading, aera-
tion location and type, sludge return, and contact basin configuration. The
modifications in practice have been identified by the major characteristics
that distinguish the particular configuration. The characteristic types and
modifications are briefly described as follows:
• Conventional. The aeration tanks are long and narrow, with plug flow
(i.e., little forward or backwards mixing).
• Complete Mix. The aeration tanks are shorter and wider, and the
aerators, diffusers, and entry points of the influent and return
sludge are arranged so that the wastewater mixes completely.
• Tapered Aeration/"' A modification of the conventional process in which
the diffusers are arranged to supply more air to the influent end of
the tank, where the oxygen demand is highest.
• Step Aeration. A modification of the conventional process in which
the wastewater is introduced to the aeration tank at several points,
lowering the peak oxygen demand.
, High Rate Activated Sludge. A modification of conventional or tapered
aeration in which the aeration times are shorter, the pollutants
loadings are higher per unit mass of microorganisms in the tank. The
rate of BOD removal for this process is higher than that of conven-
tional activated sludge processes, but the total removals are lower.
• Pure Oxygen. An activated sludge variation in which pure oxygen
instead of air is added to the aeration tanks, the tanks are covered,
and the oxygen-containing off-gas is recycled. Compared to normal air
aeration, pure oxygen aeration requires a smaller aeration tank volume
and treats high-strength wastewaters and widely fluctuating organic
loadings more effectively.
• Extended Aeration. A variation of complete mix in which low organic
loadings and long aeration times permit more complete wastewater
degradation and partial aerobic digestion of the microorganisms.
VII-63
-------
• Contact Stabilization. An activated sludge modification using two
aeration stages. In the first, wastewater is aerated with the return
sludge in the contact tank for 30 to 90 minutes, allowing finely
suspended colloidal and dissolved organics to absorb to the activated
sludge. The solids are settled out in a clarifier and then aerated in
the sludge aeration (stabilization) tank for 3 to 6 hours before
flowing into the first aeration tank.
• Oxidation Ditch Activated Sludge. An extended aeration process in
which aeration and mixing are provided by brush rotors placed across a
race track-shaped basin. Waste enters the ditch at one end, is
aerated by the rotors, and circulates.
Activated sludge is the most common end-of-pipe biological treatment
employed in the OCPSF industry. According to the Section 308 Questionnaire
data base, 143 OCPSF plants reported using activated sludge, 2 plants reported
using an oxidation ditch, and 8 plants reported using pure oxygen activated
sludge. Performance data for BOD5 and TSS removal are from the OCPSF Master
Analysis File and are presented in Table VII-18. The data show that activated
sludge treatment results in a median removal efficiency of 96 percent for BOD5
and 81 percent for TSS. For those plants meeting the BPT performance edit of'
95 percent removal of BOD5 or having an effluent BOD5 concentration no greater
than 40 mg/1, the BOD5 median removal efficiency is 98 percent and the TSS
median removal efficiency is 82 percent. (A detailed discussion of EPA's BPT,
data editing criteria is presented later in this section.)
b. Lagoons
A body of wastewater contained in an earthen dike and designed for
biological treatment is termed a lagoon or stabilization pond or oxidation
pond. While in the lagoon, the wastewater is biologically treated to reduce
the degradable organics and also reduce suspended solids by sedimentation.
The biological process taking place in the lagoon can be either aerobic or
anaerobic, depending on the design of the lagoon. Because of their low
construction and operating costs, lagoons offer a financial advantage over
other treatment methods and for this reason have become popular where
sufficient land area is available at reasonable cost.
Lagoons are used in industrial wastewater treatment for stabilization of
suspended, dissolved, and colloidal organics either as a main biological
VII-64
-------
co
EH
^N
m
§
m
1
. EH
"Y
M ^
|1
O CO
B co
CU EH
C4
*"*•
4-* rH
C ^
31?
rH *~s
4H CO
MH CO
W EH
| i p^ |
§ ^b
4-4 CO
fi CO
M H
rt
^ in
o a
B O
CU fO
(ti
rH
4-* ^x
B bO
' —
B bO
g ^-'
rH m
IH PQ
^
4-> bO
B B
CU v—'
rH &
4-1 O
«4H rH
pr*] pt^
W
o
•H
W
•rl
CO
C^ t"H CO 'i*** O» '•••' • ^ ^ g^
* * * * ^^ cxl Csl CO OO
l^ So cs S r- CO esi . *
r***
t *
•rt
U
CU
o
t
o
<4H
CU
CM
o
•x.
II
•rH
p£
1
b
TI
,£
a
a
c
•r
4->
C,
OJ
o
EH
•rt
W
CU
O OO CM NO CO C
in CM NO rt
co B
W
O
4H
CU
CM
00 CO CO CO CJS 43
CO NO iH ON -rt
II
£
K
X
O^ C5 OO C1^! CO C
CO CO O NO CU
tH CO H
U)1 • -
CU
b
*T-
r—
C/
CM -* CJS O CT> fl
co NO r~ CM -
CM NO >* a
; "i
1
o m rH o <
o m - •* o o
Sf tH O O >
in r*» o oo EH
CM «H CM NO CM
sf CM- NO
• co
m ui oo oo co
» « • • co
NO P- •"* CTv
CJS CTi 00 O>
CM iH CO in rH
CM CM NO -*
O O CT* NO CO
rxi. tH r^« r^* co
o"* in *H
ro
CO SO 'O O
NO rx o oo *
« • • •
• - • • • " " I,-
^^ ^_ ^j^ j- _J £*$ ^^ f^ ^J
f— 1 vj5 ^^ *^
•
W j^
LJ «H
•H *_j
B B CM B B (2
S 1 i 00 rtSBco
- .3 .5 -rt « a -rt -rt -rt pq
rt-SclSo <« T3 C X 0
S 55 *H nJ
'S3 S
o
MB
rt o
> o
4J CO
•rt CO
CU
0 CU
B 3
rt rH
M M-l
O CU
<4H
-------
treatment process or as a polishing treatment process following other
biological treatment systems. Aerobic, facultative, and aerated lagoons are
generally used for industrial wastewater of low and medium organic strength.
High strength wastewaters are often treated by a series of ponds; the first
one will be virtually all anaerobic, the next facultative, and the last
aerobic.
The performance of lagoons in removing degradable organics depends upon
detention time, temperature, and the nature of waste. Aerated lagoons gener-
ally provide a high degree of BOD5 reduction more consistently than the
aerobic and facultative lagoons. Typical problems associated with .lagoons are
excessive algae growth, offensive odors from anaerobic ponds if sulfates are
present and the pond is not covered, and seasonal variations of effluent
quality.
There are four major classes of lagoons that are based on the nature of
biological activity.
Aerobic Lagoons. Aerobic lagoons are shallow ponds that contain
dissolved oxygen (DO) throughout their liquid volume at all times. These
lagoons may be lined with concrete or an impervious flexible lining, depending
on soil conditions and wastewater characteristics. Aerobic bacterial
oxidation and algal photosynthesis are the principal biological processes.
Aerobic lagoons are best suited to treating soluble organics in wastewater
relatively free of suspended solids. Thus, they are often used to provide
additional treatment of effluents from anaerobic ponds and other partial
treatment processes.
Aerobic lagoons depend on algal photosynthesis, natural reaeration,
adequate mixing, good inlet-outlet design, and a minimum annual air temper-
ature above about 5°C (41°F), for a major portion of the required DO. Without
any one of these conditions, an aerobic pond may develop anaerobic conditions
or be ineffective or both. Because light penetration decreases rapidly with
increasing depth, aerobic pond depths are restricted to 0.2 to 0.3 m (0.6 to
1.0 ft) to maintain active algae growth from top to bottom. In order to
achieve effective pollutant removals with aerobic lagoons, some means of
VII-66
-------
removing algae (coagulation, filtration, multiple-cell design) is sometimes
necessary.
Anaerobic Lagoons. Anaerobic lagoons are relatively deep ponds (up to
6 meters) with steep sidewalls in which anaerobic conditions are maintained
by keeping organic loading so high that complete deoxygenation is prevalent.
Some oxygenation is possible in a shallow surface zone. If floating materials
in the waste form an impervious surface layer, complete anaerobic conditions
will develop. Treatment or stabilization results from anaerobic digestion of
organic wastes by acid-forming bacteria that break down organics. The
resultant acids are then converted to carbon dioxide, methane, and other end
products. Anaerobic lagoons are capable of providing treatment of high
strength wastewaters and are resistant to shock loads. These lagoons are
sometimes used to digest the waste sludge from an activated sludge plant.
In the typical anaerobic lagoon, raw wastewater enters near the bottom of
the pond (often at the center) and mixes with the active microbial mass in the
sludge blanket, which can be as much as 2 meters (6 feet) deep. The discharge
is located near one of the sides of the pond, submerged below the liquid
surface. Excess sludge is washed out with the effluent and recirculation of
waste sludge is not required.
Anaerobic lagoons are customarily contained within earthen dikes.
Depending on soil and wastewater characteristics, lining with various
impervious materials, such as rubber, plastic, or clay may be necessary. Pond
geometry may vary, but surface area-to-volume ratios are minimized to enhance
heat retention.
Facultative Lagoons. Facultative lagoons are intermediate depth ponds of
1 to 2.5 m (3 to 8 feet) in which the wastewater is stratified into three
zones. These zones consist of an anaerobic bottom layer, an aerobic surface
layer, and an intermediate zone. Stratification is a result of solids
settling and temperature-water density variations. Oxygen in the surface
stabilization zone is provided by reaeration and photosynthesis. The photo-
synthetic activity at the lagoon surface produces oxygen diurnally, increasing
the DO content during daylight hours, and decreasing it during the night. In
VII-67
-------
general, the aerobic surface layer serves to reduce odors while providing
treatment of soluble organic by-products of the anaerobic processes operating
at the bottom. Sludge at the bottom of facultative lagoons will undergo
anaerobic digestion, producing carbon dioxide and methane.
Facultative lagoons are customarily contained within earthen dikes.
Depending on soil and wastewater characteristics, lining the lagoon with vari-
ous impervious materials, such as .rubber, plastic, or clay, may be necessary.
Aerated Lagoons. Aerated lagoons are medium-depth basins of 2.5 to 5 m
(8 to 15 ft) in which oxygenation is accomplished by mechanical or diffused
aeration units and from induced surface aeration. Surface aerators may be
high speed, small diameter or low speed, large diameter impeller devices,
either fixed-mounted on piers or float-mounted on pontoons. Diffused aerators
may be plastic pipe with regularly spaced holes, static mixers, helical
diffusers, or other types. Aerated lagoons can be either aerobic or fac-
ultative. Aerobic ponds are designed to maintain complete mixing. Thus, all
solids are in suspension and separate sludge settling and disposal facilities
are required to separate the solids from the treated wastewater.
According to the Section 308 Questionnaire data base, lagoons are a
common secondary treatment technology in the OCPSF industry; 89 plants
reported using aerated lagoons, 24 plants reported using aerobic lagoons, and
12 plants reported using anaerobic lagoons. Performance data for BOD5 and TSS
removal from these lagoon systems were obtained from the OCPSF Master Analysis
File and are presented in Table VII-19. The data show that lagoon treatment
results in a median removal efficiency of 89 percent for BOD5 and 66 percent
for TSS, when all plants using only this secondary treatment process are
considered. For those plants meeting the BPT performance edit, the median
BOD5 removal efficiency is 90 percent and the median TSS removal efficiency is
75 percent,
c. Attached Growth Biological Systems
Attached growth biological treatment systems are used to biodegrade the
organic components of a wastewater. In these systems, the biomass adheres to
VII-68
-------
co
co
H
55
^
ifl i
Q
O ' -
PQ
•g
ON fr.
r*
1-1 *2
£* cl
11
^** rH
E™^ i~J
tt2
0 :
fi£-
CH
Cu
'a •
o
o
§
rH
nt
fj
O CO
§co
H
S-9
4-> I-H
C ^v
i-H
It
MH CO
C CO
H EH
i-H
03
!§"
a) pa
3s?
^
rH
<1> 6
i-H m
4H Q
M-l O
W CQ
y^t
rH
3 If
D ^
MHO"
a o
M pa
^*.
T3
4-> bO
C S
rH ^
MH O
MH rH
W Cr,
CO
CJ
•H
4->
CO
•rH
4->
n)
4->
CO
J_ Jj 1 — > ] lf~\ \^J +^ OA
^^ -i M ' ^^
VOVOVOCOCO T^'*c^ONVO
•n^f '
1
ro r»^. cO r^« CO
-<}• 00 f1^ '"^ O^ «* ' _^ j
V0 LO "I CN : ^ * ^^
1-1 S
CM
-k
•H
w
(U
CJ
e
§
o
MH
0)
CM
o
c»
II
•c
t
a
CO
c
c
b
n
1
0)
a
£H
4->
•H
m t~~ co en co S
VO CO CM O T-H "^
^ ^ 5 . s
§
• s
o
MH
VO CM CO O 3}
t-~ CO CM ON rH j.,
1 4J
•H
||
^
CO CM -Cf CM .-,
rH JJ c
^ Q
CO
C
in P~ rH CM VO k.
r- -* - ' •* "
CM ,
0)
o
f-l
o in on cr> vo
o vo CM - 4-»
s-i n)
0)
in M
O bO
pa o
C
in c
ON O
•H
M 4-*
0) 0)
rj (^
4-" 4J
•H C
(U 0)
CJ
co c:
03 O
E* CJ
4-> CO
•H CO
4->
0) C
0
-------
the surfaces of rigid supporting media. As wastewater contacts the supporting
medium, a thin-film biological slime develops and coats the surfaces. As this
film (consisting primarily of bacteria, protozoa, and fungi) grows, the slime
periodically breaks off the medium and is replaced with new growth. This
phenomenon of losing the slime layer is called sloughing and is primarily a
function of the organic and hydraulic loadings on the system. The effluent
from the system is usually passed to a clarifier to settle and remove the
agglomerated solids. Attached growth biological treatment systems are appli-
cable to industrial wastewaters amenable to aerobic biological treatment in
conjunction with suitable pre- and post-treatment. The process is effective
for the removal of suspended or colloidal materials, but less effective for
the removal of soluble organics. The two major types of attached growth
biological treatment processes used in the OCPSF industry are trickling
filters and rotating biologic contactors. These processes are described
below:
Trickling Filters. The physical unit of a trickling filter consists of a
suitable structure packed with an inert medium (usually rock, wood, or
plastic) on which a biological mass is grown. The wastewater is distributed
by either.a fixed-spray nozzle system or a rotating distribution system over
the upper surface of the medium and as it flows through the medium covered
with biological slime, both dissolved and suspended organic matter are removed
by adsorption. The adsorbed matter is oxidized by the organisms in the slime
during their metabolic processes. Air flows through the filter by convection,
thereby providing the oxygen needed to maintain aerobic conditions. Most
trickling filters are classified as either low- or high-rate, depending on the
organic and hydraulic loading. A low-rate filter generally has a media bed
depth of 1.5 to 3 meters (5 to 10 feet) and does not use recirculation.
High-rate filter media bed depths can vary from 1 to 9 meters (3 to 30 feet)
and require recirculation. The recirculation of effluent in high-rate filters
is necessary for effective sloughing control. Otherwise, media clogging and
anaerobic conditions could develop as a consequence of the high organic
loading rates employed.
Rotating Biological Contactors. The most common types of rotating
biological contactors consist of a plastic disk or corrugated plastic medium
VII-70
-------
mounted on horizontal shafts. The medium slowly rotates in wastewater (with
40 to 50% of its surface immersed) as the wastewater flows past. During rota-
tion, the .medium picks up a thin layer of wastewater, which flows over its
surface absorbing oxygen from the.air. A biological mass growing on the
medium surface adsorbs and coagulates organic pollutants from the wastewater.
The biological mass biodegrades the organic matter. Excess microorganisms and
other solids are continuously,removed from the film on the disk by shearing
forces created by the rotation of the disk in the wastewater. This rotation,
also mixes ,the wastewater, keeping sloughed solids in suspension until they
are removed by final clarification. :
According to the Section 308, Questionnaire data base, 8 plants report
using rotating biological contactors and 12 plants report using trickling
filters as a secondary treatment technology. Performance data fqr BOD5 and
TSS removal are from the OCPSF Master Analysis File and are presented in Table
VII-20. The data show that attached growth biological treatment results in a
median removal efficiency of 92 percent for BOD5 and 70 percent for TSS, when
all plants using only this secondary treatment process are considered. For
those plants meeting the BPT performance edits, the median BOD5 removal , .
efficiency is 92 percent and the median TSS removal efficiency is 70 .percent.
d. Secondary Clarification
The function of secondary clarifiers varies with the method of,biological
treatment utilized. Clarifiers in an activated sludge system serve a dual
purpose. In addition to providing a clarified effluent, they must also
provide a concentrated source of return sludge for process control. Adequate
area and depth must be provided to allow this compaction to occur while
avoiding rejection of solids into the tank effluent (7-6). Secondary clari-
fiers dn activated sludge systems are also sensitive to sudden changes -in flow
rates. Therefore, the use of multispeed pumps for in-plant wastewater lift
stations is strongly recommended where adequate flow equalization is not
provided (7^7).
Clarifiers in activated sludge systems must be designed not only for
hydraulic overflow rates, but also for solids loading rates. This is due
VII-71
-------
in
CO
pq
es
0) bo
9 e
l-l ^^
«W CO
V-I W
W E-i
O> bo
9 6
iH v-'
U-l CO
a co
M EH
rt
§bO
a
s^
i-H in
H-J Q
MH O
W PQ
s
bo
•-
(U
bo
CO
CO
CO
O O O O
o o o o
cr> en to o en
-* en t-i o
r- o en CM
vo in oo
CM es r*. co
• • • •
«N CM tH CvJ CM
«* CM CM i-H
in
IN
in r-» in CM en
en
CO
ey> o in o
vo m vo o
en en
o r^
o o
en
O O O
CO 13
O> 0>
W
Cti
a B co
•H -H PQ
C X O
•H «J
io
in
01
nt
- r~
oo co m o
• • • •
O O O iH
CM
CO
a CM
•g 33
•H cq
O) O»
(-4
0)
0)
M
bo
rt
u
B3
s<
M-) O
tO
•H CO
01
O 0)
gtt I
^t i
o a*
>4H
M T3I
ai a
CM nj
VII-72
-------
mainly to the need for both clarification and thickening in activated sludge
clarifiers to provide both a well clarified effluent and a concentrated return
sludge (7-6).
When the MLSS concentration is less than about 3,000 mg/1, the clarifier
size will normally be governed by hydraulic overflow rates. At higher MLSS
values, the ability of the clarifier .to thicken solids becomes the governing
factor. Therefore, solids loading rates become more critical in determining
tank size. Design size should be computed for both average and peak condi-
tions to ensure satisfactory effluent quality at all times (7-6).
Depth of clarifiers in activated sludge systems is extremely important.
The depth must be sufficient to permit the development of a sludge blanket,
especially under conditions when the sludge may be bulking. At the same time,
the interface of the sludge blanket and the clarified wastewater should be
well below the effluent weirs (7-6).
For long rectangular tanks, it is common practice to locate the sludge
withdrawal hopper about l/3«?to 1/2 the distance to the end of the tank to
reduce the effects of density currents (7-6, 7-7).
Typical design parameters for clarifiers in activated sludge systems
treating typical domestic wastewaters are also presented in Table VII-21. The
design of these clarifiers should be based upon an evaluation of average and
peak overflow rates and solids loadings. That combination of parameters that
yields the largest surface area should be used (7-6).
Clarifiers following trickling filters must effectively separate
biological solids sloughed from the filter media. .The design of clarifiers
following trickling filters is based on hydraulic overflow rates similar to
the method used for primary clarifiers. Design overflow rates must include
recirculated flow where clarified secondary effluent is used for recir-
culation. Because the influent SS concentrations are low, tank solids
loadings need not be considered. Typical design parameters for clarifiers
following trickling filters are also presented in Table VII-21 (7-6).
VII-73
-------
TABLE VII-21.
TYPICAL DESIGN PARAMETERS FOR SECONDARY CLARIFIERS
TREATING DOMESTIC WASTEWATER
Type of Treatment
Overflow Rate
(gpd/sq ft)
Average Peak
Settling Following
Trickling Filtration 400-600
Settling Following Air-
Activated Sludge
(Excluding Extended
Aeration) 400-800
Settling Following
Extended Aeration
Settling Following
Oxygen-Ac t i va t ed
Sludge with Primary
Settling
200-400
1,000-1,200
1,000-1,200
800
400-800 1,000-1,200
Solids Loading1 Depth
(Ib solids/day/sq ft) ft
Average Peak
20-30
20-30
25-35
10-12
<50 12-15
<50 12-15
<50 12-15
Allowable solids loadings are generally governed by sludge settling
characteristics associated with cold weather operations. e"-Ling
Source: Process Design Manual for Upgrading Existing Wastewater Treatment
Plants, EPA 625/l-71-004a, October 1974.
VII-74
-------
e. Operating, Managing, and Upgrading Biological Treatment Systems
This section identifies methods by which biological treatment systems in
the OCPSF industry may modify their existing facilities in order to upgrade or
improve performance. Most of the upgrades discussed pertain to activated
sludge and aerated lagoon systems, since these are the biological treatment
systems most commonly used in the OCPSF industry and the systems most amenable
to operational and design modifications. Approaches to upgrading biological
treatment units include adding unit treatment processes, modifying the design
and operational parameters of existing units, acclimating existing bacteria to
certain toxicants or using bioaugmentation (the addition of acclimated types
of bacteria bred to remain active under a variety of adverse conditions),
particle size reduction, nutrient addition, and the addition of powdered
activated carbon (PAC) to aeration units.
In some cases, the only means of improving the performance of a
biological treatment system is to add additional unit treatment processes.
Aeration basins and clarifiers are sometimes added to accommodate higher waste
loads or to address inadequacies in the original treatment plant design. The
addition of primary.unit treatment such as equalization improves system
performance by diluting slugs of concentrated wastes, minimizing routine
variations in influent wastewater flow and pollutant concentration, and
removing suspended particles. Preaeration basins are often added to raise
wastewater DO levels and improve the treatability and settling characteristics
of the wastes. Postaeration basins are added to systems to raise the DO in
treatment plant effluent before it flows into receiving streams. Microscreen
and filtration units can be added to improve suspended solids removal prior to
effluent discharge. In summary, there are a number of unit processes
available that can be added to a facility, provided that land is available, to
address specific treatment problems.
Upgrading existing bioreactor facilities can include adding chemical and
physical treatments such as the addition of polyelectrolytes to clarifiers to
improve solids settling or the installation of a surface skimmer to a pre-
treatment unit to accomplish oil and scum removal. Operational changes
affecting the quantity and species of microorganisms in a system, however, are
VII-75
-------
often the most significant with regard to improving the removal of pollutants
and increasing a treatment system's capacity to handle large raw waste loads.
Experience at some facilities indicates that operation of an activated sludge
plant to maintain a stable mixed liquor fauna (i.e., maintain a specific
distribution of bacterial species), rather than operation based on a constant
aeration rate or MLSS concentration, yields more consistent treatment of BOD
and priority pollutants (7-8). Thus, operational changes and unit treatment
modifications should be planned giving appropriate consideration to this
approach. Many of the concepts for improving the performance of biological
units discussed below are presented in the context of activated sludge and
aerated lagoon systems; however, in many cases they also apply to other types
of biological units, such as fixed film reactors.
As previously discussed, flow equalization is important in improving the
treatability of a waste stream by minimizing variations in wastewater
characteristics, such as temperature, pH, and pollutant concentrations. One
facility in the OCPSF industry improved the equalization of its wastewater by
removing several feet of sedimentation from a primary clarifier, thus
increasing the wastewater detention time. This plant also added heat
exchangers upstream of the treatment units to lower the wastewater temperature
and provide a more uniform wastewater temperature year round.
Modifications to the operations of activated sludge units include
changing influent flow patterns; altering the division, mixing, and aeration
characteristics of the tanks; and recycling sludge from the secondary
clarifier to one or more locations in the treatment train. Step aeration,
introducing primary effluent at several locations in the aeration basin, can
be used to upgrade the performance of a plant with high pollutant loadings
(7-9). Distribution of the waste equalizes the loading in the aeration basin
and enables the microorganisms to function more efficiently.
In situations where a treatment system needs to be modified to handle an
increased waste load, a conventional single tank activated sludge process can
be converted into a two-stage contact stabilization process. The main advan-
tage of contact stabilization is that it operates with a much shorter
hydraulic retention time and hence enables the facility to treat a larger
VII-76
-------
waste load. In other situations where oxygen requirements are not being met
and the facility has extra capacity, oxygen supply can be improved by creating
a complete mix activated sludge system from a contact stabilization or
conventional activated sludge unit. Another approach to improving oxygen
supply is to convert a standard air supplied aeration system to a pure oxygen
system.
Pure oxygen systems are recommended for situations where wide fluctua-
tions occur in the organic loading to a plant and for strong industrial waste-
water. Since they are more efficient than conventional aeration systems, they
can be used to increase the treatment capacity of existing plants. A means of
further improving a pure oxygen or air supplied aeration system is to use
diffusers that produce smaller diameter bubbles (and hence increase the
surface area to bubble volume ratio), and to increase the contact time between
the bubble and the wastewater.
In some treatment train configurations, it is possible to create a second
biological treatment unit by recycling sludge from a secondary clarifier to a
preaeration unit. As presented in the discussion of summer/winter issues,
this was done by plant #2394 in the OCPSF industry to improve the performance
of its treatment plant during cold weather. An additional benefit of
recycling sludge in this manner is that there is usually a decrease in the
total sludge volume generated. Plant #2394 used 100 percent recycle and hence
had no waste sludge during winter months.
Fixed film biological treatment units sometimes have problems associated
with waste distribution and waste loading. Low flows in trickling filter
plants may result in poor distribution of wastewater over the filter media.
Recirculation of part of the treatment plant effluent will increase the flow
through the plant and improve the motion of the distribution arm. An approach
to increasing the capacity or improving the performance of some trickling
filter plants is to replace traditional filter media usually consisting of
stones with synthetic media designed to have a much larger surface area.
Efficient operation of a bioreactor is dependent on maintaining viable
populations of bacteria. Organic priority pollutant removal is often
VII-77
-------
problematic as the pollutants often inhibit the growth of organisms respon-
sible for their degradation (7-10). To efficiently degrade these organics,
the inhibitory levels should be determined and should not be exceeded in plant
operations. In addition, bacteria can be acclimated to certain toxicants by
subjecting the activated sludge to an acclimation program or by using
"pre-acclimated" bacteria, the latter process being called bioaugmentation.
Bioaugmentation has also been used to supplement plants in cold weather with
specialized bacteria that maintain high levels of biodegradation activity at
wastewater temperatures as low as 40°F. In addition, bioaugmentation has been
proven to improve oxygen transfer, reduce sludge generation, and improve
sludge settling characteristics. Furthermore, bioaugmentation will greatly
reduce the time needed for recovery from a shock loading. Preserved bacteria
can be added to a biological treatment system as needed to maintain existing
populations and to increase biodegradation capabilities in the event of a
chemical upset.
The efficiency of a biological system can be improved by reducing the
particle size of solids in the influent through pretreatment with coagulation/
flocculation, sedimentation, or other processes. Rates of adsorption,
diffusion, and biochemical reaction are all enhanced by smaller particle size.
Particles smaller than 1 x 10~6 meter in diameter can be biochemically
degraded at a much faster> rate than larger particles (7-11). This is due to
the increase in surface area to mass ratio as particle size decreases. Higher
quality secondary effluent from the biological treatment unit will result in
subsequent improvements in the performance of downstream units such as filtra-
tion and activated carbon units.
Secondary clarification systems can also be modified or operated
differently in order to upgrade or improve TSS effluent performance. An
Agency study of full-scale municipal treatment systems shows that rectangular
clarifier modifications such as reaction baffles and other flow-modifying
structures at clarifier inlets resulted in a 13.8 percent reduction in
effluent TSS. Also, the additional installation of a stop-gate in a channel
upstream of the aeration basins to reduce large flow transients to a rectang-
ular secondary clarifier resulted in 31.5 percent lower effluent TSS levels
than the unmodified clarifier without the stop-gate. In another case, this
VII-78
-------
study also shows that slowing the rotational speed of hydraulic sludge removal
mechanisms .in circular clarifiers to 56 percent of its design speed reduced
effluent TSS by 10,5 percent. Also, the additional installation of a
cylindrical ring baffle/flocculation chamber in secondary clarifiers resulted
in 38.5 percent lower effluent TSS levels than the unmodified secondary
clarifier (7-7).
For a biological system to function properly, nutrients such as organic
carbon, nitrogen, and phosphorus must be available in adequate amounts. While
domestic wastewaters usually have an excess of nutrients, industrial waste-
waters are sometimes deficient. If a deficiency is identified, the perfor-
mance of an industrial wastewater treatment plant can be improved through
nutrient addition. According to the Section 308 Questionnaire data base,
114 OCPSF plants utilize nutrient addition prior to biological treatment.
Removal of organics can be enhanced by mixing powdered activated carbon
(PAC) in the aeration basin of a biological treatment system (7-12). PAC
improves treatment in the activated sludge process because of its adsorptive
and physical properties. Lighter weight organics, such as phenols, appear to
adsorb reversibly on the carbon. Use,of PAC can dampen the shock effects of
concentrated slugs of inhibiting organics on the bacteria culture, as the
organics will initially adsorb on the carbon. The PAC can be bioregenerated
as these lighter weight organic species desorb from the PAC and are degraded.
Heavier organics, such as the residual metabolic end products, appear to
adsorb irreversibly on the PAC. PAC also helps to remove pollutants by
extending the contact time between the pollutant and the biomass. When
adsorbed by the carbon, pollutants settle into the sludge and contact time
with the biomass is extended from hours to days. The waste sludge that
contains powdered carbon is removed from the activated sludge system,
dewatered, and either disposed of or regenerated. The regenerated carbon may
require an acid wash to remove metals as well as other inorganic materials to
improve the adsorption capacity.
e.
Summer/Winter
In commenting on the 1983 proposal and subsequent notices, many commen-
ters asserted that EPA incorrectly evaluated the effect of temperature on
VII-79
-------
Diological treatment systems and incorrectly concluded that temperature is ,,„,
important in the context of effluent limitations guidelines. They claimed
that one element of this incorrect analysis was EPA's deletion of nine plants
from the data base simply because they had been issued "Best Professional
Judgement" NPDES permits with separate compliance standards for summer and
winter months. They claim that this is an arbitrary decision that virtually
ensures that the effect of temperature will not be considered in estimating
effluent variability.
EPA has studied the effects of temperature variations on biological
treatment system performance, in the OCPSF industry and disagrees with these
comments. With regard to operations in warm climates, the Agency believes
that warmer than average temperatures do not have any significant effect on
biological treatment efficiency or variability. However, algae blooms in
ponds can be a wastewater treatment problem in ponds located in warm climates.
Nonetheless, polishing ponds are not part of the technology basis for BPT
limitations. Also, EPA was not able to associate algae bloom problems'with
any elements of biological treatment (aerated lagoons, clarification, equali-
zation basins, etc.). Consequently, EPA believes that algae growth problems
in warm climates are not relevant to the promulgated BPT regulations.
In order to evaluate winter performance of biological treatment systems,
EPA has analyzed BOD5 removal efficiency, BOD5 effluent concentration, and
operational changes for 21 plants reporting daily data and other plants
located in various parts of the country. These analyses indicated that there
is a slight reduction in average BOD5 removal efficiency and a small increase
in average effluent BOD5 concentrations during winter months for some plants.
However, other plants were able to maintain a BOD5 removal efficiency of
95 percent or greater and effluent BOD5 concentrations characteristic of good
operation during the entire year. The analysis also suggests that the plants
with lower efficiencies are affected as much by inefficient operation
practices as by winter temperature considerations. A discussion of
inefficient operating practices used by some plants as well as practices
employed by plants achieving superior all year performance is presented below.
The adoption of practices used by plants with higher winter efficiencies
should result in improved winter effluent quality.
VII-80
-------
EPA has determined that temperature effects can be mitigated by opera-
tional and technological changes so that compliance with BPT limitations using
biological treatment is possible for all OCPSF plants with well-designed and
well-operated biological systems. As also discussed below, the potential
effects of winter operations are included in the plant-specific factors that
affect derivation of the variability factors used to establish effluent
limitations guidelines. In addition, EPA has developed costs for plants that
need to upgrade their winter-time biological treatment operation to comply
with the promulgated BPT limitations.
: .Regarding the deletion of nine summer/winter plants' data from the data
base, the Agency notes that because these plants were subject to meeting two
different sets of permit limits, they had no incentive to attempt to achieve
uniform limitations throughout the year. Not suprisingly then, the daily data
from these plants exhibit a two-tier pattern. These data can be characterized
by-two means, and the variability of these data over a 12-month period is
fundamentally different from the data from plants required to meet only one
set of permit limits. Consequently, the data generated during these periods
are not representative of well-operated biological treatment, which as noted
ibove, is capable of uniform treatment throughout the year as demonstrated by
£ number of plants. Another problem with daily data from these plants is that
during certain periods of the spring and fall, these plants may be able to
Operate their treatment plants at less than full efficiency because they are
required to meet the less stringent set of permit limits.
: In summary, the Agency believes that it has accounted adequately for the
effect of temperature changes on biological treatment performance in its
variability analysis by including in the variability data, base a number of
well-designed and well-operated plants from climates with significant tempera-
ture variation. The inclusion of data from plants with summer/winter permits
would result in an overestimate of the variability of biological treatment
operations in the OCPSF categories.
The detailed analyses described below are based on two sets of data that
were analyzed in order to determine the effect of temperature on the treatment
of BOD and TSS. The first set included the OCPSF daily data base, which
VII-81
-------
contained daily data from 69 plants. Of these, 48 were excluded from the
final BPT daily data base analysis for a variety of reasons, including greater
than 25 percent non-process wastewater dilution, summer/winter NPDES permit
limits, changes in treatment system during sampling, non-representative
treatment, and effluent data after post-biological tertiary treatment. As a
result, daily data from 21 plants formed the basis of the variability
component of the BPT limits and were included in the summer/winter analysis.
These 21 plants are #s 387, 444, 525, 682, 741, 908, 970, 1012, 1062, 1149,
1267, 1407, 1647, 1973, 1977, 2181, 2430, 2445, 2592, 2626, and 2695. The
second data set includes 131 plant responses to a Section 308- Survey question
regarding average winter and average summer performance and operating para-
meters that were gathered to highlight practices used to accommodate cold
weather conditions.
The principal parameters evaluated for correlation with temperature were
average effluent BOD5 and TSS concentration, and BOD5 removal efficiency. In
addition, two plants that had made operational changes to increase winter
efficiency were also evaluated.
525s Removal Efficiency. Of the 21 plants with long-term daily data,
14 had sufficient BOD5 influent and effluent data (total BOD5 values were
used) to enable the calculation of BOD5 monthly removal efficiencies. Six
plants <#s 387, 444, 1149, 1267, 2626, and 2695) were not used because they
had no BOD5 influent values, and plant ;#908 was eliminated because its
geographic location in Puerto Rico made any seasonal distinctions meaningless
The plants that were used had a minimum of three influent and effluent
values each month; if there were time periods where fewer values were avail-
able, these specific time periods were excluded from the analysis (Plant 1062
had only one influent measurement between 1-1-79 and 7-31-79 and plant 2592
had no influent sampling between 12-1-79 and 7-9-80). For each plant where
sampling occurred over a period exceeding 1 year, values for the same month
but different years were averaged together.
The monthly efficiencies were derived by use of the formula
VII-82
-------
Fraction removed = 1 -
[average BOD effluent for the month]
average BOD influent for the monthJ
'The result of the efficiency analysis is presented in Table VII-22.
As can be seen, the annual average BOD5 removal efficiency is 95 percent.
Seven of the.fourteen plants (#s 682, 970, 1062, 1647, 1977, 2181, and 2430)
had greater than.95 percent removal of BOD5 throughout the year. If the '
winter months are defined to be January-February-March and the summer months
are defined to be June-July-August, two plants had removal efficiencies in the
winter months that were greater than or equal to those in the summer: months.
Plant 1062 had 97 percent removal efficiency in both the winter and summer
months, and Plant 2430 had 99 percent removal efficiency in the winter months
and 98 percent removal efficiency in the summer. In addition, five plants
(ts 682, 970, 1647, 1977 «and 2181) had average winter removal efficiencies
within 1 percent of their average summer removal efficiencies.
The 14 plants are located in three different geographical regions. Plant
data were analyzed by region, with subset I including data from,the five
plants located in the ncnrth (W, IL, RI, lA, IN), subset II including data
from the six plants located in the south '(TX, GA, LA, SC), and subset III
including data from the three plants located in the middle-latitudes (VA, NG).
These results are presented in Tables VII-23, VII-24, and VII-25. Monthly
average removal efficiencies for each plant were obtained, and these were
combined into an overall monthly average for each subset. Plants located in
the northern region had the highest average removal efficiency (northern
plants - 98 percent; southern plants - 95 percent; middle latitude - 89
percent). In the northern region, four of the five plants (682, 1062, 1647,
and 2181) had removal efficiencies greater than 95 perc.ent throughout the ;
^Although it was also possible to obtain monthly efficiencies by calculating
daily efficiencies and-averaging them for each month, such a method would
have resulted in elimination of many data points when only influent or efflu-
ent values, not both, were available for a specific day- Also, because
retention times are generally greater than 1 day, and because wastewaters are
mixed during treatment, an effluent value cannot necessarily be correlated
with an influent value for that same day or for any other particular time.
VII-83
-------
1 I
oooooooooo o
S5S§88§88S8:8: fe
ooeoo'ooo
o o o o o
o o o d d
CMJOOOOOOOOOOO
CM * OOOOOO«-«-
ooooeiociodo.cse ca
«-;ooooooo
oooo o
OOOOOOOOOOO O
ooooooooo
o o o o
Ri
8 -S ft S ft
'
^-•00000 ooooo'
ooooooooo o
ooooooraoo
o o o o o o o
o o o o
ooooooooo
ft
ooo o
ooooeoeaood
£ t * 2 £ S
II
2 „ a m
VII-84
-------
1
(3
82 ~
& =
1-1 —I
o —
u. *
u- <
UJ _
in •
> •§
UJ (/>
8
OOOOC3C>C30CSOP O
8;
§ g fe
OC3OOC3OOOOOOO O
8 85 fc 8 fc 8 8 8 8 S'fc 8
K &* Ot OO
K Qs ^K CS
esooooociei o 'o o o «a
g 8 S £: 8 8.
8
8: 8
C3OOC5OOOOOOOO
I * - ... a 1 I
1
i- O ••-
01 !o o.
u- I <
>.
.
~5 ? 01 o o o
-3 < » O ae o
VII-85
-------
cj
>
« w
SI
>- i>
ooooooooooo
§
tvi
o>
oooooooooo
ooooooooooo
oooooooooooo o
s I -s - n^«sjs|| Z
3fe""-..'l>>-34->oSS C
i|fefe.^S = 58-t5gg S
^"•3:«z-3-3
-------
u
UJ
a:
>
UJ
eg
2r &
.c «
4-* C-
II
*— (M CO «— eOCO^^Qmin-JOO; Qj
5«So>wc8o>o>oj-ojoj«o co
OOOOOOOOOOOO O
8
cJcsdcSooeaoeooo
oooooooooooo o
fc
C3C3CJOOOOOOOOO O
^* *
iii
o
-------
year. In the southern region, only two of the six plants (1977 and 2430) had
greater than 95 percent removal efficiencies throughout the year; in the
middle latitudes, one.out of the three plants (970) had greater than 95
percent removal efficiency. This analysis shows that removal efficiency was
affected primarily by nonclimate-related factors.
A similar analysis was performed using the data base derived from plants
that responded to the OCPSF 308 Questionnaire on summer/winter operations.
Question C-12 of the questionnaire 'asked each respondent to select a 3-month
period in the summer and a 3-month period in the winter of the same year. The-
summer period was generally selected as-June-July-August or July-August-
September, although a few respondents selected May-June-July. The winter
period was generally selected as January-February-March, although some
respondents selected various other 3-month periods from October through
February. For these two periods, the respondent was to provide summary data
for a variety of parameters, including average daily,total BOD5 influent and
effluent concentrations, TSS influent and effluent concentrations, MLSS con-
centration, mixed liquor volatile suspended solids (MLVSS) concentration, and
food to microorganism ratio (F/M). Plants were included in the analysis if
there were both influent and effluent total BOD5 values so that a BOD5 removal
efficiency could be calculated. Of all plants for which information was
available from Question C-12, 131 had sufficient information to enable the
calculation of BOD5 removal efficiency. When estimated values were given,
they were used. For the four plants using recycled waste streams (296, 2551,
1617, 2430), only the initial influent and final effluent values were used;
although this might result in artificially high efficiencies, it represented
the only logical approach. Two plants (1038 and 1389) had two different sets
of values, so each set was used. Two plants (227 and 909) had influent data
from one biological treatment system and 'effluent data from another, and were
not used in the analysis.
VII-88
-------
The results of the analysis are as follows:
Summer
Winter
Plant Category
All Plants
Southern Plants
Northern Plants
Middle Latitude
Plants
N !
131
52
46
33
Avg
Sfficiencj
0.89
0.91
0.86
0.89
Std Avg Std
r Dev Efficiency Dev
0.31
0.14
0.48
0.18
0.86
0.86
0.85
0.87
0.25
0.21
0.32
0. 19
Southern plants were located in Alabama, Florida, Georgia, Louisiana,
Mississippi, South Carolina, and Texas. Northern plants were located in
Connecticut, Iowa, Illinois, Indiana, Michigan, New Jersey, New York, Ohio,
Pennsylvania, Rhode Island, and West Virginia. Middle latitude plants were
located in Arkansas, Delaware, Kentucky,. Maryland, North Carolina, Oregon,
Tennessee, Virginia, and Washington.
These results are consistent with the results of the 14-plantdaily data
analysis discussed previously. The BOD5 removal efficiencies for all plants
are 3 percent less during the winter period than the .summer period (86% vs.
89%). The regional removal efficiencies are 1 to 5 percent less in the winter
period than in the summer period. The greatest regional variation in
efficiency occurs in the south. The standard deviation.of the efficiency is
large relative to the efficiency difference within each category, reflecting
the large variations among plants within the same category. These results
tend to indicate that while northern and middle latitude plants would have
larger swings in temperature going from season to season, these swings have
been compensated for through operation and process modifications as indicated
by the similar summer and winter removal efficiencies (86% vs. 85%). The
larger difference between summer and winter removal efficiencies for southern
plants (91% vs. 86%) indicate that these facilities have not "adequately
addressed the smaller temperature swings by operational and process modifica-
tions.
VII-89
-------
These findings support several conclusions. There may be differences
between efficiencies attainable in summer and in winter, but these differences
are nonetheless small. The large standard deviations obtained reflect differ-
ences in operating practices among plants. Plants that operate efficiently do
so year-round, and have been able to minimize or at least partially compensate
for temperature effects through equipment and operational treatment system
adjustments. In addition, plants located in the colder northern climate show
minimal efficiency differences between winter and summer months, which
provides further evidence that temperature effects are minimal. The daily
data assessment also indicates minimal efficiency variations during the spring
and autumn months, when temperature fluctuations would tend to be greatest;
this result casts doubt on the theory that fluctuations, rather than continued
cold, would reduce BOD5 removal efficiency by preventing the formation of a
stable microbial population.
Average Effluent BOD., and TSS
The effect of temperature on effluent BOD5 and TSS levels was evaluated
previously in the July 1985 document entitled "Selected Summary of Information
in Support of the OCPSF Point Source Category Notice of Availability of New
Information." EPA calculated rank correlation by subcategory for BOD5 efflu-
ent and TSS effluent versus heating degree days, a measure typically used by
power companies to estimate heating bills. The results of the analysis were
consistent with the assumption that temperature is not a factor. With the
exception of effluent TSS for specialty chemicals, all calculated rank
correlations were not significant. In the case of specialty chemicals, the
correlation was positive and significant. However, the positive correlation
implies that TSS increases as temperature decreases. Since engineering
considerations dictate that TSS should not decrease as temperature increases,
this result is considered spurious.
A new analysis was conducted, employing data from. 20 of the 21 plants in
the data base used for the calculation of BPT variability factors. The only
plant not used was #908, because of its location in Puerto Rico. BOD, and TSS
effluent averages were compared to months rather than heating degree days (see
Tables VII-26 and VII-27). The annual average BCD, and. TSS effluent concen-
trations are 22 mg/1 and 31 mg/1, respectively. Seven of the 20 plants (525,
VII-90
-------
1
I i ~. 8 S * S 3.3. B Sj « Xj S,. . 8|
C. i N> O3 JO CO CO CO w o* ^ ^ ^_ (^
ft) i fO CM CM *•• *^ *^ *^
i/^ i ?5l «~ ^o K» w *~ |^ jo JT ^ 2 ^ . i
ru • K) CO *Q M ^ ^"" "^ *^ S _ £
, ^ f\J »~ «~ C\J »" *~ <™ t''
io.O-OO-iK'l'-ggJSgD! K
S j |i;8|H|8-§§8^ 3
«:8M8883S888SB3 8
felocDed^-oi^-*""-*"10" *
K | 8 58 IS B 3. « » S 8 3 S^ 58
«>, i' g a s 3 s t; § s § s " s &.
o ' in^-o^roo-jo^jj^®^;^- , ,$t
fell^r-ljj^jijio^jJrOwp y>
', w ° t\J >-
jn j S ft to 3 5j '5j 8 !» 3 t= •» ^ g
i fc 8 S 8 K § 52 !5 § ~ B S ^
a
VII-91
-------
<=>
fc i s
s
*o K» in h- r\i m
«- T •» k- in K
- : S K
D. *- * in o* *- «—
«- fM CM f^ ^
2 fc Rj
S
CM ,- ^ ?: g Ri
g
S
.to m in N- to
8:
IA O T- r-
iK^^-niKiRincOroRRj?
IN. i »~ino*OeOMN-cMS.o*O*o
eo * rnF-iA^oroNOolKJoiog
if r
•— g> Q. •"
V U
O) O
jD O)
VTI-92
-------
682, 970, 1062, 1973, 2181, and 2430) have monthly average BOD5 effluent
concentrations less than 22 rag/1 throughout the year, while four of the
20 plants (387, 1012, 1407, and 2626) have monthly average. BOD5 effluent
concentrations less than 37 mg/1 throughout the year. Also, if winter months
are defined as January-February-March and summer months are defined as
June-July-August, three plants (1062, 1149, and 2430) have lower average BOD5
effluent concentrations for the winter months than for the summer months. In
addition, two plants (387 and 2626) have average BOD5 effluent concentrations
for the winter within 3 mg/1 of the summer average BODg effluent concentra-
tions, while four plants (444, 1973, 2626, and 2695) have average TSS effluent
concentrations for the winter months within 3 mg/1 of the summer average TSS
effluent concentrations.
Another analysis was performed comparing each plant's average BODg and
TSS effluent concentrations in the winter and summer months to its annual
average BOD5 and TSS effluent targets that provide the basis for BPT effluent
limitations. These annual compliance targets are presented in Appendix VII-A
of this document. Eight of the 20 plants (525, 682, 1062, 1407, 1647, 1973,
2181, and 2430) had both winter and summer average BOD5 effluent concentra-
tions below their annual average; BOD5 effluent compliance targets, while eight
plants (387, 444, 525, 1CTL2, 1407, 1973, 2181, and 2626) had both summer and
winter average TSS effluent concentrations below their annual average TSS
effluent compliance targets.
The plants were then divided into geographical regions,and the same
analyses performed. Subset I consisted of six northern plants from West
Virginia, Illinois, Rhode Island, Iowa, and Indiana; subset II consisted of
10 southern plants from Texas, Georgia, Louisiana, and South Carolina; and
subset III consisted of four middle latitude plants from Virginia and North
Carolina (see Tables VII-28, VII-29, VII-30, VII-31, VII-32, and VII-33). The
annual average BOD5 effluent concentrations were 13 mg/1, 30 mg/1, and 16 mg/1
for the northern, southern, and middle latitude plants, respectively; annual
average TSS effluent concentrations were 38 mg/1, 31 mg/1, and 19 mg/1 for the
northern, southern, and middle latitude plants, respectively. Approximately
66 percent, 70 percent, and 50 percent of the plants in the northern,
southern, and middle latitude regions, respectively, have annual average BOD5
VII-93
-------
g ~
8
II
*OintnoocMOOoco«-o
8
o co in o «-
t^ in CM •* •*
«- i- IM «- «-
tn r*- >o O< 5
CM «- «-
»- t^ o s. «-
O» CD O *O «—
in in N- M CM
«- CM >» «—
«- «- >O
^O co in
fc fe g f£
CM vO CO CO
If-
£. £
0) >- 3 -4-> O 5 5
>. C —' O> CL 0-> > o
raSDDmood)
z-3-3<«ozo
VII-94
-------
S
111 C
>- OJ
>
Ul
_l
en
— S, « — -, ~ .- — ^ -- - - -
^ RJn£cQe3!5«-»-«-»-«-K> O* **3" lf\ CO C? ^3 CO *O OJ
I ^ ooJCON-vON.srmxfMMM^
o_
5
S ------
ooeoco«*«o ^""cu^
trt ••«•»••«•••* ^J
O"»f*"Ooin*OmiAr*iw^^o^ co
«_ ^— . v" «~
SQS^^oj?^00^^^-*- «-
I
to
3JS— »
—
-
VII-95
-------
g
s
OC 'N
S-H
a
:•
§
111 o'
ro w- ro
«— %o o*
g
a.
?S
Ul
I
co
M
2
§
VII-96
-------
I
13 r*
ta z
at •—
-I
_ 0)
> 2
NO
8 § 5
3
CO
ON
10
-------
t>
5>
in «» >o o
ro ro CM e\i
N. co >o
O O> O N.
ro o to o.
S!
KJ
2: fc
S
§
-* »* >O IV.
-------
§
<9
£
m §
CD 2C
1— >
£ I
UJ -0
3 3
U- i^>
U. 4-*
UJ (D
5 «'
ae — '
ac '£
to «••
ro —
1
5 1
UJ I
_J U>
CO
ff
>- 0)
4-1 «-
O >
SE ^
c.
m
c
(0
-3
^o *4* ro ^o
O O O tn
CM CM CM CM
CM CM rM *\l
Jci S !Q 3
O>, »- «- ps.
\o «- o in
co co >o in
NO tn ^O O*
5 5Q S fc
ro in ro «-
•o- in in m
tn o >o co
F-. tn -** o
.
O 10 Q- tO
u. Z < SE
^ •* -3t
Os *- v6
o sj- ro
o «- o
>o ro «-
O CM •*
§^— «—
t^ CO
v- *O CM
O f-. CM
O CO r^
in •* rw
ro CM i-
8 tS SJ
S Si »
(A
* D
III
3 5 § SS
^* ro N- co
O** CO J1*- •**
ro in so in
o ro ro o*
ro in o in
\O N^ r*- *O
sO in m NO
f** *— N» CO
in po, c\j «-
v~ i"~ •*d" (\l
fc NO NQ O
in st- co *-
^
S t_ n> of
4-* O S S
O_ 4-> >• O
G) O O 0)
w O a: Q
tj
c>
°
vt
s
iv!
ro
tn
. in
ro
in
S
™
Di
tn
<
4->
5
o.
VII-99
-------
concentrations less than the regional annual average BOD5 effluent concentra-
tion; approximately 66 percent, 66 percent, and 50 percent of the plants in
the northern, southern, and middle latitude regions, respectively, have annual
average TSS effluent concentrations below the regional annual average TSS
effluent concentrations.
Additional Parameters
Evaluating other parameters using the 21-plant daily data base was not
possible since BODg, TSS, and flow were the only parameters monitored. The
Question C-12 data base provides average summer and winter values for MLSS,
HLVSS, and F/M. For all plants used in the previous C-12 data analysis for
BOD5 efficiency and for which values for MLSS, MLVSS, or F/M were available
for both summer and winter periods, average values for MLSS, MLVSS, and F/M
were determined.
Several editing rules were used. If estimates were given, they were
used. For Plant #1340, two different biological treatment processes had the
same BOD5 values, but had two different sets of MLSS, MLVSS, and F/M values.
Both sets were used. For Plant #296, which recycled waste streams, the MLSS,
MLVSS, and F/M values for each recycled stream were used. For Plants #1389
and #1038, -where two sets of BOD5 values were used, two sets of MLSS, MLVSS,
and F/M values were also used.
Based on these rules, average MLSS, MLVSS, and F/M values are as follows:
HLSS (mg/1)
Summer
4634
Winter
4950
MLVSS (mg/1)
Summer
3003
Winter
3444
Summer
1.024
F/M
Winter
0.863
An attempt was made to correlate the summer and winter values for MLSS,
MLVSS, and F/M to the summer and winter values for BOD5 removal efficiency.
This exercise yielded no conclusive results; the analysis found some plants
with poor winter performance to have higher MLSS concentrations and lower F/M
ratios (which should help to compensate for lower temperatures), while other
poor winter performers had the opposite trend in operating conditions. There
also appeared to be no correlation between plant location (northern or
VII-100
-------
southern) and seasonal operating parameters. This exercise also found plants
in northern climates achieving high year-round performance with very little
variation in seasonal MLSS, MLVSS, and F/M values. Therefore, it seems that
good plant performance is a function of a combination of factors (including
system design, operating parameters, and operating procedures) whose separate
contributions cannot be readily determined based on the level of information
gathered in this segment of the Section 308 Questionnaire.
Operational Changes
Two plants (948 and 2394) were identified as having made operational or
process changes in an effort to improve efficiency and provide at least
partial compensation for temperature.
Plant #948, which has a warm process effluent, has instituted several
operational changes in winter months to improve the performance of its
biological treatment system. First, it turns off some of its cooling towers
to compensate for greater heat loss during winter months. The facility also
decreases the number of aerators by 5 percent since there is significant heat
loss during the aeration process. The MLSS level and sludge age are increased
by decreasing the sludge wastage rate. These measures increase the sludge's
capacity to oxidize and metabolically assimilate organic material. A disad-
vantage of the increased sludge age is that sludge settling characteristics
are adversely affected. The plant largely compensates for this by increasing
the pplyelectrolyte dosage to the influent to the clarifier in the winter.
A second facility, Plant #2394, has also instituted process modifications
to improve the performance of its activated sludge system in the winter. In
the summer,-the plant uses a preaeration basin followed by a single stage
activated sludge unit and secondary elarifiers. In the summer, sludge from
the elarifiers is recycled ,to the activated sludge unit. In the winter,
sludge from the elarifiers is recycled to the preaeration unit, thus con-
verting it.into a second biological unit. In summary, the installation of
additional piping to allow flexibility in the sludge recycle point allows the
plant to have a one-stage biological treatment system in the summer and a
two-stage system in the winter.
VII-101
-------
Data are not available for Plant #948 to correlate its operational
changes with removal efficiency. Monthly monitoring data are available for
Plant 12394, although the plant was excluded from the 21-plant data base for
calculating BPT variability factors because the treatment system was modified
during the period of record and the effluent data were collected after terti-
ary treatment. Monthly BOD5 influent and effluent levels (IBOD5 and EBOD ),
TSS effluent levels (ETSS), and removal efficiencies for Plant #2394 are
presented in Table VII-34 for the period December 1981 to March 1984. The
results are inconclusive. They show reduced efficiency during the months of
January and February. They also show an efficiency increase of 19 percent
between January 1982 and January 1983, and an increase of 13 percent between
February 1982 and February 1983. The efficiency for January 1984 then drops
by 7 percent from the preceding January, but the February 1984 efficiency of
95 percent is the same as the efficiency of the preceding February. The sharp
efficiency increase between winter 1982 and winter 1983 suggests the effec-
tiveness of the operational changes, but the reasons for the decrease between
January 1983 and January 1984 cannot be determined from the available data.
It is not known if production changes occurred during that period.
Conclusion
Results of the BOD5 removal efficiency, BOD5 effluent, and operational
changes analyses performed above show a slight reduction in efficiency at some
plants during the months of January"and February. Efficiencies vary widely
among plants, and many plants have attained efficiencies of 95 percent or
greater for all months of the year. This suggests that the plants with lower
efficiencies are affected as much by inefficient operating practices as by
winter temperature considerations. Adoption of certain practices used by
plants with higher winter efficiencies by these plants should result in
improved winter efficiency.
Technologies and operating techniques exist that, if properly applied,
can compensate for temperature. Plants operating in cold weather conditions
should recognize that excessive storage prior to treatment may reduce the
temperature of the biotreatment system. Cold weather operation may require
insulation of treatment units, covering of open tanks, and tracing of chemical
VII-102
-------
TABLE VII-34.
MONTHLY DATA FOR PLANT #2394
1981
1982
1983
1984
December
January
February
March
April
May
June
July
August
September
October
November
December
January
February
March
April
May
June
July
Augus t
September
October
November
December
January
February
March
Average
Influent
BOD
(mg/1)
396
311
475
484
468
364
416
350
608
427
570
530
521
377
457
420
387?
404
436
332
474
364
415
388
351
295
397
354
Average
Effluent
BOD
(mg/1)
59
76
84
38
1 9
5
5
2
2
3
9
9
14
20
21
13
8
'5
4
3
3
3 .
4
8
11
35
21
15
Average
Effluent
TSS
(mg/1)
26
20
20
22
24
14
19
13
8
7
8
10
15
15
14
14
22
17
17
13
8
10
13
21
15
24
25
26
BOD5
Removal
Efficiency
(%)
0.85
0.76
0.82
0.92
0.98
0.99
0.99
0.99
1.00
0.99
0.98
0.98
0.97
0.95
0.95
0.97
0.98
0.99
0.99 ' •• •'
0.99 ,
0.99
0.99
0.99
0.98
0.97
0.88
• 0.95
0.96
VII-103
-------
feed lines. Insulation may include installing tanks in the ground rather than
aboveground, using soil around the walls of aboveground units, or enclosing
treatment units. During colder periods, maintenance of higher MLSS concen-
trations and suitable, reduced F/M may be necessary. Plant-specific
techniques, such as those used at Plants #948 and #2394, should also be
applied.
Another case study, cited in vendor literature, discusses cold weather
modifications for a biological treatment system at a West Virginia polyester
resin manufacturer. During the winter, the plant uses its equalization basin
for biological contact stabilization before the wastewater enters the
biological aeration basin. The plant replaced some of its aerators with
mechanical aerators especially designed for cold weather operation and added
similar aerators to the equalization basin for winter use. The new aerators
designed specifically for winter conditions provide "aeration, mixing, and 02
transfer without the temperature loss of conventional aerators during cold
weather." The West Virginia facility now achieves "a 99 percent BOD removal,
with influent BOD at 2,500 mg/1 and effluent at 20 mg/l~even in the winter."
Part of the improvement in effluent quality was attributed to warmer basin
temperature (7-13).
Two other points should be made. First, temperature is only one of many
factors that impacts wastewater treatment performance. Waste load variations,
biomass acclimation, flow variations, waste treatability, and temperature of
the wastewater as well as adequacy of treatment system design and operation
must all be considered. The interaction among these factors makes it diffi-
cult to isolate any one factor separately. Temperature considerations must be
viewed as specific to a given site in the context of these factors, rather
than as specific to a given geographic area.
Secondly, EPA has taken the cost of improving winter efficiency into
account by using the minimum State temperature in the K-rate equation for
estimating costs for full-scale and second-stage biological systems and by
adding a cost factor for biological upgrades. The cost factor ranges from
1.0 to 2.0 and is also based upon a State's minimum average ambient
temperature. Both State minimum temperature and the biological upgrade cost
factor are discussed in more detail in Section VIII.
VII-104
-------
4. Polishing and Tertiary Treatment Technologies
Polishing technologies consist of polishing ponds, filtration, and
chemically assisted clarification (CAC). Tertiary treatment includes only
activated carbon treatment.
a. Polishing Ponds
Polishing ponds are bodies of wastewater, generally limited to 2 to
3 feet in depth, used for the removal of residual suspended solids by
sedimentation. They are usually used as a tertiary treatment step following
biological treatment. Depending on the nature of the pollutant to be removed
and the degree of removal required, the polishing treatment system can consist
of one unit operation or multiple unit operations in series.
, According to the Section 308 Questionnaire data base, 64 OCPSF plants
reported using polishing ponds as an end-of-pipe treatment. Originally, 18 of
these 64 plants were used to establish treatment performance limits for BPT
Option II. However, following the December 9, 1986, Federal Register Notice
of Availability, the Agency carefully reviewed the BPT data base identifying
plants that reported having polishing ponds, and evaluated the data that they
provided. The 18 plants used to calculate BPT Option II effluent limitations
met the preliminary BFT effluent criteria, which was 95 percent removal of
BOD5 across the treatment system or an effluent BOD5 concentration equal to or
less than 50 mg/1 and an effluent TSS concentration .equal to or less than
100 mg/1.
The Agency reviewed the information provided in response to the Section
308 Questionnaires and contacted permit writers in the Regions and/or States
in which the facilities were located. The results of this effort identified
16 of the 18 plants as not containing BPT Option II treatment systems. Only
two plants are actually using their ponds as a final polishing step to remove
suspended solids and BOD5 from the effluent produced by a biological system
operating at a BPT Option I level. A summary of the results of this evalu-
ation is given in Table VII-35. A' description of the 16 plants without the
BPT Option II technology follows. Seven of the 16 plants combine treated
wastewater from the biological treatment system with other wastewaters in a
VII-105
-------
TABLE VII-35.
MATRIX OF 18 PLANTS WITH POLISHING
PONDS USED AS BASIS FOR BPT OPTION II LIMITATIONS
Plant ID
157
267
284
384
500
811
866
948
990
1020
1061
1438
1695
1698
1717
2471
2528
4017
Pond
Serves as
Equalization
Basin
X
X
X
X
X
X
X
Pond Pond Pond Pond
Serves as Serves as Known to Serves as
Secondary Reaeration Have Algae a Final
Clarifier Basin Problem Polish
X
X
X
X
X
X
X
X
X
X
X
TOTAL
VII-106
-------
final pond. Since these ponds mix different wastewaters, they achieve some
dilution of treated process wastewater prior to discharge. Because the actual
removal of the pollutants through biodegradation or settling cannot be
demonstrated, these ponds cannot be characterized as polishing ponds. Another
plant uses a "polishing pond" as a reaeration basin to increase the level of
dissolved oxygen (DO) in its effluent and to prevent a depressed oxygen level
from occurring in the receiving stream. Finally, one plant is known to have
an algae problem associated with its pond operation during the summer months,
that indicates that this plant may not be meeting the BPT Option II criteria
during part of the year.
As for the remaining 30 plants that reported having polishing ponds that
were not used to form the basis for the BPT Option II limits, four plants that
reported effluent BOD5, TSS, and flow data did not meet the BPT Option II
criteria. Fifteen plants did not report any BOD5 or TSS data; seven of these
15 plants use their ponds as a secondary clarification step, and six plants
use their ponds as a final mixing step. The remaining 11 plants were not used
because three plants have BPT Option III treatment (filtration); one plant
recycles water back to its production processes from the pond; one plant is an
indirect discharger; two plants discharge from their polishing ponds into
subsequent treatment stages; and four plants do not use biological treatment.
Based on the above information, the Agency concluded that the use of polishing
ponds to provide additional removal ,of conventional pollutants (BODg and TSS)
beyond that achievable by well-designed and well-operated biological treatment
(Option I) is not successfully demonstrated in the OCPSF industry.
b. Filtration
Filtration is an established unit operation for achieving the removal -of
suspended solids from wastewaters. The removal is accomplished by the passage
of water through a physically restrictive medium (e.g., sand, coal, garnet, or
diatomaceous earth) with resulting entrapment of suspended particulate matter
by a complex process involving one or more removal mechanisms, such as
straining, sedimentation, interception, impaction, and adsorption. Continued
filtration reduces the porosity of the bed as particulate matter removed from
the wastewater accumulates on the surface of the grains of the media and in
VII-107
-------
the pore spaces between grains. This reduces the filtration rate and
increases the head loss across the filter bed. The solids must be removed by
"backwashing" when the head loss increases to a limiting value. Backwashing
involves forcing wash water through the filter bed in the reverse direction of
the original fluid flow so that the solids are dislodged from the granular
particles and are discharged in the spent wash water. When backwashing is
completed, the filter is returned to service.
Filtration is an established -wastewater treatment technology currently in
full-scale use for industrial waste treatment. Filtration has several
applications: 1) pretreatment to remove suspended solids prior to processes
such as activated carbon adsorption, steam stripping, ion exchange, and
chemical oxidation; 2) removal of residual biological floe from settled
treatment process effluents; 3) removal of residual chemically coagulated floe
from physical/chemical treatment process effluents; and 4) removal of oil from
oil separation and dissolved air flotation effluents.
According to the Section 308 Questionnaire data base, 41 OCPSF plants use
filtration as a polishing technology. EPA evaluated BPT Option III (bio-
logical treatment plus multimedia filtration) technology to determine if this
option could achieve, in'°a practicable manner, additional conventional pollu-
tant removal beyond that achievable by well-designed, well-operated biological
treatment with secondary clarification. Eleven plants in the BPT data base
use BPT Option III technology and meet the final BPT editing criteria. Thus,
this option would require EPA to regulate all seven subcategories based upon a
very small data set. As shown in Table VII-36, the median effluent TSS
concentration value for these plants is 32 mg/1. Even if three additional
plants are included in this data base because they use Option I treatment plus
either ponds or activated carbon followed by1filters, the resulting median TSS
value is 34 mg/1. These results, when compared to the performance of
clarification only following biological treatment (median value of 30 mg/1),
clearly show that the efficiency of filtration following good biological
treatment and clarification is not demonstrated for this industry. Moreover,
on the average, OCPSF plants with more than Option I treatment in EPA's data
base (biological treatment plus filtration) have not demonstrated significant
BOD5 removal beyond that achievable by Option I treatment alone. The median
VII-108
-------
TABLE VII-36.
OPTION III OCPSF PLANTS WITH BIOLOGICAL TREATMENT
PLUS FILTRATION TECHNOLOGY THAT PASS THE BPT EDITING CRITERIA
Plant ID
2551
1943
102
2536
883
2376
1343
2328
909
1148
844
Median value
Effluent TSS
(mg/1)
9
16
18
18
27
32
36
37
41
46
54
32
Effluent BOD
(mg/1)
11
22
7
3
20
27
8
19
21
37
5
19
VII-109
-------
concentration value for these plants is 19 mg/1 compared to a median
BOD,
value of 23 mg/1 BOD5 for the plants with Option I technology in place which
meet the 95 percent/40 mg/1 BOD5 editing criteria. Therefore, EPA does not
believe that the data support any firm estimate of incremental pollutant
removal benefits and incremental costs for BPT Option III.
One commenter suggested that, in light of the apparent poor incremental
performance of filters in the OCPSF industry, EPA should transfer data from
non-OCPSF filtration operations, specifically from domestic sewage treatment.
EPA has evaluated the additional removal achievable by multimedia filtration
on the effluent from the biological treatment of domestic sewage. Data found
in EPA's "Process Design Manual for Suspended Solids Removal" (EPA 625/1-
75-003, January 1975) indicates that multimedia filtration achieves a median
of 62 percent removal of TSS from biological treatment effluent TSS levels of
25 mg/1 or less.
The Agency also considered transferring multimedia filtration performance
data from the pharmaceutical manufacturing point source category for use in
the development of BPT Option III (biological treatment plus filtration)
limitations. Daily data across multimedia filtration systems at three
pharmaceutical plants demonstrated that effluent concentrations of TSS from
advanced biological treatment in that industry could be reduced by 50 percent
over a 15 to 100 mg/1 influent concentration range by multimedia filtration
(no removal of BOD5 across multimedia filtration was demonstrated). This
concentration range covers the range of performance of OCPSF plants that meet
the Agency's Option I 95 percent/40 mg/1 (BOD5) and 100 mg/1 (TSS) editing
criteria to define well-designed and well-operated biological treatment.
However, the OCPSF industry filtration data do not indicate any
substantial TSS or BOD5 removal beyond that achieved by Option I technology.
This indicates that differences in the biological solids in the OCPSF industry
may be responsible for the lack of filtration effectiveness. For example, if
the OCPSF biological floe (solids) were to break into smaller sized or
colloidal particles, they could pass through the filter substantially
untreated. While EPA cannot be certain.whether this occurs, the data indicate
VII-110
-------
that filters in this industry are not as effective in removing OCPSF waste-
water solids as they may be for domestic sewage or certain other industry
wastewater solids. EPA does not believe that the appropriateness of
transferring data from these other wastewaters to the OCPSF industry is
demonstrated.
c. Chemically Assisted Clarification (GAG)
Coagulants are added to clarifiers (chemically assisted clarifiers) to
enhance liquid-solid separation, permitting solids denser than water to settle
to the bottom and materials less dense than water (including oil and grease)
to flow to the surface. Settled solids form a sludge at the bottom of the
clarifier, which can be pumped out continuously or intermittently. Oil and
grease and other floating materials may be skimmed off the surface.
Chemically assisted clarification may be used alone or as part of a more
complex treatment process. It may also be used as:
• The first process applied to wastewater containing high levels of
settleable suspended solids.
• The second stage of most biological treatment processes to remove the
settleable materials, including microorganisms, from the wastewater;
the microorganisms can then be either recycled to the biological
reactor or discharged to the plant's sludge handling facilities.
9 The final stage of most chemical precipitation (coagulation/
flocculation) processes to remove the inorganic floes from the
wastewater.
As discussed in Section VIII, chemically assisted clarification was a
component of the model wastewater treatment technology for estimating the BPT
engineering costs of compliance. First, when biological treatment was in
place (with or without secondary clarification), an additional chemically
assisted clarification unit operation iwas costed if the reported TSS effluent
concentration was more than 3 mg/1 above the plant's long-term average
compliance target. Second, for plants that do not need biological treatment
to comply with their BOD5 compliance targets, chemically assisted clarifi-
cation unit operations were costed if the reported TSS effluent concentrations
were more than 3 mg/1 above the long-term average compliance target.
VII-111
-------
Although chemical addition was not frequently reported by plants in the
OCPSF industry, chemically assisted clarification is a proven technology for
the removal of BOD5 and TSS in a variety of industrial categories, partic-
ularly in the pulp and paper industry. Case studies of full-, pilot-, and
laboratory-scale chemically assisted clarification systems in the pulp and
paper industry as well as other industrial point source categories are
discussed in the following sections.
Full-Scale Systems
Several full-scale, chemically assisted clarification systems have been
constructed in the pulp, paper, and paperboard industry and in other indus-
trial point source categories. Data on the capability of full-scale systems
to remove conventional pollutants are presented below.
Recent experience with full-scale, alum-assisted clarification of
biologically treated kraft mill effluent suggests that final effluent levels
of 15 mg/1 each of BODg and TSS can be achieved. The desired alum dosage to
attain these levels can be expected to vary depending on the chemistry of the
wastewater to be treated. The optimum chemical dosage is dependent on pH.
Chemical clarification following activated sludge treatment is currently
being employed at a groundwood (chemi-mechanical) mill. According to data
provided by mill personnel, alum is added at a dosage .of about 150 mg/1 to
bring the pH to an optimum level of 6.1. Polyelectrolyte is also added at a
rate of 0.9 to 1.0 mg/1 to improve flocculation.
Neutralization using NaOH is practiced prior to final discharge to bring
the pH within acceptable discharge limits. The chemical/biological solids are
recycled through the activated sludge system with no observed adverse effects
on biological organisms. Average reported results for 12 months of sampling
data (as supplied by mill personnel) show a raw wastewater to final effluent
BOD5 reduction of 426 to 12 mg/1, and TSS reduction of 186 to 12 mg/1.
VII-112
-------
Treatment system performance at the mill was evaluated as part of a study
conducted for the EPA (7-14). Data obtained over 22 months show average final
effluent BOD5 and TSS concentrations of 13 and 11 mg/1, respectively. As part
of this study, four full-scale chemically assisted clarification systems in
other industries were evaluated. Alum coagulation at a canned soup and juice
plant reduced final effluent BOD5 concentrations from 20 to 11 mg/1, and TSS
levels from 65 to 22 mg/1. Twenty-five mg/1 of alum plus 0.5 mg/1 of poly-
electrolyte are added to the biologically treated wastewater to achieve these
final effluent levels. Treatment plant performance was evaluated at a winery
where biological treatment followed by chemically assisted clarification was
installed. Final effluent levels of 39.6 mg/1 BOD5 and 15.2 mg/1 TSS from a
raw wastewater of 2,368 mg/1 BOD5 and 4,069 mg/1 TSS were achieved. The
influent wastewater concentrations to the clarification process were not
reported. The chemical dosage was 10 to 15 mg/1 of polymer (7-14). A
detailed summary of the results of the study of full-scale systems is pre-
sented in Table VII-37 (7-14).
In October'1979, operation of a full-scale chemically assisted
clarification system treating effluent from an aerated stabilization basin at
a northeastern bleached kraft mill began. This plant was designed and
constructed after completion of extensive pilot-scale studies. The purpose of
the pilot plant was to demonstrate that proposed water quality limitations
could be met through the use of chemically assisted clarification.' After
demonstrating that it was possible to meet the proposed levels, studies were
conducted to optimize chemical dosages. The testing conducted showed that the
alum dosage could be reduced significantly by the addition of acid for pH
control, while still attaining substantial TSS removal. In the pilot-scale
study, it was shown that total alkalinity, a measure of a system's buffering
capacity, was a reliable indicator of wastewater variations and treatability.
Through this study, a direct relationship between total alkalinity and alum
demand was shown. High alkalinity (up to 500 mg/1) caused by the discharge of
black liquor or lime mud results in high alum demands. Therefore, a sub-
stantial portion of alum dosage can be used as an expensive and ineffective
means of reducing alkalinity (pH) to the effective pH point (5 to 6) for
optimum coagulation. The use of acid to assist in pH optimization can mean
substantial cost savings and reduction in the alum dosage rate required to
VII-113
-------
c
o
Is
tn C
12
1 $
2 ^
Sg
•s-s
ra
E
V)
12
jTi*
Sfi-
l1!^
11?
X
tivt
oj.
lllut
TSS
8
o"
-
*!,
ill
- <« 3
a-Bjgl
fill
il
Jli
! 3
3 a
Us ill
INtfl-
a 1' *
ll
<
lit
f
.Jin
Sllllf
"~rsC~
=|ilif=
Sa I Ss-33
•• E- '
ill
sf
if
si
sf i!
9 9
at
SI
^
ill
SKS
ill
if!
ill
tj—t
e a
5-3
li
Jjg
I'll i
hi—
1 1
5 •
ill
*lriir
SySsSmSKS
lil
If
s <
-{!'
I le s|
lls
;s' fli
lisii
n«
-------
effect coagulation. In one instance, use of concentrated sulfuric acid for pH
reduction decreased alum demand by 45 percent. Acid addition was also
effective in reducing alum dosage for wastewaters with low alkalinity
(approximately 175 mg/1) (7-15).
Table VII-38 summarizes effluent quality of the full-scale system since
startup; this system has been operated at an approximate alum dosage rate of
350 mg/1 without acid addition. Recent correspondence with a mill represen-
tative indicated that, with acid addition, this dosage rate could be reduced
to 150 mg/1 (7-16). However, this lower dosage rate has not been confirmed by
long-term operation.
Scott et al. (7-17) reported on a cellulose mill located on the shore of
Lake Baikal in the USSR. The mill currently produces 200,000 kkg (220,000
tons) of tire cord cellulose and 11,000 kkg (12,100 tons) of kraft pulp per
year. Average water usage is 1,000 kl/kkg (240 kgal/t). The mill has strong
and weak wastewater collection and treatment systems. The average BOD5 for
the weak wastewater system is 100 mg/1, while the strong wastewater BOD5 is
400 mg/1. Only 20 percent of the total wastewater flow is included in the
strong wastewater system. Each stream receives preliminary treatment con-
sisting of'neutralization of pH to 7.0, nutrient addition, and aerated
equalization. Effluent from equalization is discharged to separate aeration ,
and clarification basins. These basins provide biological treatment using a
conventional activated sludge operation. Aeration is followed by secondary
clarification. Suspended solids are settled, and 50 percent of the sludge is
returned to the aeration process. Waste sludge is discharged to lagoons. The
separate streams are combined after clarification and are treated for color
and'suspended solids removal in reactor clarifiers with 250 to 300 mg/1 of
alum and 1 to 2 mg/1 of polyacrylamide flocculant, a nonionic polymer. The
clarifiers have an overflow rate of approximately 20.4 m per day/m
(500 gpd/ft2).
Chemical clarification overflow is discharged to a sand filtration
system. The sand beds are 2.9 m (9.6 ft) deep with the media arranged in five
layers (7-18). The sand size varies from 1.3 mm (0.05 in) at-the top to 33 mm
(1.3 in) at the bottom. The filter is loaded at 0.11 m3 per minute/m
VII-115,
-------
TABLE VII-38.
FINAL EFFLUENT QUALITY OF A CHEMICALLY ASSISTED
CLARIFICATION SYSTEM TREATING BLEACHED KRAFT WASTEWATER
Date
Average
for Month
BOD. (mg/1)
Maximum Day
Average
for Month
TSS (mg/1)
Maximum Day
September 1979
October 1979
November 1979
December 1979
January 1980
February 1980
March 1980
April 1980
May 1980
11
8
9
21
8
7
13
9
11
21
12
18
83
16
14
46
16
22
87
40
28
21
28
31
44
32
38
254
92
47
56
36
68 -
113
96,
80
VII-116
-------
(2.7 gpm/ft2). Effluent from sand filtration flows to a settling basin and
then to an aeration basin; both basins are operated in series and provide a
7r-hour detention time.
The effluent quality attained is as follows:
Parameter ". Raw Waste
BOD5 (mg/1) 300
Suspended Solids (mg/1) 60
PH -
Final Effluent
2
• '• 5
6.8 - 7.0
Individual treatment units are not monitored for specific pollutant
parameters. '
Pilot- and Laboratory-Scale Systems ,
Several laboratory- and pilot-scale studies of the application of
chemically assisted clarification have been conducted. Available data on this
technology to remove conventional pollutants based on laboratory- and pilot-
scale studies are presented below.
As part of a study of various solids reduction techniques, Great Southern
Paper Co. supported a pilot-scale study of chemically assisted clarification
(7-19). Great Southern operates an integrated unbleached kraft mill,
Treatment consists of primary clarification and aerated stabilization followed
by a holding pond. The average suspended solids in the discharge from the
holding pond were 65 mg/1 for the period January 1, 1973, to December 31,
1974. In tests on this wastewater, 70 to 100 mg/1 of alum at a pH of 4.5
provided optimum dosages; the removals after 24 hours of settling ranged from
83 to 86 percent. Influent TSS of the sample tested was 78 mg/1. Effluent
TSS concentrations ranged from 11 to 13 mg/1;
- In a recent EPA-sponsored laboratory study, alum, ferric chloride, and
lime in combination with five polymers, were evaluated in further treatment of
biological effluents from four pulp and paper mills (7-20). Of the three
chemical coagulants, alum provided the most consistent flocculation at minimum
dosages, while lime was the least effective of the three. However, the study
VII-117
-------
provides the optimum chemical dosage for removal of TSS from biologically
treated effluents. These, inconclusive findings are the result of a number of
factors, including the lack of determination of optimum pH to effect removal
of TSS; the lack of consideration of higher chemical dosages when performing
laboratory tests even though data for some mills indicated that better removal
of TSS was possible with higher chemical dosage (a dosage of 240 mg/1 was the
maximum considered for alum and ferric chloride, while 200 mg/1 was the
maximum dosage used for lime); the testing of effluent from one mill where the
TSS concentration was 4 mg/1 prior to the addition of chemicals; and the elim-
ination of data based simply on a visual determination of proper flocculation
characteristics.
Laboratory data on alum dosage rates for chemically assisted
clarification have been submitted to the Agency in comments on the pulp,
paper, and paperboard contractor's draft report (7-21). Data submitted for
bleached and unbleached kraft pulp and paper wastewaters indicate that
significant removals of suspended solids occur at alum dosages in the range of
100 to 350 mg/1 (7-22, 7-23, 7-24). For wastewaters resulting from the
manufacture of dissolving 'sulfite pulp, effluent BOD5 and TSS data were
submitted for dosage rates of 250 mg/1; however, it was stated that dosages
required to achieve an effluent TSS concentration on the order of 15 mg/1
would be in the range of 250 to 500 mg/1 (7-25). During the pulp, paper, and
paperboard rulemaking, NCASI assembled jar test data for several process types
and submitted it to the Agency (7-26). Data for chemical pulping subcategories
indicated that alum dosages in the range of 50 to 700 mg/1 will effect
significant removals of TSS. The average dosage rate for all chemical pulping
wastewaters was 282 mg/1. Data submitted for the groundwood, deink, and
nonintegrated-fine papers subcategories indicate that dosages in the range of
100 to 200 mg/1 .will significantly reduce effluent TSS.
Data on the frequency of this technology are not available for the OCPSF
industry although data on the frequency of other similar technologies
(coagulation, flocculation, clarification, chemical precipitation) have been
previously presented. However, based upon the above information and upon the
general performance of clarifiers in treating TSS, EPA has concluded that
chemically assisted clarification can treat TSS in non-end-of-pipe biological
plants to meet the BPT TSS limits.
VII-118
-------
d. Activated Carbon Adsorption
Activated carbon adsorption is a physical separation process in which
organic and inorganic materials are removed from wastewater by sorption or the
attraction and accumulation of one substance on,the surface of another. There
are essentially three consecutive steps in the sorption of dissolved materials
in wastewater by activated carbon. The first step is the transport of the
solute through a surface film to the: exterior of the carbon. The second step
is the diffusion of solute within the pores of the activated carbon. The
third and final step is sorption of the solute on the interior surface bound-
ing the pore and capillary spaces of the activated carbon, ,While the primary
removal mechanism is adsorption, biological degradation and filtration also
may reduce the organics in the solution.
Activated carbon is considered to be a non-polar sorbent and tends to
.sorb the least.polar and least soluble organic compounds; it will sorb most,
but not all, organic compounds. As activated carbon adsorbs organics from
wastewater, the carbon pores eventually become saturated and the exhausted
carbon must be regenerated for reuse or replaced with fresh;carbon. The
adsorptive capacity of the carbon can be restored by chemical or thermal
regeneration. :_•'-••
s t f ;, * '. - '
There are two forms of activated carbon in common use—granular and
powdered. Granular carbon is generally preferred for.most wastewater applica-
tions because it can be readily regenerated. The two forms of carbon used and
different process configurations are described below.
Granular Activated Carbon. Granular carbon is about 0.1 to 1 mm in
diameter and is contacted with wastewater in columns or beds. The water to be
treated is either filtered .down (downflow) or forced up (upflow) through the
carbon column or bed. Additional design configurations of carbon contact
columns include gravity or pressure flow, fixed or moving beds, and single
(parallel) or multi-stage (series) arrangements. In a typical downflow
countercurrent operation, two columns are operated.in series with a common
spare column. When breakthrough occurs for the second column (i.e., the
concentration of a, target pollutant in the effluent is higher than the
VII-119
-------
desired concentration), the exhausted column is removed from service for
regeneration of the carbon. The partially exhausted second column becomes
the lead column, and the fresh spare column is added as a second column in the
series. When breakthrough is again reached, the cycle is repeated. The fixed
bed downflow operation, in addition to adsorption, provides filtration but
may require frequent backwashing. In an upflow configuration, the exhausted
carbon is removed at the bottom of the column, and virgin or regenerated
carbon is added at the top, thereby providing countercurrent contact in a
single vessel.
Powdered Activated Carbon. Powdered carbon is about 50 to 70 microns in
diameter and is usually mixed with the wastewater to be treated. This
"slurry" of carbon and wastewater is then agitated to allow proper contact.
Finally, the spent carbon carrying the adsorbed impurities is settled out or
filtered. In practice, a multi-stage, countercurrent process is commonly used
to make the most efficient use of the carbon's capacity.
Carbon adsorption systems have been demonstrated as practical and
economical for the reduction of dissolved organic and toxic pollutants from
industrial wastewaters. Activated carbon can be used to remove chemical
oxygen demand (COD), biochemical oxygen demand (BOD), and related parameters;
to remove toxic and refractory organics; to remove and recover certain
organics; and to remove selected inorganic chemicals from industrial waste-
water. Compounds that are readily removed by activated carbon include
aromatics, phenolics, chlorinated hydrocarbons, surfactants, organic dyes,
organic acids, higher molecular weight alcohols, and amines. Activated carbon
can also be used to remove selected inorganic chemicals, such as cyanide,
chromium, and mercury. A summary of classes of organic compounds adsorbed on
carbon are presented in Table VII-39, and a summary of carbon adsorption
capacities (the milligram of compound adsorbed per gram of carbon) is
presented for powdered carbon in Table VII-40.
The major benefits of carbon treatment involve its applicability to a
wide variety of organics and its high removal efficiencies. The system is
compact, and recovery of adsorbed materials is sometimes practical. The
limitations of the process include ineffective removal of low molecular weight
VII-120
-------
TABLE VII-39.
CLASSES OF ORGANIC COMPOUNDS ADSORBED ON CARBON
Organic Chemical Class
Examples of Chemical Class
Aromatic Hydrocarbons
Polynuclear Aromatics
Chlorinated Aromatics
Phenolics
Chlorinated Phenolics
High Molecular Weight Aliphatic
and Branch Chain Hydrocarbons*
Chlorinated Aliphatic Hydrocarbons
High Molecular Weight Aliphatic
Acids and Aromatic Acids*
High Molecular Weight Aliphatic
Amines and Aromatic Amines*
High Molecular Weight Ketones,
Esters, Ethers, and Alcohols*
Surfactants
Soluble Organic Dyes
benzene, toluene, xylene
naphthalene, anthracenes ,
biphenyls
chlorobenzene, polychlorinated
biphenyls, aldrin, endrin,
toxaphene, DDT
phenol, cre'sol, resorcenol, and
polyphenyls
trichlorophenol,
pentachlorophenol
gasoline, kerosene
1,1,1-trichloroethane,
trichloroethylene, carbon
tetrachloride, perchloroethylene
tar acids, benzoic acid
aniline, toluene diamine
hydroquinone, polyethylene
glycol
alkyl benzene sulfonates
methylene blue, Indigo carmine
*High Molecular Weight includes compounds in the range of 4 to 20 carbon atoms
VII-121
-------
TABLE VII-40.
SUMMARY OF CARBON ADSORPTION CAPACITIES
Compound
bis(2-Ethylhexyl)
phthalate
Butylbenzyl phthalate
Heptachlor
Heptachlor epoxide
Endosulfan sulfate
Endrin'
Fluoranthene
Aldrin
PCB-1232
beta-Endosulfan
Dieldrin
Hexachlorobenzene
Anthracene
4-Nitrobiphenyl
Fluorene
DDT
2-Acetylaminofluorene
alpha-BHC
Anethole*
3,3-Dichlorobenzidine
2-Chloronaph thalene
Phenylmercurie Acetate
Hexachlorobutadiene
gamtna-BHC (lindane)
p-Nonylphenol
4-Dimethylaminoazobenzene
Chlordane
PCB-1221
DDE
Acridine yellow*
Benzidine dihydrochloride
beta-BHC
N-Bu tylph thalate
N-Ni trosodiphenylamine
Adsorption8
Capacity (mg/g)
Compound
Adsorption*
Capacity (mg/g)
11,300
1,520
1,220
1,038
686
666
664
651
630
615
606
450
376
370
330
322
318
303
300
300
280
270
258
256
250
249
, 245
242
232
230
220
220
220
220
Phenanthrene• 215
Dimethylphenylcarbinol* 210
4-Aminobiphenyl 200
beta-Naphthol* 200
alpha-Endosulfan 194
Acenaphthene 190
4,4' Methylene-bis-
(2-chloroaniline) 190
Benzo(k)fluoranthene 181
Acridine orange 180
alpha-Naphthol IQQ
4,6-Dinitro-o-cresol 169
alpha-Naphthylamine 160
2,4-Dichlorophenol 157
1,2,4-Trichlorobenzene 157
2,4,6-Trichlorophenol 155
beta-Naphthylamine 150
Pentachlorophenol 150
2,4-Dinitrotoluene 146
2,6-Dinitrotoluene 145
4-Bromophenyl phenyl ether 144
p-Nitroaniline* 140
1,1-Diphenylhydrazine 135
Naphthalene 132
l-Chloro-2-nitrobenzene 130
1,2-Dichlorobenzene 129
p-Chlorometacresol 124
1,4-Dichlorobenzene 121
Benzothiazole* 120
Diphenylamine 120
Guanine* 120
Styrene 12Q
1,3-Dichlorobenzene 118
Acenaphthylene 115
4-Chlorophenyl phenyl ether 111
Diethyl phthalate no
VII-122
-------
TABLE Vli-40.
SUMMARY OF CARBON ADSORPTION CAPACITIES (Continued)
Compound
Adsorption3
Capacity (mg/g)
Compound
Adsorption3
Capacity (mg/g)
2-Nitrophenol
Dimethyl phthalate
Hexachloroethane
Chlorobenzene
p-Xylene
2,4-Dimethylphenol
4-Nitrophenol
Acetophenone
1,2,3,4-Tetrahydro-
naphthalene
Adenine*
Dibenzo(a,h)anthracene
Nitrobenzene
3,4-Benzofluoranthene
1,2-Dibromo-3-chloro-
propane
Ethylbenzene
2-Chlorophenol
Tetrachloroethene
o-Anisidine*
5 Bromouracil
Benzo(a)pyrene
2,4-Dini trophenol
Isophorone
Trichloroethene
Thymine*
Toluene
5-Chlorouracil*
N-Ni trosodi-n-propylamine
bis(2-Chloroisopropyl)
ether
Phenol
99
97
97
91
85
78
76
74
74
7.1
69
68
57
53
53
51
51
50
44
34
33
32
28
27
26
25
24
24
21
Bromoform 20
Carbon tetrachloride 11°
bis(2-Chloroethoxy)
methane H
Uracil* . . 11
Benzo(ghi)perylene 11
1,1,2,2-Tetrachloroethane li
1,2-Dichloropropene 8.2
Dichlorobromomethane 7.9
Cyclohexanone* 6.2
1,2-Dichloropropane 5.9
1,1,2-Trichloroethane 5,-8
Trichlorofluorome thane ->.6
5-Fluorouracil* 5.5
1,1-Dichloroethylene 4.9
Dibromochloromethane 4^8
2-Chloroethyl vinyl
ether 3.9
1,2-Dichloroethane 3.6
1,2-trans-Dichloroethene 3.1
Chloroform ' ; 2.6
1,1,1-Trichloroethane 2.5
1,1-Dichloroethane
Acrylonitrile
Methylene chloride
Acrolein
Cytosine*
Benzene
Ethylenediaminetetra-
acetic acid
Benzoic acid
Chloroethane
N-Dimethylnitrosamine
1.8
1.4
1.3
1.2
1.1
1.0
0.86
0.76
' 0.59
6.8 x 10-5
VII-123
-------
TABLE VII-40.
SUMMARY OF CARBON ADSORPTION CAPACITIES (Continued)
NOT ADSORBED
Acetone cyanohydrin
Butylamine
Cyclohexylamine
Ethanol
Hydroquinone
Triethanolamine
Adipic acid
Choline chloride
Diethylene glycol
Hexamethylenediamine
Morpholine
Ion
Na+
K+
Ca++
Mg++
^ "mineralized" distilled water containing the following
Cone, (mg/1)
92
12.6
100
25.3
Ion
P04
S04
Cl-
Alkalinity
Cone, (mg/1)
10
100
177
200
Source:
concentration of
Isotherms for Toxic Organics." MERL, April 1980.
VII-124
-------
or highly soluble organics, low tolerance for suspended solids in the waste-
water, and relatively high capital and operating costs. Preliminary treatment
to reduce suspended solids and to remove oil and grease will often improve the
effectiveness of the activated carbon system.
Treatability tests should be performed on specific waste streams to
determine actual performance of an activated carbon unit. The degree of
removal of different organic compounds varies depending on the nature of the
adsorbate, the pH of the solution, the temperature of the solution, and the
wastewater characteristics. If the wastewater contains more than one organic
compound, these compounds may mutually enhance adsorption, may act relatively
independently, or may interfere.with one another.
According to the Section 308 Questionnaire data base, 21 OCPSF plants
reported using carbon adsorption as a tertiary treatment technology. Table
VII-41 presents tertiary activated carbon performance data for an OCPSF plant
sampled during the EPA 12-Plant Study.
E. Total Treatment System Performance
1. Introduction
The last two sections presented descriptions and performance data for
those in-plant and end-of-pipe treatment technologies currently used or avail-
able for the reduction and removal of conventional, nonconventional, and
priority pollutants discharged by the OCPSF industry. The performance data
presented were primarily for those pollutants that the technologies were
primarily designed to remove. For example, BOD5 and TSS data were presented
for activated sludge; metals data were presented for chemical precipitation;
and volatile priority pollutant data were presented for steam stripping.
This section discusses the removal of pollutants from all treatment
technologies by presenting the performance of total treatment systems. The
treatment systems studied are those used to promulgate the BPT and BAT
effluent limitations. In addition,- the performances of those treatment
systems within the OCPSF industry that do not use biological treatment are
also presented.
VII-125
-------
TABLE VII-41.
END-OF-PIPE CARBON ADSORPTION PERFORMANCE
DATA FROM PLANT NO. 3033
Pollutant
Name
Average
Influent Concentration
to Activated Carbon
(ug/1)
Average
Effluent Concentration
from Activated Carbon
(ug/1)
Bis(2-chloroethyl)ether (18)
1,2-Dichloropropane (32)
2,4-Dimethylphenol (34)
Methylene Chloride (44)
Phenol (65)
Bis(2-ethylexyl)Phthalate (66)
13.64
10.46
13.92
12.21
11.42
14.31
10.00 (ND)
10.00 (ND)
10.00 (ND)
11.46
10.00 (ND)
13.00
VII-126
-------
2. BPT Treatment Systems
EPA has promulgated concentration-based BPT effluent limitations based on
selected biological end-of-pipe technologies that are designed primarily to
address the conventional pollutants BOD5 and TSS. These are supplemented by
those in-plant controls and technologies that are commonly used to assure the
proper and efficient operation of the end-of-pipe technologies, such as steam
stripping, activated carbon, chemical precipitation, cyanide destruction, and
in-plant biological treatment. Activated sludge and aerated lagoons are the
primary examples of such biological treatment.
The. performance of BPT treatment systems is represented by the long-term
BOD and TSS averages for each subcategory and the overall maximum monthly and
daily maximum variability factors presented in the limitations development
part of this section.
3. Nonbiological Treatment Systems
Approximately 84 plants rely exclusively upon end-of-pipe physical/
chemical treatment or did not report any in-place treatment at all. These
facilities must comply with the BPT effluent limitations guidelines based on
biological treatment system performance. Some of these plants generate low
levels of BOD5, thus finding physical/chemical treatment more effective in
reducing TSS loadings. Without nutrient addition, biological systems
generally cannot function unless influent BOD5 is high enough to sustain their
biota. Other plants have determined, based on an analysis of the types and
volumes of pollutants that they discharge, that physical/chemical treatment is
more economical, easier to operate, or otherwise more appropriate. Some of
these plants can control conventional pollutants effectively without using the
biological component of the BPT Option I technologies. However, other plants
seem to rely on dilution of process wastewater prior to discharge rather than
the appropriate Option I treatment. A listing of available BOD5 and TSS
effluent data and in-place controls reported by those plants with nonbiolog-
ical treatment systems is presented in Table VII-42. Forty-one of the
physical/chemical treatment only plants reported discharge BOD5 concentration
data, and 46 provided TSS concentration data. After adjusting the reported
wastewater concentration data for non-process wastewater dilution, 29 percent
VII-127
-------
TABLE VII-42.
TREATMENT TECHNOLOGIES FOR DIRECT NONBIOLOGICAL PLANTS*
Plant
ID
76
87
105
114
155
159
225
259
260
294
373
447
451
502
536
Effluent BOD.
(mg/1)
-
929
-
15
-
429
96
350
20
57
62
23,628
-
93
31
Effluent TSS
(rag/I) Type of Controls Reported
Neutralization
44 Equalization, neutralization, primary
clarification, carbon adsorption
Str-eam stripping, neutralization, primary
clarification
89 Filtration
282 Neutralization, API separation, dissolved
air flotation
Filtration, chemical precipitation, steam
stripping, equalization, coagulation,
neutralization, oil separation, primary
clarification, filtration, carbon adsorp-
tion, second stage of an indicated
treatment unit
s>
46 Steam stripping, distillation, equaliza-
tion, settling pond, neutralization,
screening, oil skimming
Filtration, coagulation, API separation,
surface impoundment
8 Cooling tower, API separation
119 Reuse for steam, coagulation, flocculation,
neutralization, oil separation, primary
clarification
155 Neutralization, oil separation, oil
skimming
22,898 Neutralization, filtration
Chemical precipitation, primary clarifi-
cation, flocculation
38 Water scrub, neutralization
1 Neutralization
VII-128
-------
TABLE VII-42.
TREATMENT TECHNOLOGIES FOR DIRECT NONBIOLOGICAL PLANTS
(Continued)
Plant Effluent BOD5
ID (mg/1)
Effluent TSS
(mg/1)
Type of Controls Reported
569
614
657
663
669
709
727
775
814
819
859
876
16
7
56
91
84
225
90
Steam stripping, primary clarification
Distillation, equalization, acidification/
aeration, neutralization, filtration,
equalization
17 Collection basin, neutralization, oil
separation
47 Equalization, flocculatioh, neutralization,
dissolved air flotation, mechanical skim-
ming, spray cooling, polishing pond
42 Filtration, steam stripping, neutraliza-
tion, oil skimming, dissolved air flota-
tion, air stripping
98 Settling pond, neutralization, API separ-
ation, filtration, carbon adsorption
108 Equalization, flocculation, chemical pre-
cipitation, grit removal, oil skimming,
clarification, air stripping,
neutralization, polishing pond
6 Chemical precipitation, neutralization,
primary clarification
Carbon adsorption, neutralization, oil
skimming, oil separation, API separation,
coagulation, flocculation
128 Chemical precipitation, equalization, neu-
tralization, oil separation, carbon adsorp-
tion
4,369 Equalization, neutralization, primary
clarification
76 Formaldehyde treatment, carbon absorption,
equalization, neutralization, primary
clarification
VII-129
-------
TABLE VII-42.
TREATMENT TECHNOLOGIES FOR DIRECT NONBIOLOGICAL PLANTS*
(Continued)
Plant Effluent BOD,.
ID (mg/1)
Effluent TSS
(rag/1)
Type of Controls Reported
877
913
938
1618
1688
1774
4
142
8
54
27
942
962
991
992
1249
1439
1532
1569
71
17
-
-
-
302
110
18
66
25
-
-
-
1,463
-
44
11
46
Dissolved air flotation
Chemical oxidation, steam stripping, equal-
ization, phase separation, neutralization
Steam stripping, equalization, floccula-
tion, hypochlorite addition, filtration,
neutralization, primary clarification,
settling pond
Steam stripping, neutralization, oil skim-
ming, primary clarification
Equalization, primary clarification
Solvent decantation
Distillation, equalization, neutralization
Equalization, neutralization
Settling, solvent extraction, equalization,
neutralization, steam stripping
Steam stripping, mercury treatment, neu-
tralization, carbon adsorption
Distillation, equalization, neutralization,
primary clarification, blending and air
stripping, filtration
Oil skimming
Steam stripping, equalization, floccula-
tion, neutralization, primary clarification
Equalization, flocculation, neutralization,
primary clarification, filtration
VII-130
-------
TABLE VII-42.
TREATMENT TECHNOLOGIES FOR DIRECT NONBIOLOGICAL PLANTS
(Continued)
Plant Effluent BOD.
ID
(mg/1)
Effluent TSS
(mg/1)
Type of Controls Reported
1776
1785
1794
1839
2030
2055
2062
2073
2090
2206
2268
2345
168
6
862
50
2400 5,640
2419
2527
100 Steam stripping, grit removal, oil skim-
ming, neutralization
- Chemical precipitation, chromium reduction,
steam stripping, ion exchange, carbon ad-
sorption, equalization, neutralization
Oil skimming, API separation
Steam stripping, gravity settling
Chemical precipitation, chromium reduction,
air stripping, neutralization, flocculation
Steam stripping, coagulation, flocculation,
recycle basin, clarification, polishing
pond
Chemical precipitation, steam stripping,
carbon adsorption, coagulation, floccula-
tion, neutralization, pH adjustment .
40 HOPE skimmer, polishing pond, pH adjustment
50 Distillation, equalization, neutralization,
grit removal
Oil skimming, oil separation
264 Equalization, sedimentation, neutraliza-
tion, filtration
29 Steam stripping, solvent extraction, floc-
culation, redox reactor, redox towers,
neutralization, polishing pond, noncontact
coolers
1,175 Solvent extraction, distillation
Equalization, neutralization, oil skimming,
dissolved air flotation
Oil skimming, aerobic spray field
VII-131
-------
TABLE VII-42.
TREATMENT TECHNOLOGIES FOR DIRECT NONBIOLOGICAL PLANTS*
(Continued)
Plant Effluent BOD,
ID
(mg/1)
2735
2767
2770
2771
2786
16
140
80
4010
Effluent TSS
(rag/1)
Type of Controls Reported
2531
639
145
Equalization,
flocculation,
neutralization.
primary clarification, carbon adsorption
2533
2590
-
16
31
13
Equalization,
Sulfur recovei
screening
:y, single stac
re flash .
2606
2647
2668
2680
-
47
939
48
-
51
5,866
26
21
31
17
13
55
176
equalization, stormwater impoundment, neu-
tralization, oil separation, filtration,
carbon adsorption
Neutralization
Filtration, distillation
Steam stripping, distillation
Decant sump, equalization, steam stripping,
neutralization, carbon adsorption
Pellet skimming, neutralization, oil
skimming, dissolved air flotation,
clarification
Neutralization
Distillation, equalization, neutralization,
oil skimming, primary clarification
Equalization, neutralization, primary
clarification
Filtration, chemical precipitation, air
stripping, steam stripping, equalization,
neutralization, oil skimming, oil
separation, API separation, dissolved air
flotation, polishing pond, (nutrient
addition prior to a septic tank for part of
the plant flow)
Depolymerization, distillation, pH adjust-
ment, neutralization, centrifugation
n ' 446'.601' 611' 664> 956, 1033, 1327, 1593, 1670, 1986,
, and 2660 report no in-place treatment technology.
VII-132
-------
of the physical/chemical treatment plants were determined to require no
further treatment to comply with the individual plant BPT Option I BOD5 long-
term average effluent compliance targets (discussed later in this section and
in Section VIII). For another 69 percent of the plants, the engineering costs
of compliance were based on activated sludge treatment systems because their
discharge BOD5 concentrations (after correction for non-process wastewater
dilution) ranged from 15 to 23,600 mg/1 above their individual plant BPT
Option I BOD5 long-term average effluent compliance targets. The remaining
2 percent of the plants were costed for contract hauling because their
wastewater flows were less than 500 gallons per day (gpd).
In the case of TSS, 38 percent of the 46 physical/chemical treatment only
plants that reported TSS data were determined to require no further treatment
to comply with the individual plant BPT Option I TSS long-term average efflu-
ent compliance targets. For 49 percent of the plants, the engineering, costs
of TSS compliance were associated with the activated sludge treatment-system
costed for BOD5.control. For another 7 percent of the plants, t>e engineering
costs of TSS compliance were based on chemically assisted clarification
treatment systems} for 4'percent of the plants, costs were based on copper
sulfate addition to polishing ponds; and for 2 percent, on contract hauling
because the wastewater flows were less than 500 gpd.
Currently, 14 plants do not report any in-place treatment at all; of
these/two plants reported BOD5 and TSS concentrations. One plant would
require no treatment and the other plant would require biological treatment to
comply with their respective BPT compliance targets.
The Agency did not establish alternative limitations for facilities that
do not utilize or install biological treatment systems to comply with the BPT
effluent limitations. Some industry commenters criticized the Agency for not
exempting,or establishing, alternative BOD5 limitations for stand-alone
"chlorosolvent" manufactures. They claim that "chlorosolvent" wastewaters
cannoV sustain a biomass and should not be subject to limitations based on
biological treatment, but did not provide supporting data. The Agency
identified only three stand-alone "chlorosolvent" facilities (plants 569, 913,
and 2062) using the commenters definition of "chlorosolvents" as chlorinated
VII-133
-------
Cl and C2 hydrocarbons. These three plants use only physical/chemical
controls to achieve their current discharge levels. However, of these three
plants, only plant 913 reported BODS data that provided a long-term average of
4 mg/1 BOD5. Since this is significantly below the plant's BPT long-term
effluent compliance target of 21 mg/1 BOD5, the Agency concluded that plant
913 would comply with the BOD5 effluent limitations without the use of
biological treatment. The only other identified stand-alone chlorinated
organics plant that did not use biological treatment was plant 1569, a manu-
facturer of chlorinated benzenes. This plant reported a long-term average
BOD5 discharge concentration of 18 mg/1, a level already below its BPT long-
term effluent compliance target of 27 mg/1 BOD5. The Agency also identified
three other manufacturers that produced "chlorosolvents" along with other
products (plants 1532, 2770, and 2786); they reported long-term average BOD5
discharge concentrations of 110, 140, and 80 mg/1, respectively—sufficient5
levels to sustain biota. In fact, the Agency identified 13 OCPSF plants that
utilize biological treatment systems with reported influent BOD5 concentration
less than 125 mg/1. The influent concentrations for seven of these plants
range from 60 to 80 mg/1 BOD5. Furthermore, another plant (725) sampled by
EPA has an activated sludge system that treats wastewater with a 37 mg/1 BOD5
average influent concentration. The product mix at this facility included
tetrachloroethylene and chlorinated paraffins.
•Gfr
The nonbiologicar wastewater treatment performance information for OCPSF
plants that reported influent and effluent BOD5 and/or TSS data is listed in
Table VII-43. As shown, the ranges of BOD5 and TSS percent removals are 27 to
98 percent and 0 to 91 percent, respectively. Some of these systems include
clarification treatment, but in combination with other physical/chemical
wastewater treatment unit operations.
In an effort to identify performance data for physical/chemical
clarification treatment systems treating BOD5 and TSS, the Agency was able to
obtain influent and effluent BOD5 and TSS data for clarification systems at
pulp, paper, and paperboard mills. Table VII-44 presents performance data for
clarification systems at 27 mills, and the data show that clarification
systems can obtain significant removals of both TSS and BOD5 as well as
reducing TSS levels in raw wastewaters to levels comparable to BPT Option I
VII-134
-------
', TABLE VII-43.
PERFORMANCE OF OCPSF NONBIOLOGICAL WASTEWATER TREATMENT SYSTEMS
Plant ID
Reported Reported
Pollutant Influent Efficient .•%,•;. In-Place
Parameter (mg/1) (mg/1) Removal Treatment*
657
669
938
BOD
TSS
BOD
TSS
BOD
TSS
22
47
2804
451
226
1688
1776
2055
BODC
TSS"
BODC
TSS"
BOD
TSS
235
100
237
16 27 Collection basin,
17 i - 64 neutralization, oil
separation
56 98 Filtration, steam
42 91 stripping, neutralization,
oil skimming, dissolved air-
flotation, air stripping
Steam stripping, equaliza-
27 88 tion, flocculation,
hypochlorite addition,
filtration, neutralization,
primary clarification,
settling pond
142 — Steam stripping, equaliza-
46 80 tion, flocculation,
neutralization, primary
clarification
Steam stripping, grit
100 0 removal, oil skimming,
neutralization , .
168 29 Steam stripping, coagula-
tion,, flocculation, recycle
basin, secondary clarifi-
cation, polishing pond
individual plants may treat all process wastevater or a portion of the
process wastewater by the reported treatment unit operations. Reported
influent data may not precede all listed unit operations.
VII-135
-------
8
E
-
^
wT!
Q Q C
cn co c
C; b; fc? S3 E?
O» O» CT* \7\ CM
rH T-l i-l rH <-H
in
r-l rH M CM
i u-i
' CO CO
CM CM
I CM sj-
en
VII-136
-------
long-term average levels in a wastewater matrix containing low BODg levels.
In addition, for these plants BOD5 effluent values are also comparable to BPT
Option I long-term average levels.
Based on the discussion and the performance data presented above, the
Agency concludes that:
• There are a limited number of OCPSF plants with either no treatment or
physical/chemical treatment in-place (which have BOD and TSS effluent
data) that are not in compliance with the BOD5 and TSS BPT long-term
average effluent compliance targets and have not had BPT compliance
costs estimated based on biological treatment.
• There are a limited number of OCPSF plants with either no treatment or
physical/chemical treatment in-place (which have BOD5 and TSS effluent
data) that are in compliance with BOD5 but not in compliance with TSS
BPT Option I long-term average effluent compliance targets.
• BPT Option I long-term averages for BOD5 and TSS, which are based on
the performance of biological treatment, can be attained by physical/
chemical treatment systems either in-place or used by the Agency to
estimate BPT compliance costs (i.e., chemically assisted clarifica-
tion).
Furthermore, compliance with BAT toxic pollutant effluent limitations
guidelines based on installation of physical/chemical or biological treatment
or improvements in the design and operation of in-place treatment would also
result in incidental reductions of conventional pollutants.
For these reasons, the Agency has decided not to establish a separate set
of BPT effluent limitations for OCPSF plants that do not require biological
treatment to comply with BPT.
A. BAT Treatment Systems
The Agency promulgated BAT limitations for two subcategories that were
largely determined by raw waste characteristics. First, the end-of-pipe
biological treatment subcategory includes plants that have or will install
biological treatment to comply with BPT limits. Second, the non-end-of-pipe
biological treatment subcategory includes plants that either generate such low
levels of BOD5 that they do not need biological treatment or choose to use
VII-137
-------
physical/chemical treatment alone to comply with the BPT limitations for BOD5.
The BAT limitations are based on the performance of the biological treatment
component plus in-plant control technologies that remove priority pollutants
prior to discharge to the end-of-pipe treatment system. These in-plant
technologies include steam stripping to remove volatile and semivolatile
priority pollutants, activated carbon for various base/neutral priority
pollutants, chemical precipitation for metals, cyanide destruction for
cyanide, and in-plant biological treatment for removal of polynuclear aromatic
(PNA) and other biodegradable priority pollutants. Table VII-45 presents a
list of the regulated BAT toxic pollutants and the technology basis for the
final BAT Subcategory One and Two effluent limitations for each. Tables
VII-46 and VII-47 present a summary of the long-term weighted average effluent
concentrations for the final BAT toxic pollutant data base for BAT Subcategory
One and Subcategory Two. The minimum, maximum, and median of the plant's
weighted average effluent concentrations were calculated for each pollutant to
display the performance of well-operated treatment systems in the OCPSF
industry.
F. WASTEWATER DISPOSAL
1. Introduction
The method of treatment for direct and indirect dischargers was discussed
in Sections C and D. In this section the treatment processes and disposal
methods associated with zero or alternate discharge in the OCPSF industry are
described. Zero or alternate discharge at the OCPSF plant is defined as no
discharge of contaminated process wastewater to either surface water bodies or
to POTWs. Table VII-48 presents the frequency of waste stream final discharge
and disposal techniques. This section describes deep well injection (56 OCPSF
plants), contract hauling (128 plants), incineration (93 plants), evaporation
(29 plants), surface impoundment (25 plants), and land application (19 plants).
2. Deep Well Injection
Deep well injection is a process used for the ultimate disposal of
wastes. The wastes are disposed by injecting them into wells at depths of up
to 12,000 ft. The wastes must be placed in a geological formation that
prevents the migration of the wastes to the surface or to groundwater
VII-138
-------
TABLE VII-45.
LIST OF REGULATED TOXIC POLLUTANTS AND THE TECHNOLOGY BASIS
FOR BAT SUBCATEGORY ONE AND TWO EFFLUENT LIMITATIONS
Poll't.
No. Pollutant Name
BAT
Subcategory One
End-of-Pipe
Biological Treatment Plus
BAT
Subcategory Two
1 Acenaphthene
3 Acrylonitrile
4 Benzene
6 Carbon Tetrachloride
7 Chlorobenzene
8 1,2,4-Trichlorobenzene
9 Hexachlorobenzene
10 1,2-Dichloroethane
11 1,1,1-Trichloroethane
12 Hexachloroethane n
13 1,1-Dichloroethane
14 1,1,2-Trichloroethane
16 Chloroethane
23 Chloroform
24 2-Chlorophenol
25 1,2-Dichlbrobenzene
26 1,3-Dichlorobenzene
27 1,4-Dichlorobenzene
29 1,1-Dichloroethylene
30 1,2-Trans-Dichloroethylene
31 2,4-Dichlorophenol
32 5 1,2-Dichloropropane
33 1,3-Dichloropropene
34 2,4-Dimethylphenol
35 2,4-Dinitrotoluene
36 2,6-Dinitrotoluene
38 Ethylbenzene
In-,Flant Biological
In-Plant Biological
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping**
Steam Stripping
Steam Stripping
Steam Stripping
(Biological Only)
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
(Biological Only)
Steam Stripping
Steam Stripping
InrPlant Biological
(Biological Only)
(Biological Only)
Steam Stripping
In-Plant Biological
In-Plant Biological
Steam Stripping
Steam Stripping*
Steam Stripping*
Steam Stripping*
Steam Stripping
Steam Stripping*
Steam Stripping
Steam Stripping*
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
Reserved
Steam Stripping*
Steam Stripping*
Steam Stripping*
Steam Stripping
Steam Stripping
Reserved
Steam Stripping*
Steam Stripping*
In-Plant Biological
Reserved
Reserved
Steam Stripping*
VII-139
-------
TABLE VII-45.
LIST OF REGULATED TOXIC POLLUTANTS AND THE TECHNOLOGY BASIS
FOR BAT SUBCATEGORY'ONE AND TWO EFFLUENT LIMITATIONS
(Continued)
Poll't.
No. Pollutant Name
BAT
Subcategory One
End-of-Pipe
Biological Treatment Plus
BAT
Subcategory Two
39 Fluoranthene
42 Bis(2-Chloroisopropyl)Ether
44 Methylene Chloride
45 Methyl Chloride
52 Hexachlorobutadiene
55 Naphthalene
56 Nitrobenzene
57 2-Nitrophenol
58 4-Nitrophenol
59 2,4-Dinitrophenol
60 4,6-Dini tro-o-Cresol
65 Phenol
66 Bis(2-Ethylhexyl)Phthalate
68 Di-N-butyl Phthalate
70 Diethyl Phthalate
71 Dimethyl Phthalate
72 Benzo(a)Anthrancene
73 Benzo(a)Pyrene
74 3,4-Benzofluoranthene
75 Benzo(k)Fluoranthene
76 Chrysene
77 Acenaphthylene
78 Anthracene
80 Fluotene
81 Phenanthrene
In-Plant Biological
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
In-Plant Biological
Steam Stripping and
Activated Carbon
Activated Carbon
Activated Carbon
Activated Carbon
Activated Carbon**
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological'
Steam Stripping*
Steam Stripping
Steam Stripping
Steam Stripping*
In-Plant Biological
Steam Stripping and
Activated Carbon
Activated Carbon
Activated Carbon
Activated Carbon
Activated Carbon
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
In-Plant Biological
VII-140
-------
TABLE VII-45.
LIST OF REGULATED TOXIC POLLUTANTS AND THE TECHNOLOGY BASIS
FOR BAT SUBCATEGORY ONE AND TWO EFFLUENT LIMITATIONS
(Continued)
Poll't.
No. Pollutant Name
BAT
Subcategory One
End-of-Pipe
Biological Treatment Plus
BAT
Subcategory Two
84 Pyrene
85 Tetrachloroethylene
86 Toluene
87 Trichloroethylene
88 Vinyl Chloride
119 Total Chromium
120 Total Copper
121 Total Cyanide
122 Total Lead
124 Total Nickel
128 Total Zinc
In-Plant Biological
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
Hydroxide Precipi-
tation***
Hydroxide Precipi-
tation***
Alkaline Chlori-
nation***
Hydroxide Precipi-
tation***
Hydroxide Precipi-
tation***
Hydroxide Precipi-
tation***
In-Plant Biological
Steam Stripping
Steam Stripping
Steam Stripping
Steam Stripping
Hydroxide Precipi-
tation***
Hydroxide Precipi-
tation***
Alkaline Chlori-
nation***
Hydroxide Precipi-
tation***
Hydroxide Precipi-
tation***
Hydroxide Precipi-
tation***
*Steam stripping performance data transferred based on Henry's Law Constant
groupings.
**Transferred from Subcategory Two.
***Metals and cyanide limitations based on hydroxide precipitation and
alkaline chlorination, respectively, only apply at the process source.
VII-141
-------
TABLE VH-46.
SIMWRY OF THE LOWS-TERM WEIGHED AVERAGE EFFLUENT OONCEmMTICNS FOR THE
FINAL BAT TOXEC POLLUTANT DATA BASE FOR BAT SUBGATBGORY ONE
Pollutant
Umber
1
3
4
6
7
8
9
10
11
12
14
16
23
24
25
26
27
29
30
31
32
33
34
35
36
38
39
42
44
45
52
Pollutant Name
Acenaphthene
Acrylonitrile
Benzene
Carbon Tetrachloride
Chlorobenzene
1,2, 4-Trichlorobenzene
Hexachlorobenzene
1, 2-Dichloroethane
1, 1, 1-Trichloroethane
Hexachloroethane
1, 1,2-Tridiloroethane
Chloroethane
Chloroform
2-Chlorophenol
1, 2-Dichlorobenzene
1 , 3-Dichlorobenzene
1,4-Dichlorobenzene
1 , 1-Dichloroethylene
1 , 2-Trans-dichloroe'thylene
2,4-Wchlorophenol
1 , 2-Dichloropropropane
1 , 3-Dichloropropene
2,4-Dimethylphenol
2,4-Dinitrotoluene
2,6-Dinitrotoluene
Ethylbenzene
Fluoranthene
Bis(2-Chloroisopropyl)Ether
Methylene Chloride
Methyl Chloride
Hexachlorobutadiene
Number of
Plants
3
5
17
3
2
3
1
9
2
2
3
4
8
3
7
1
1
5
3
3
6
3
4
2
2
14
3
1
8
1
2
Median of
Est. Long-
Term Means
(ppb)
10.000
50.000
10.000
10.000
10.000
42.909
10.000
25.625
10.000
10.000
10.000
50.000
12.208
10.000
47.946
24.800
10.000
10.000
10.000
17.429
121.500
23.000
10.794
58.833
132.667
10.000
11.533
156.667
22.956
50.000
10.000
Minimum of
Est. Long-
Term Means
(ppb)
10.000
50.000
10.00
10; 00
10.00
10.00
10.00
10.00
10.00
10.00
10.00
50.00
10.00
10.00
10.00
24.80
10.00
10.00
10.00
10.00
13.19
10.25
10.00
10.00
10.00
10.00
10.13
156.67
10.00
50.00
10.00
Maximum of
Est. Long-
Term Means
(ppb)
13.00
122.67
16.62
10.00
10.00
69.46
10.00
1228.33
10.00
10.00
10.00
50.00
43.00
93.30
88.20
24.80
10.00
11.60
77.67
21.62
923.00
63.33
13.47
107.67
255.33
10.00
12.27
156.67
206.67
50.00
10.00
Vn-142
-------
TABLE VIP-46.
SUMMARY OF 1HE LOWS-TERM WEIGHTED AVERAGE EFFLUENT CONCHflEATIONS FCR THE
FINAL BAT TOXLC POLLUTANr DATA BASE FOR BAT SUBCATEGORY ONE
(Continued)
Pollutant
Number
55
56
57
58
59
65
66
68
70
71
72
73
74
75
76
77
78
80
81
84
85
86
87
88
Pollutant Name
Naphthalene
Nitrobenzene
2-Nitrophenol
4-Nitrophenol
2,4-Dinitrophenol
Phenol
Bis(2-Ethylhe5^1)Phthalate
Di-N-Butyl Phthalate
Diethyl Phthalate
Dimethyl Phthalate
Benzo(a)Anthracene
Benzo(a)Pyrene
3, 4-Benzofluoranthene
Benzo(K)Fluoranthene
Chrysene
Acenaphthylene
Anthracene
Fluorene
Phenanthrene
Pyrene
Tetrachloroethylene
Toluene
Trichloroethylene
Vinyl Chloride
Number of
Plants
10
4
2
3
3
22
2
2
2
2
2
1
1
1
3
3
3
3
6
3
3
24
4
3
Median of
Est. Long-
Term Means
(ppb)
10.000
14.000
27.525
50.000
50.000
10.363
47.133
17.606
42.500
10.000
10.000
10.333
10.267
10.000
10.000 .
10.000
10.000
10.000
10.000
11.333
10.423
10.000
10.000
50.000
Minimum of
Est. Long-
Term Means
(ppb)
10.00
14.00
20.00
50.00
50.00
10.00
43.45
13.09
23.67
10.00
10.00
10.33
10.27
10.00
10.00
10.00
10.00
10.00
10.00
10.33
10.00
10.00
10.00
50.00
Maximum of
Est. Long-
Term Means
(ppb)
10.21
149.67
35.05
145.00
105.35
120.00
50.81
22.12
61.33
10.00
10.00
10.33
10.27
10.00
10.00
13.00
10.00
10.00
17.92
16.00
227.00
102.67
16.00
174.00
VH-143
-------
TABLE VII-47.
SUflfiRY OF THE LONG-TERM WEIGHTED AVERAGE EFFLUENT COMmRATIONS FOR THE
FINAL BAT TOXIC POLLUTANT DATA. BASE FOR BAT SUBCATEQORY TWO
Pollutant
foiriber
1
3
4
6
7
8
9
10
11
12
13
14
16
23
25
26
27
29
SO
32
33
34
38
39
42
44
45
52
Pollutant Name
Acenaphthene
Acrylonitrile
Benzene
Carbon Tetrachloride
Chlorobenzene
1,2, 4-Tridilorobenzene
Hexachlorobenzene
1 , 2-Dichloroethane
1, 1, 1-Trichloroethane
Hexachloroe thane
1 , 1-Dichloroethane
1, 1, 2-Trichloroethane
Chloroe thane
Chloroform
1 , 2-Dichlorobenzene
1,3-Dichlorobenzene
l,4^chlorobenzene
1 , 1-Dichloroethylene
1, 2-Trans-dichloroethylene
1 , 2-Dichloropropane
1 , 3-^ti.chloropropene
2,4-Dimethylphenol
Ethylbenzene
Fluoranthene
Bis(2-Chloroisopropyl)Ether
Methylene Chloride
Methyl Chloride
Hexachlorobutadiene
Number of
Plants
1
1
4
_
_
_
_
2
1
_
1
2
2
2
_
_
_
2
2
_
_
1
_
1
_
3
1
:
Median of
Est. Long-
Term Means
(ppb)
10.000
50.000
28.576
64.500
64.500
64.722
64.722
64.722
10.000
64.722
10.000
10.293
50.000
44.108
64.722
64.500
64.500
10.052
11.052
64.722
64.722
10.000
64.500
11.533
64.722
10.800
50.000
64.500
Minimum of
Est. Long-
Term Means
(ppb)
10.000
50.000
10.00
64.50
64.50
64.72
64.72
62.77
10.00 '
64.72
10.00
10.00
50.00
11.81
64.72
64.50
64.50
10.00
10.00
64.72
64.72
10.00
64.50
11.53
64.72
10.00
50.00
64.50
Maximum of
Est. Long-
Term Means
(ppb)
10.00
50.00
200.33
64.50
64.50
64.72
64.72
66.67
10.00
64.72
10.00
10.59
50.00
76.41
64.72
64.50
64.50
10.10
12.10
64.72
64.72
10.00
64.50
11.53
64.72
30.33
50.00
64.50
VLT-144
-------
TABLE Vn-47.
SUMMARY OF THE LONG-TERM WEIGHTED AVERAGE EFFLUENT CONCENTRATIONS FOR THE
FINAL BAT TOXIC POLLUTANT DATA BASE FOR BAT SUBCATBQORY TWO
(Continued)
Pollutant
Number
55
56
57
58
59
60
65
66
68
70
71
72
73
74
75
76
77
78
80
81
84
85
86
87
88
Pollutant Name
Naphthalene
Nitrobenzene
2-Nitrophenol
4-Nitrophenol
2 , 4-Dini trophenol
4 , 6-Dini tro-0-Cresol
Phenol.
Bis(2-Ethylhe?yl)Phthalate
Di-N-Butyl Phthalate
Diethyl Phthalate
Dimethyl Phthalate
Benzo(a)Anthracene
Benzo(a)Pyrene
3 , 4-Benzofluoranthene
Benzo(k)Fluoranthene
Chrysene
Acenaphtnylene
Anthracene
Fluorene
Phenanthrene
Pyrene
Tetrachloroethylene
Toluene
Trichloroethylene
Vinyl Chloride
Number of
Plants
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
Median of
Est. Long-
Term Means
(ppb)
10.000
948.675
20.000
50.000
373.000
24.000
10.000.
43.455
13.091
23.667
10.000
10.000
10.333
10.267
10.000
10.000
10.000
10.000
10.030
10.000
10.333
18.429
12.418
11.586
64.500
Minimum of
Est. Long-
Term Means
(ppb)
10.00
712.60
20.00
50.00
373.00
24.00
10.00
43,45 ,
13.09
23.67
10.00
10.00
10.33
10.27
10.00
10.00
10.00
10.00
10.00
10.00
10.33
18.43
10.951
10.00
50.00
Maximum of
Est. Long-
Term Means
(ppb)
10.00
1184.75
20.00
50.00
373.00
24.00
10.00
43.45
13.09
23.67
10.00
10.00
10.33
10.27
10.00
10.00
10.00
10.00
10.00
10.00
10.33
18.43
13.88
13.17
79.00
VH-145
-------
TABLE VII-48.
FREQUENCY OF WASTE STREAM FINAL DISCHARGE
AMD DISPOSAL TECHNIQUES
Disposal Technique
No. of Plants No. of Plants Total No.
(Full Response) (Part A) of Plants
Direct Discharge to Surface Water
Discharge to Publicly
Owned Treatment Works
Discharge to Privately Owned
Off-Site Treatment Facilities
Deep Well Injection
Contract Hauling
Incineration
Land Application
Evaporation
Surface Impoundment
Recycle
250
287
6
32
82
63
0
13
8
36
54
106
35
24
46
30
19
16
17
0
304
393
41
56
128
93
19
29
25
36
NOTE: Combined direct and indirect discharges have been counted with the
direct dischargers; otherwise, remaining disposal techniques can be
double-counted for applicable plants.
. VII-146
-------
supplies. The most suitable site for deep well. injection is a porous zone of
relatively low to moderate pressure that is sealed above and-below by unbroken
impermeable strata. Limestones, sandstones, and dolomites are among the rock
types most frequently used because of their relatively high porosity. The
formation chosen must have sufficient volume to contain the waste without
resulting in an increase in the hydraulic pressure, which could lead to a
crack in the confining rock layers. .
The most significant hindrance to the application of deep well injection
is the potential for groundwater and surface water contamination; -Careful ..,-.
control of the process is necessary to prevent any contamination, and
injection should only be used in certain geographically acceptable areas. The
process is also limited to waste streams with low levels of suspended solids
to prevent plugging of the well screen which can cause unstable operation.
Pretreatment such as filtration can prevent .clogging of the screen and the
disposal aquifer. Another practical limitation is that waste streams to be
injected should have a pH value between 6.5 and 8.0 to prevent equipment
corrosion. In general, all streams subject to deep well injection are.treated
through equalization, neutralization, and filtration before disposal. Deep
well injection may be particularly attractive for disposal of inhibitory or
toxic organic waste streams.
According to the Section 308 Questionnaire data base, 56 OCPSF plants use
deep well injection as a means for ultimate disposal for all or a portion of
their wastes. . . --.. ;:.:.... •;.•...,,..-•. -..•..-.--••- :•,;..•. • -,._ . -r , ; - 4..-v
3. Off-Site Treatment/Contract Hauling
Off-site treatment refers to wastewater treatment at a site other than
the generation site. Off-site treatment may occur at a cooperative or
privately owned centralized facility. Often a contract hauler/disposer is
paid to pick up the wastes at the generation site and to haul them to the
treatment facility. The hauling may.be accomplished by truck, rail, or barge.
VII-147
-------
Off-site treatment/contract hauling is usually limited to low volume
wastes, many.of which may require specialized treatment technologies for
proper disposal. Generators of these wastes often find it more economical to
treat the wastes at off-site facilities than to install their own treatment
system. Sometimes, adjacent plants find it more feasible to install a
centralized facility to handle all wastes from their sites. The costs usually
are shared by the participants on a prorated basis.
According to the Section 308 Questionnaire data base, 128 plants use con-
tract hauling and off-site treatment as a final disposal technique for part or
all of their wastes.
4. Incineration
Incineration is a frequently used zero discharge method in the OCPSF
industry. The process involves the oxidation of solid, liquid, or gaseous
combustible wastes primarily to carbon dioxide, water, and ash. Depending
upon the heat value of the material being incinerated, incinerators may or may
not require auxiliary fuel. The gaseous combustion or composition products
may require scrubbing, particulate removal, or another treatment to capture
materials that cannot be discharged to the atmosphere. This treatment may
generate a waste stream tnat ultimately will require some degree of treatment.
Residue left after oxidation will also require some means of disposal.
Incineration is usually used for the ultimate disposal of flammable
liquids, tars, solids, and hazardous waste materials of low volume that are
not amenable to the usual end-of-pipe treatment technologies. To achieve
efficient destruction of the waste materials by incineration, accurate and
reliable information on the physical and chemical characteristics of the waste
must be acquired in order to determine appropriate operating conditions for
the process (e.g., feed rates, residence time, and temperature) and the
required destruction efficiency.
VII-148
-------
According to the Section 308 Questionnaire data base, 93 OCPSF plants use
incineration as an ultimate disposal technique.
5. Evaporation
Evaporation is a concentration process involving removal of water from a
solution by vaporization to produce a concentrated residual solution. The
energy source may be synthetic (s.team, hot gases, and. electricity) or natural
(solar and geothermal). Evaporation equipment can range from simple open
tanks or impoundments to sophisticated multi-effect evaporators capable of
handling large volumes of liquid. The evaporation process is designed on the
basis of the quantity of water to be evaporated, the quantity of heat required
to evaporate water from solution, and the heat transfer rate. The process
offers the possibility of total wastewater elimination with only the remaining
concentrated solution requiring disposal and also offers the possibility of
recovery and recycle of useful chemicals from wastewater.
According to the Section 308 Questionnaire date base, 29 OCPSF plants use
evaporation as a final disposal technique.
6. Surface Impoundment ,
Impoundment generally refers to wastewater storage in large ponds.
Alternate or zero discharge from these facilities relies on the natural losses
by evaporation, percolation into the ground, or a combination thereof.
Evaporation is generally feasible if precipitation, temperature, humidity, and
wind velocity combine to cause a net loss of liquid in the pond. Surface
impoundments are usually of shallow depth and large surface area to encourage
evaporation. If a net loss does not exist, recirculating sprays, heat, or
aeration can be used to enhance the evaporation rate to provide a net loss.
The rate of percolation of water into the ground is dependent on the subsoil
conditions of the area of pond construction. Since there is a great potential
for contamination of the shallow aquifer from percolation, impoundment ponds
are frequently lined or sealed to avoid percolation and thereby make the
basins into evaporation ponds. Solids that accumulate over a period of time
in these sealed ponds will eventually require removal. Land area requirements
are a major factor limiting the amount of wastewater disposed of by this
method.
VII-149
-------
According to the Section 308 Questionnaire data base, 25 OCPSF plants
report using surface impoundments as a final disposal technique.
7. Land Application
Land treatment is the direct application of wastewater onto land with
treatment being provided by natural processes (chemical, physical, and
biological) as the effluent moves through a vegetative cover or the soil.
Land application greatly reduces or eliminates BOD5 and suspended solids,
results in some nutrient removal, may result in some heavy metal removal, and
can recharge groundvater. A portion of the wastewater is lost to the atmo-
sphere through evapotranspiration, part to surface water by overland flow, and
the remainder percolates to the groundwater system.
Land disposal of industrial wastewaters must be compatible with land use
and take into consideration the potential for environmental pollution, damage
to crops, and entrance into the human food chain. To protect soil fertility
and the food chain during land disposal, it is necessary to determine the
capacity of soils to remove nitrogen, the potential toxicity of organic and
inorganic contaminants to plant life and soil, and the deleterious effects of
dissolved salts, including sodium, on plants and soil.
According to the Section 308 Questionnaire data base, 19 OCPSF plants
report using land application as a final disposal technique.
G. SLUDGE TREATMENT AND DISPOSAL
Solid residues (sludge) are generated by many wastewater treatment
processes discussed in previous sections of this chapter. Sludge is generated
primarily in biological treatment, chemical precipitation (coagulation/
flocculation), and chemically assisted clarifiers. Sludge must be treated to
reduce its volume and to render it inoffensive before it can be disposed.
Sludge treatment alternatives include thickening, stabilization, conditioning,
and dewatering. Disposal options include combustion and disposal to land.
The frequency of these treatment and disposal alternatives, according to the
Section 308 Quesionnaire data base, is presented in Table VII-49.
VII-150
-------
•a
=tt=
'o
<3
51-3
ll
•B
IS
3-0
si
=«=
g.
CM O O rH O O O •* ' e>».
O 00 O 00 0,0 °
Q rH
00 rH. O CM £ °,
O O O
-a- o
CS,
O cMrH O 00 esvo
I
£ Z
vn-151
-------
Sludge thickening is the first step in removing water from sludges to
reduce their volume. It is generally accomplished by physical means,
including gravity settling, flotation, and centrifugation. The principal
purposes of stabilization are to make the sludge less odorous and putrescible,
and to reduce the pathogenic organism content. The technologies available for
sludge stabilization include chlorine oxidation, lime stabilization, heat '
treatment, anaerobic digestion, and aerobic digestion. Conditioning involves
the biological, chemical, or physical treatment of a sludge to enhance
subsequent dewatering techniques. The two most common methods used to
condition sludge are thermal and chemical conditioning. Dewatering, the
removal of water from solids to achieve a volume reduction greater than that
achieved by thickening, is desirable to prepare sludge for disposal and to
reduce the sludge volume and mass to achieve lower transportation and disposal
costs. Some common dewatering methods include vacuum filtration, filter
press, belt filter, centrifuge, thermal, drying beds, and lagoons. Combustion
serves as a means for the ultimate disposal of organic constituents found in
sludge. Some common equipment and methods used to incinerate sludge include
fluidized bed reactors, multiple hearth furnaces, atomized spray combustion,
flash drying incineration, and wet air oxidation. Environmental impacts of
combustion may include discharges to the atmosphere (particles and other toxic
or noxious emissions), surface waters (scrubbing water), and land (ash).
Disposal of sludge to land may include the application of the sludge (usually
biological treatment sludge) on land as a soil conditioner and as a source of
fertilizer for plants, or the stockpiling of sludge in landfills or permanent
lagoons. In selecting a land disposal site, consideration must be given to
guard against pollution of groundwater or surface water supplies.
According to the Section 308 Questionnaire data base, 116 plants report
treating their sludge by thickening or dewatering (26 by thickening, 4 by
centrifugation, 4 by filtration, 22 by digestion, and 50 by dissolved air
flotation). Of the 104 plants reporting sludge disposal methods, 21 use
on-site landfills, 15 employ incineration, 18 use contract hauling, and 50
dispose of sludge at off-site landfills.
VII-152
-------
H. LIMITATIONS DEVELOPMENT
This section describes the methodology used to develop BPT, BAT, and PSES
effluent limitations and standards and includes discussions of data editing
criteria, derivation of long-term averages, and derivation of "Maximum for
Monthly Average" and "Maximum for Any One Day" variability factors.
1. BPT Effluent Limitations
As discussed in Section VI, the Agency decided to control BOD5 and TSS
under BPT. This section discusses the data editing rules and methodology used
to derive the final BPT effluent limitations guidelines for BOD5 and TSS.
a. Data Editing Criteria
Two sets of data editing rules were developed for BPT; one set was used
to edit the data base, which was utilized to calculate the long-term averages
(LTA) BOD5 and TSS values for each subcategory, while the second set was used
to edit the BPT daily data base, which was utilized to derive variability
factors.
b. LTA Data Editing
The two major forms of data editing performed on the LTA data base
obtained through the 1983 Section:308 Questionnaire were the dilution adjust-
ment assessments made for each full-response, direct discharge OCPSF facility
which submitted BOD,., or"TSS influent and/or effluent data and a BPT perform-
ance edit.
Dilution Adjustment - Since the limitations apply to all process
wastewater as defined in Section V, the Agency grouped all volumes of process
and non-process wastewater for the purpose of adjusting reported plant-level
BOD and TSS concentrations for dilution by nonprocess wastewater. This also
permitted the Agency to estimate engineering costs of compliance based on the
proper process wastewater flows and conventional pollutant concentrations.
For example, if BOD5 was reported as 28 mg/1 at the final effluent sampling
location with 1 MGD of process wastewater flow that was combined with 9 MGD of
unconlaminated nonprocess cooling water flow, then the BOD5 concentration in
VII-153
-------
the process wastewater alone was actually 280 mg/1 before dilution. This
conservatively assumes that the cooling water flow is free of BODg and TSS.
However, in the Agency's judgment, many of the sources and flows reported
as nonprocess wastewater by plants in their respective Section 308 Question-
naires are contaminated by process sources of BOD5 and TSS. Table VII-50
presents a list of the miscellaneous wastewaters reported in the Section 308
Questionnaires as nonprocess, which EPA has determined to be either contam-
inated (and therefore process wastewater) or uncontaminated with conventional
pollutants. The Agency reviewed this list after receiving public comments on
both NOAs criticizing some of its assignments and determined that, in general,
its assignments were correct.
Since the limitations apply to process wastevater (which includes
"contaminated nonprocess" wastewater) only, the relative contributions of
process wastewater versus "uncontaminated nonprocess" wastewater were deter-
mined at the influent and effluent sample sites. These data were used to
calculate plant-by-plant "dilution factors" for use in adjusting pollutant
concentrations at influent and effluent sampling locations as appropriate.
The general procedure for determining sample-site dilution factors and
adjusting BOD5 and TSS values was as follows:
• Sum uncontaminated nonprocess wastewater flows for an individual plant
(e.g., Plant No. 61 uncontaminated nonprocess wastewater flow =
0.280 MGD)
• Sum process wastewater flow for an individual plant (e.g., Plant No
61 process wastewater flow =0.02 MGD)
• Divide the sum of uncontaminated nonprocess wastewater flows by the
total process wastewater flow to determine dilution factor (e.g., for
Plant No. 61, 0.280 MGD/ 0.02 MGD = 14.0)
• Apply the sample-site dilution factor (plus 1) by multiplying by the
reported BOD or TSS value to be adjusted (e.g., for Plant No. 61,
196 mg/1 effluent BOD5 x (14.0 + 1) = 2,940 mg/1 effluent BOD .
VII-154
-------
TABLE VII-50.
CONTAMINATED AND UNCONTAMINATED MISCELLANEOUS "NONPROCESS" WASTEWATERS
REPORTED IN THE 1983 SECTION 308 QUESTIONNAIRE
Contaminated "Nonprocess" Wastewaters
(therefore designated as
process wastewater)
Uncontaminated Nonprocess Wastewaters
Air Pollution Control Wastewater (B5)
Sanitary (receiving biological treat-
ment) (B4)
Boiler Slowdown
Sanitary (indirect discharge)
Steam Condensate
Vacuum Pump Seal Water
Wastewater Stripper Discharge
Bi from Vertac
Boiler Feedwater Lime
Softener Slowdown
Contaminated Water Offsite
Condensate
Storage, Lans, Shops
Laboratory Waste
Steam Jet Condensate
Water Softener Backwashing
Miscellaneous Lab Wastewater
Raw Water Clarification
Landfill Leachate
Water Treatment
Technical Center
Scrubber Water
Utility Streams
Washdown N-P Equipment
Contact Cooling Water
Vacuum Steam Jet Slowdown
Densator Slowdown
Bottom Ash-Quench Water
! ,
Demineralizer Washwater
Non-Contact Cooling Water (Bl)
Sanitary (no biological treatment,
direct discharge) (B4)
Cooling Tower Slowdown (B2)
Stormwater Site Runoff (S3).
Deionized Water Regeneration
Miscellaneous Wastewater (conditional)
Softening Regeneration
Ion Exchange Regeneration
River Water intake
Make-up Water
Fire Water Make-up
Tank Dike Water
Demineralizer Regeneranf'
Dilution Water
Condensate Losses
Shipping Drains
Water Treatment Slowdown .
Cooling Tower .Overflow
Chilled Water Sump Overflow
Air Compressor and Conditioning Blow
Firewall Drainings
Other Non-contact Cooling
Miscellaneous Leaks and .Drains
Boiler House Softeners ' .
Fire Pond Overflow
Boiler Regeneration Backwash
Groundwater (Purge)
Firewater Discharge
Freeze Protection Water
VII-155
-------
TABLE VII-50.
CONTAMINATED AND UNCONTAMINATED MISCELLANEOUS "NONPROCESS" WASTEWATERS
REPORTED IN THE 1983 SECTION 308 QUESTIONNAIRE
(Continued)
Contaminated "Nonprocess" Wastewaters
(therefore designated as
process wastewater)
Uncontaminated Nonprocess tfastewaters
Water Softening Backwash
Lab Drains
Closed Loop Equipment Overflow
Filter Backwash
Demineralizer Wastewater
Laboratory Offices
Demineralizer Slowdown
Utility Clarifier Slowdown
Steam Generation
RO Rejection Water
Power House Slowdown
Inert Gas Gen. Slowdown
Contaminated Groundwater
Potable Water Treatment
Unit Washes
Non-Contact Floor Cleaning
Slop Water from Dist. Facilities
Laboratory and Vacuum Truck
Ion Bed Regeneration
Tankcar Washing (HCN)
Film Wastewater
Generator Slowdown
Air Sluice Water
Research and Development
Quality Control
Steam Desuperheating
Pilot Plant
Other Company Off-site Waste
Ion Exchange Resin Rinse
H2 and CO Generation
Demineralizer Spent Regenerants
Lime Softening of Process
Miscellaneous Service Water
Recirculating Cooling System
HVAC Slowdown Lab Utility
Condenser Water Backwash
Deonfler Regenerant
Raw Water Filter Backwash
Distribution
VII-156
-------
TABLE VII-50.
CONTAMINATED AND UNCONTAMINATED MISCELLANEOUS "NONPROCESS" WASTEWATERS
REPORTED IN THE 1983 SECTION 308 QUESTIONNAIRE
(Continued)
Contaminated "Nonprocess" Wastewaters
(therefore designated as
process wastewater)
Uncontaminated Nonprocess Wastewaters
Iron Filter Backwash
Area Washdown
Vacuum Pump Wastewater
Garment Laundry
Hydraulic Leaks
Grinder Lubricant
Utility Area Process
Contact Rainwater
Alum Water Treatment
Incinerator H20
Product Wash
Backflush from Demineralizer
Water Clarifier Blowdown
Water Treatment Filter Wash
Equipment Cooling H20
Belt Filter Wash
Ejector
OCPSF Flow from Another Plant
. VII-157
-------
Plant-specific dilution factor calculations and adjustments are
summarized in Appendix VII-B.
BPT Performance Edits - As stated earlier in Section VII, the Agency has
chosen BPT Option I (which is based on the performance of biological treatment
only) as the technology basis for the final BPT effluent limitations. After
selecting the technology basis, the Agency developed the associated limita-
tions based on the "average-of-the-best" plants that use the BPT Option I
technology. A performance criterion was developed to segregate the better
designed and operated plants from the inadequate performers. This was done to
ensure that the plant data relied upon to develop BPT limitations reflected
the average of the best existing performers. Since the data base also
included plants that are inadequate performers, it is necessary to develop
appropriate criteria for differentiating poor from good plant performance.
The BOD5 criteria used for the March 21, 1983 Proposal, the July 17, 1985 and
the December 8, 1986 Federal Register NOAs was to include in the data base any
plants with a biological treatment system that, on the average 1) discharged
50 mg/1 or less BOD5 after treatment, or 2) removed 95 percent or more of the
BOD5 that entered the end-of-pipe treatment system.
The Agency has received two diametrically opposed sets of comments on the
proposed data editing criteria used to develop BPT limitations. EPA proposed
to select plants for analysis in developing limitations only if the plants
achieve at least a 95 percent removal efficiency for BOD5 or a long-term
average effluent BOD5 concentration below 50 mg/1. On one hand, many industry
commenters argued that these criteria were too stringent, were based upon data
collected after 1977 from plants that had already achieved compliance with BPT
permits and thus raised the standard of performance above what it would have
been had the regulation been promulgated in a timely manner, and had the
effect of excluding from the BPT data base some well-designed, well-operated
plants. An environmental interest group argued, in contrast, that the
criteria were not stringent enough, in that they resulted in the inclusion of
the majority of plants in the data base used to develop effluent limitations.
VII-158
-------
The data collected by EPA for the BPT regulation were indeed, as industry
commenters have noted, based largely on post-1977 data. EPA had originally
collected data in the early and mid-1970s that reflected OCPSF pollutant
control practices at that time. As a result of industry challenges to EPA's
ensuing promulgation of BPT (and other) limitations for the OCPSF industry,
EPA began a new regulatory development program, which included a new series of
data-gathering efforts (see Section I of this document). Industry commenters
are correct in noting that the data are thus taken to a large extent from
OCPSF plants that had already been issued BPT permits that required compliance
by July 1977 with BPT limitations established by the permit writers on a
case-by-case basis. It is thus fair to conclude that the performance of at
least some of these plants was better when EPA collected the data for the new
rulemaking effort than it had been in the mid-1970s when the original BPT
regulations were promulgated.
EPA does not believe that the use of post-1977 data is improper. First,
the Clean Water Act provides for the periodic revision of BPT regulations when
appropriate. Thus it is within EPA's authority to write BPT regulations after
1977 and to base them on the best information available at the time. More-
over, it is not unfair to the industry. The final BPT regulations are based
on the same technology that was used to effectively control BOD5 and TSS in
the 1970s—biological treatment preceded by appropriate process controls and
in-plant treatment to ensure effective, consistent control in the biological
system, and followed by secondary clarification as necessary to ensure
adequate control of solids. The resulting effluent limitations are not neces-
sarily more (or less) stringent than they would have been if based on pre-1977
data. Many of the plants that satisfy the final data editing criteria
discussed below, and thus are included in the BPT data base, would not have
satisfied those criteria in the mid-1970s. The improved performance wrought
by the issuance of and compliance with BPT permits in the 1970's has resulted
in EPA's ability in 1987 to use,data from a larger number of plants to develop
the BPT limitations. Approximately 72 percent of the plants for which data
were obtained pass the final BOD5 editing criteria (95 percent/40 mg/1 for
biological only treatment); the editing criteria have excluded other plants
that, despite having BPT-type technology in-place, were determined not to meet
the performance criteria used to establish the data base for support of BPT
VII-159
-------
limitations. EPA concludes that the use of post-1977 data has resulted in a
good quality but not unrealistic BPT data base.
EPA has modified the BOD5 editing criteria to make them slightly more
stringent. However, it must be noted that EPA does not consider the selection
of editing criteria to be a strict numerical exercise based upon exclusion of
data greater than a median or any other such measure. EPA specifically
disagrees with the comment that data reflecting BPT performance must
necessarily constitute performance levels better than a median. The criteria
represent in numerical terms what is essentially an exercise of the Agency's
judgment, informed in part by industry data, as to the general range of
performance that should be attained by the range of diverse OCPSF plants
operating well-designed biological systems properly. The numerical analyses
discussed below should thus be regarded as an analytical tool that assisted
EPA in exercising its judgment.
The data to which the criteria have been applied reflect the performance
of plants that have been issued BPT permits requiring compliance with BPT
permit limits. It is not unreasonable to expect, therefore, that the class of
facilities identified as the "best" performers in the industry is considerably
larger than it would have been had the data been collected in the mid-1970s.
This result is consistent with the purpose and intent of the NPDES program:
to require those plants performing below the level of the best performers to
improve their performance. Moreover, it should be noted that while the major-
ity of OCPSF plants pass the initial screening criteria, a majority of OCPSF
plants (approximately 70 percent) will nonetheless need to upgrade their
treatment systems' performance to comply with the BPT effluent limitations
guidelines, based upon the reported effluent data (for 1980), and the long-
term average targets for BOD5 and TSS. The fact that a majority of plants
will need to upgrade years after they received their initial BPT permits
indicates that the result of the adoption of the data base used to develop the
limitations is appropriately judged the best practicable treatment.
The editing criteria were applied to the "308" survey data, composed of
annual average BOD5 and TSS data from plants in the OCPSF industry. The
purpose of the editing criteria was to establish a minimum level of treatment
VII-160
-------
performance acceptable for admission of a plant's data into the data base that
would be used to determine BPT limitations. First, only data from plants with
suitable treatment (i.e., biological treatment) were considered for inclusion
in the data base. For these plants, the use of both a percent removal
criterion and an average effluent concentration criterion for BOD5 is
appropriate, since well-operated treatment can achieve either substantial
removals and/or low effluent levels. In addition, use of only a percent
removal criterion would exclude data from plants that submitted usable data
but did not report influent data. The use of an effluent level criterion
allowed the use of data from such plants in estimating the regression
equation.
Following review of the data base, EPA continues to believe that
95 percent BOD5 removal is an appropriate editing criterion. Over half the
plants in the "308" survey data that reported both influent and effluent BOD5
achieve better than 95 percent removal. The median removal for these plants
is 95.8 percent, which reflects good removal from an engineering point of
view.
The Agency also continues to believe that a cut-off for.,average effluent
BOD concentration is necessary to establish an acceptable standard of
performance in addition to percent removal. In order to establish a cutoff
value for the final regulation and respond to various .comments, the Agency
re-examined the "308" survey data. There are data from a total of 99 full
response direct discharging plants with end-of-pipe biological treatment only
(the selected BPT technology, as discussed below) that reported average
effluent BOD5 and a full range of information regarding production at the
plant. All of these data were used in the evaluation of the BOD5 cutoff, even
in cases of plants that did not report influent values and for which removal
efficiencies could therefore not be estimated. The median BOD5 average
effluent for these 99 plants is 29 mg/1. There is no engineering or statis-
tical theory that would support the use of the median effluent concentration
as a cutoff for developing a regulatory data base. In fact, there are many
plants that, in the Agency's best judgment, achieve excellent treatment and
have average effluent values greater than the overall median of 29. There are
many reasonable explanations for differences in average effluent levels at
VII-161
-------
well operated plants. Differences in a plant's BPT permit limitations coupled
with individual company waste management practices and wastewater treatment
system design and operation practices, in addition to the type of products and
processes at each plant, contribute to differences in average effluent levels
achieved. To obtain insight into differences in BOD5 values among different
subcategories, the data were grouped into different subsets based on
subcategory production at each plant. The results of this analysis are
summarized in Parts A and B of Table VII-51.
The Agency grouped the data two different ways for analysis. Thus, the
data were assigned by plant into two different groupings, each with different
subgroups, and the medians of the average BOD5 effluent values in each sub-
group were determined. The first grouping placed plants into three subgroups
(plastics, organics, and mixed) and the second into five subgroups (fibers/
rayon, thermoplastics, thermosets, organics, and mixed). All plants
considered in the analysis had biological treatment only in place. The
assignment of a plant to a subgroup was determined by the predominant
production at the plant (i.e., whether a plant had 95% or more of its
production in the subgroup). For instance, if a plant has 95 percent or more
plastics production, it was placed in the plastics subgroup. Those plants not
containing 95 percent or more of a subgroup production were classified as
mixed.
The largest subset median average effluent BOD5 in both groupings is
42.5 mg/1, which suggests that the proposed 50 mg/1 criterion is high.
In the absence of a theoretical engineering or statistical solution that
would determine what value should be used in a regulatory context, the Agency
examined some reasonable alternatives suggested by the results displayed in
Parts A and B of Table VII-51. The Agency considered using different editing
criteria for different product subgroups, such as those listed in Part A of
Table VII-51, but decided to use a single criterion to define the final data
base.
VII-162
-------
TABLE VII-51.
SUMMARY STATISTICS FOR DETERMINATION
OF BPT BOD EDITING CRITERIA BY GROUPS
Subset
Number of
Plant Averages
Median of Plant
Average Effluent
BOD (mg/1)
A. Summary of Groups for Three Groupings
Plastics 30
Organics 42
Mixed (all remaining plants) 27
All Plants 99
20.5
42.5
35
29
B. Summary of Groups for the Five Groupings
Rayon/Fibers
Thermoplastics
Thermosets
Organics
Mixed (all remaining plants)
All plants ...-..:
7
17
3
42
30
99
14
18
32
42.5
35.5
29
VII-163
-------
An important reason for using a single edit criterion for all subcate-
gories is that this facilitates setting an edit criterion for the group of
plants that do not fall primarily into a single subcategory. These mixed
plants comprise a significant segment of the industry; thus, regulations must
be based on data from this segment as well. Editing criteria that are
subcategory-specific cannot be applied to mixed plants. The Agency did,
however, examine BOD5 levels by subgroups to gain insight into what uniform
editing criterion would be appropriate.
For the subgroups exhibiting relatively high BOD5 levels (organics and
mixed plants), EPA determined that a 40 mg/1 BOD5 edit would be appropriate.
This value is between the median for these two subgroups. Given the fact that
plants with substantial organics production tend to have fairly high influent
BOD5 levels or complex, sometimes difficult to biodegrade wastewaters, EPA
believes that a more stringent edit would not be appropriate for these two
groups. However, EPA believes that a less stringent edit would be inappro-
priate, since many plants in these subgroups meet the 40 mg/1 criterion.
The other subgroups have median values below 40 mg/1, and EPA examined
them closely to determine whether they should be subject to more stringent
edits than the organics and mixed subgroups. EPA concluded that they should
not for the reasons discussed below.
BOD,
The thermosets subgroup contains three plants, whose average effluent
levels are approximately 15, 32, and 34 mg/1, respectively. EPA believes
all three should be retained in the data base. This is particularly important
because a major source of wastewater at the plant with the lowest value is
only melamine resin production; several other types of resins fall under the
thermoset classification. Thus, including all three plants' data provides
improved-coverage of thermoset operations in the data base. An edit of
30 mg/1 arbitrarily excludes data from the two plants whose performance
slightly exceeds 30 mg/1 and would result in melamine resin production being
the predominant thermoset production represented in the data base.
VII-164
-------
The average BOD5 effluent values for rayon/fibers and thermoplastics are
lower than the average values for thermosets, organics, and mixed. The Agency
evaluated the effects of these subgroups by uniformly editing the industry
data base at 30, 35, 40, and 50 mg/1, using the BPT regression approach to
calculating subcategory long-term average values. The long-term averages
calculated for rayon/fibers and thermoplastics are relatively insensitive to
the use of the 30, 35, 40, and 50 mg/1 edited data bases. That is, the
long-term averages are roughly the same regardless of which of these edits is
used.
After considering the effect of the various editing criteria on the
different subgroups discussed above, EPA has concluded that a 95 percent/
40 mg/1 BOD editing criterion is most appropriate. Moreover, in defining
BPT-level performance, this criterion results in a data base that provides
adequate coverage of the industry.
As discussed previously, the Agency also saw a need to edit the data base
for TSS performance. Some commenters recommended additional editing for TSS,
and the Agency agrees that this is justified. The Agency is using two edits
for the TSS data. The primary edit is that the data must be from a plant that
meets the BOD5 edit (i.e., achieves either 95 percent removal of BOD5 or
40 mg/1). Second is an additional requirement that the average effluent TSS
must be 100 mg/1 or less. As a result of this edit, TSS data from 61 plants
are retained for analysis.
In a well-designed, well-operated biological treatment system, achievable
effluent TSS concentration levels are related to achievable effluent BOD5
levels and, in fact, often are approximately proportional to BOD,.. This is
reflected in the OCPSF data base for those plants that meet the BOD5 perfor-
mance editing criteria (provided that they also exhibit proper clarifier
performance, as discussed below). By using TSS data only from plants that
have good BOD5 treatment, the Agency is thus establishing an effective initial
edit for TSS removal by the biological system. However, as BOD5 is treated
through biological treatment, additional TSS may be generated in the form of
biological solids. Thus, some plants may need to add post-biological
secondary clarifiers to ensure that such biological solids are appropriately
treated.
VII-165
-------
Thus, while the 95/40 BOD5 editing ensures good BOD5 treatment and a
basic level of TSS removal, plants meeting this BOD5 editing level will not
• necessarily meet a TSS level suitable for inclusion in the data base used to
set TSS limitations. To ensure that the TSS data base for setting limitations
reflects proper control, EPA proposed in the December 8, 1986, Notice to
include only data reflecting a long-term average TSS concentration of less
than or equal to 100 mg/1.
The December 1986 Notice requested comment on the use of the 100 mg/1 TSS
editing criterion and, as an alternative, use of 55 mg/1 TSS concentration as
the editing criterion along with setting the TSS limitations based upon the
relationship between BOD5 and TSS. Some commenters criticized both the 100
mg/1 and 55 mg/1 as overly stringent, and asserted that such additional TSS
edits were unnecessary since the BOD5 edit was sufficient to assure that TSS
was adequately controlled. These commenters, while agreeing that there was a
relationship between BOD5 and TSS, also recommended a slightly different
methodological approach for analyzing the BOD5/TSS relationship.
The Agency disagrees with the commenters who argued in effect that all
TSS data from plants that meet the BOD5 criteria be included in the data base
for setting TSS limitations. The Agency has examined the data and has
concluded that an additional TSS edit is required at a level of 100 mg/1.
Support for this is evident in the reasonably consistent BODg and TSS
relationship for plants in the data set that results from the 95/40 BOD edit,
for TSS values of 100 mg/1 or less. For TSS values above 100 mg/1, there is a
marked change in the pattern of the BOD5/TSS relationship. Below 100 mg/1
TSS, the pattern in the BOD5/TSS data shown in Figure VII-2 is characterized
by a homoscedastic or reasonably constant dispersion pattern along the range
of the data. Above the 100 mg/1 TSS value, there is a marked spread in the
dispersion pattern of the TSS data. The Agency believes that this change in
dispersion (referred to as heteroscedastic) reflects insufficient control of
TSS in some of the treatment systems. The Agency has concluded that the
100 mg/1 TSS edit provides a reasonable measure of additional control of TSS
required in good biological treatment systems that have met the BODg edit
criterion.
VII-166
-------
?
o
o
tn
s
n
o
in
cxi
o
o
CXI
o
in
tu
CD H-
< Z
CC UJ
U.
fe
to
CO
O)
E
<
1
1
8
01
UJ
U.
U.
UJ
in
a
a
m
UJ
CO
UJ
H-
QJ £ e
^ •- B
UJ « *
O =
CO a
5 c-°
^ g co 00
2 2. •-' «•
3 H« O) O
CD ^ ^2 "^
E o 2
121
<^™ ^..»
(0 CD
"o °" ^
s ^ *
a. -A\
s
VII-167
-------
The Agency considered a more stringent TSS editing criterion of 60 mg/1,
rather than 100 mg/1. The Agency's analysis demonstrated that this is not
appropriate. Most fundamentally, this criterion would result in the exclusion
of plants that EPA believes are well-designed and well-operated plants.
Moreover, the relationship between BOD5 and TSS is well defined for plants
with TSS less than 100 mg/1 and BOD5 meeting the 952/40 mg/1 criteria.
The Agency gave serious consideration to the statistical method
recommended by a commenter for the analysis of the BOD../TSS relationship.
This commenter recommended a linear regression relationship between the
untransformed (not converted to logarithms) BOD5 and TSS data. The Agency has
retained the use of a linear regression relationship between the natural
logarithms of the BOD5 and TSS data. The logarithmic appproach is similar to
that recommended by the commenter, but resulted in a somewhat better fit to
the data.
In response to comments, the Agency also considered an editing criterion
based on secondary clarifier design criteria (i.e., clarifier overflow rates
and solids loadings rates). While the Agency agrees that using these design
criteria, if available, may have provided an appropriate editing criterion,
very little data were supplied by industry in response to the Agency's request
for data regarding these design criteria or were otherwise contained in the
record.
Daily Data Base Editing ;
Prior to the calculation of BPT variability factors, the BPT daily data
base was reviewed to determine if each plant's BOD5 and TSS data were
representative of the BPT technology performance.
The BPT daily data base contains daily data from 69 plants. The sources
of the data were the Supplemental Questionnaire, public comment data from
plants and the State of South Carolina, and data obtained during the EPA
12-Plant Study. The daily data, which included flow, BOD5, and TSS, were
entered on a computer data base. The sampling site for each parameter was
identified by a treatment code that was entered along with the data. The
VII-168
-------
treatment code allowed specific identification of the sampling site within the
treatment plant. For example, effluent data were identified as sampled after
the secondary clarifier, after a polishing pond, after tertiary filtration, at
final discharge, etc.
After the data base was established, the data at each sampling site
were compared with the treatment system diagrams obtained in the 1983 Section
308 Questionnaire. The comparison served to verify that the data corresponded
to the sampling sites indicated on the diagrams, and to determine if the data
were representative of the performance of OCPSF waste treatment systems. Non-
representative data were those data from effluent sampling sites where the
treatment plant effluent was diluted (>25 percent) with uncontaminated
non-process waste streams prior to sampling; treatment systems where a
significant portion of the wastewater treated by the treatment system
(>25 percent) was uncontaminated non-process or non-OCPSF wastewater;
treatment systems where side streams of wastewaters entered the treatment
system midway through the process, and no data were available for these waste
streams; and treatment systems where the influent sampling site did not
include all wastewaters entering the head of the treatment system (e.g., data
for a single process waste stream rather than all of the influent waste
streams).
Examination of the data available for each plant and the treatment system
diagrams provided the basis for exclusion of some of the plants from further
analysis. The criteria used were:
• Performance based on more than BPT Option I controls
• Data not representative of the performance of the plant's treatment
system
• Treatment systems not representative of the treatment technology
normally used in the OCPSF industry (e.g., effluent data did not
represent one wastewater treatment system, such as multiple
end-of-pipe treatment systems)
• Insufficient data due to -infrequent sampling (less than once per week
while operating) or omission of one or more parameters from testing
(BOD5, TSS, or flow)
• Treatment plant performance below that expected from the treatment
technology in operation (i.e., fail to meet the editing criteria of
95/40 for BOD. and 100 mg/1 for TSS).
VII-169
-------
Of the plants excluded from the data base, most were excluded for two or more
reasons. Other editing rules for plants retained in the data base included:
• Use of the most recent 12 months of all reported daily data when more
than 1 year of data was available. This allowed the Agency to use the
data from treatment systems with the most recent treatment system
improvements.
•, When historical reported long-term average and Section 308 Supplemen-
tal Questionnaire daily data were both available for a plant, the
Supplemental daily data were used to calculate the long-term average
because they provided a reproducible basis for calculating the
averages.
• When daily BOD5 or TSS values- were received or calculated
[concentration = C*(mass -f flow)] in decimal form, they were rounded
to the nearest milligram per liter.
Plots of concentration versus time and other analyses revealed that most
observations clustered around the mean with excursions far above or below the
mean. In the case of influent data, the excursions were believed to be
related to production factors such as processing unit startups and shutdowns,
accidental spills, etc. Effluent excursions, particularly those of several
days duration, were believed to be related to seasonal trends, upsets of the
treatment system, and production factors. Verification of the cause of the
excursions and of the apparent outliers in the data bases was deemed necessary
in order to supplement the analysis of the data with engineering judgment and
plant performance information. Each plant was contacted and asked to respond
to a series of questions regarding their treatment system, its performance,
and the data submitted. The plants were asked about seasonal effects on
treatment system performance and compensatory operational adjustments, winter
and summer NPDES permit limits, operation problems (slug loads, sludge
bulking, plant upsets, etc.), production changes and time of operation, plant
shutdowns, and flow metering locations. Data observations that were two
standard deviations above and below the mean were identified, and the plants
were asked to provide the cause of each excursion. The results of this effort
are described below.
VII-170
-------
The plant contacts and analysis of the data that were identified as being
more than two standard deviations above and below the mean revealed some of
the strengths and weaknesses of treatment in the industry. Plants within the
OCPSF.industry, regardless of products manufactured at an individual plant,
experience common treatment system problems. Daily data compiled over at
least a year show operational trends and problems, plant upsets, and seasonal
trends that would not be apparent for plants sampled less than daily.
Equalization and diversion basins are commonly used to reduce the effects of
slug loads on the treatment system and to prevent upsets. Influent data
obtained before equalization or diversion may show high strength wastes, but
the effluent may not because of equalization and diversion. Seasonal effects
tend to be more pronounced in southern climates because treatment systems
there generally may not be designed for cold weather. Operational techniques
to compensate for reduced efficiency are similar and should be practiced
industry-wide whenever needed or impossible with the existing treatment
system.
While common operational problems appear to be consistent across the
industry, responsive treatment system design and operation changes are not
fully documented within the data base. For example, some treatment systems
incorporating similar unit operations produced substantially different
effluent quality. The reasons for this may include strength and type of raw
wastes, capacity of the treatment system (under- or overloaded), knowledge and
skill of operating personnel, and design factors. While the raw waste type
can be categorized somewhat by dividing the OCPSF industry into subcategories,
the degree to which the other factors affect plant performance may not be
readily apparent in the data. For example, the daily data may not show
seasonal trends because of plant design or operational adjustments which
adequately compensate for cold weather.
Sampling and analytical techniques are another potential problem area of
the data base, particularly for the BOD5 data. The OCPSF industry manufac-
tures and uses a multitude of toxic substances that can affect a bioassay such
as the BOD5 test. Also, certain facilities sometimes collect unrefrigerated
.BOD composite samples which will affect the results of the analysis.
However, since the majority of the effluent data were collected for NPDES
VII-171
-------
permit compliance and approved analytical methodologies (such as standard
methods or EPA's test method) and QA/QC procedures are stipulated in each
facility's NPDES permit, it was assumed that the effluent data utilized were
collected and analyzed in an acceptable manner.
Table VII-52 presents a summary of the plants that were excluded from the
BPT daily data base and the reasons for the exclusion. Appendix VII-C
presents a plant-by-plant accounting of all 69 BPT daily data plants and
provides detailed explanations of each plant's inclusion or exclusion.
Based on the BPT daily data base editing, daily data from a total of
21 plants remain to calculate BOD5 variability factors and 20 plants remain to
calculate TSS variability factors (one plant does not meet the TSS editing
criterion). For these plants, all reported daily data from the most recent
12 months of sampling were included in the calculation of variability factors
because the Agency could not obtain sufficient information through plant
contacts and followup efforts to provide an adequate basis for deleting any
specific daily data points.
Derivation of Subcategory BOD., and TSS Long-Term Averages (LTAs)
As presented previously in Section IV, the Agency's final revised
subcategorization approach also included a methodology for calculation of BPT
BOD5 and TSS LTAs for each subcategory, which are used together with vari-
ability factors to derive facility subcategorical daily and monthly maximum
limitations. Recall from Section IV that the final subcategorization model is
given by:
= a
To estimate the average l^BOD.J corresponding to a set of the independent
variables wi;j, I4i, and Ib£, the random error term e.^ is deleted. The
estimates of the coefficients a, !.,, B, and D are used with the values of the
' independent variables to obtain the estimate.
VII-172
-------
0)
w
CO
CO
5
co
Q
i_
«^
12
£
jn
u.
co
Q
"3
"5
Exclusior
1
15
c
_o
cc
|2
T3 g ^ C i
lljii
ill! •' '
l||
CF 5 ... ....
0 w S
5 g £
I-
, s S | '
p« .
2 -
nO B«0 —
iljp
IliP
.1 1-2
O £ (/>•§*.
ssi
EZ-1 ... ....
isl
w.ES
5°-
- 2
||^S
j!]i : • •
iiss§iii^i§ii§gg|i9S^i§§5
E| ^^^^^.^^^ « CM
•
.
'
• •
si8ssiig8g8S§!8s§S|||p8g
. i
VII-173
-------
The LTA BOD5 for subcategory k is based on a plant that has 100 percent
of its OCPSF production in subcategory k. Therefore, to obtain the LTA BOD5
for subcategory k, set
If J=k
0,
Also, because the subcategorical LTA BOD5 is based on a plant that satisfies
the BOD5 95/40 criterion (set I41=l) and that has biological only treatment
(set Ibi-l), it follows that the BOD5 LTA for subcategory k is given by
BOD5 LTAk = exp [a + Tk + B + D],
where a, Tk, B, and D are estimates of the model parameters given in Appendix
IV-A, Exhibit 1. The estimates are derived from the data base of 157 full-
response, direct discharge OCPSF facilities that have at least biological
treatment in place, and that provided BOD5 effluent and subcategorical produc-
tion data. The parameter estimates are restated below and the subcategorical
LTAs for BOD5 are given in Table VII-53.
Parameter
a+Tl: Thermoplastics
a+T2: Thermosets
a+T3: Rayon
a+T4: Other Fibers
a+T5: Commodity Organics
a+T6: Bulk Organics
a+T7: Specialty Organics
B: Performance Shift
C: Treatment Shift
Estimate
4.27270510
5.22885710
4.32746980
4.03782486
4.49784137
4.66262711
4.92138427
-1.94453768
0.41834828
The subcategory LTAs for TSS are based on the final subcategorization
regression model for TSS, which was presented in Section IV as:
In (TSSi) = a + b [ln(BOD..)] + e. .
VII-174
-------
The estimates of the regression parameters a and b are derived from the
61 OCPSF plants that have at least biological treatment in place, meet the
95/40 editing criteria for BOD5, and have TSS effluent concentrations of at
most 100 mg/1. The estimates of parameters a and b are presented in Appendix
IV-A, Exhibit 2, and they are:
and
a = 1.84996248
b = 0.52810227.
Now, this model is used to provide subcategorical TSS LTAs corresponding to
the subcategorical BOD5 LTAs. Again, e.^ is set to zero in the model, and
TSS LTAR = exp (a + b lln(BOD5 LTAk)]
for k=l, 2, ..., 7. The calculated TSS LTA values are given in Table VII-54.
These subcategorical BOD,, and TSS LTAs allow the determination of
•* -.
plant-specific BOD5 and TSS LTAs, even for a plant that has production in more
than one subcategory. These plant-specific LTAs are then used with variability
factors to derive the effluent limitations guidelines presented in Section IX.
In particular, for a'^specific plant, let w.. be the proportion of that
plant's production in subcategory j. The plant-specific LTAs are given by:
Plant BOD,. LTA - £ w.(BOD LTA )
and
Plant TSS LTA =
w..(TSS LTA..),
where BOD LTA. and TSS LTA. are the BOD. and TSS long-term averages presented
5 ] 3 s
in Tables VII-53 and VII-54, respectively. This approach is analogous to the
building-block approach typically used by permit writers.
VII-175
-------
TABLE VII-53.
BPT SUBCATEGORY LONG-TERM AVERAGES (LTAs) FOR BOD
Subcategory
BOD LTA (mg/1)
Thermoplastics
Therraosets
Rayon
Other Fibers
Commodity Organics
Bulk Organics
Specialty Organics
16
41
16
12
20
23
30
TABLE VII-54.
BPT SUBCATEGORY LONG-TERM AVERAGES (LTAs) FOR TSS
Subcategory
TSS LTA (mg/1)
Thermoplastics
Thermosets
Rayon
Other Fibers
Commodity Organics
Bulk Organics
Specialty Organics
27
45
27
24
31
33
38
VII-176
-------
Calculation of BPT Variability Factors
After establishing a, final BPT daily data base, data from 21 plants for
BOD and 20 plants for TSS were retained to calculate variability factors
using the statistical methodology shown in Appendix VII-D. These statistical
methods assume a log-normal distribution; hypothesis tests investigating this
assumption are discussed in Appendix VII-E. The Agency has been using the
95th percentile average "Maximum for Monthly Average" and the 99th percentile
average "Maximum for Any One Day" variability factors for BOD5 and TSS,
regardless of the subcategory mix of each plant. However, many industry
commenters argued that effluent variability was subcategory-specific and
should be taken into account in variability factor calculations. In response
to these comments, the Agency performed an alternative variability factor
analysis which calculated production proportion-weighted variability factors
by category (plastics or organics) and subcategory for the 21 daily data
plants for BOD5 and the 20 plants for TSS. Table VII-55 presents the results
of this analysis which compares overall average variability factors with the
subcategory production proportion-weighted variability factors. This
comparison shows that subcategory-specific variability factors are not
substantially different from the overall average variability factors. This
would be expected since subcategory differences would be reflected more in the
long-term average values, while variability factors are dependent ,on treatment
system performance which is fairly consistent given that all plants use
biological treatment and perform well (i.e., after the 95/40/100 editing
rule). Based on the results of this alternative subcategory weighted
variability factor analysis, the Agency has decided to retain its approach of
calculating overall average variability factors and applying them to all OCPSF
facilities.
Individual plant variability factors are listed in Tables VII-56 and
VII-57 for BOD5 and TSS, respectively. As shown in the tables, the average
BOD Maximum for Monthly Average and Maximum for Any One Day variability
factors are 1.47 and 3.97, respectively. The average TSS Maximum for Monthly
Average and Maximum for Any One Day variability factors are 1.48 and 4.79,
respectively.
VII-177
-------
W
U-l £3
O O
•H
a 4-» .c
3 O bO
CO 3 -H
•O *CT\ O i
^3 ^bo rH in o O CM r-» CM r- CM
QOO) VO O r-l CO CO CO CO OO
O M t* ^
m &<
4™* ^^ K?t- fcj^ f^\ ^«* ^^. i *v j. . —. ^^
c to
.Q to
CMCOVOCOt-lrHVO in
C3 ^^ ••---.- ^*"^ ^^ ^^
o o
s «
r~-iocooocyvi— co\o co
j^-5r-;COiric^cM coco cr\
rt w CD ^H co vo r-to co m o
QE-i iri>sr>TCO-3--*in -
-------
rt o
-H nJ
*%
•H XI O
nj CD rt
Q -H PH
(-4
rt
rH
cn
cn
tn
R
vo
eo
co
S
oj
•*
O
•
CO
J5-
en
a
M-4 O
O -H
M rt
ja M
s
5
CO
i
§
'
*
a
a
§
•H
•H G rt
T3 O
C C -H
Bj
G) 3 rt ^^
S rH !-l rH
M-l +-" -x
u-H-1 C bO
Q W
r^t o
B
inO)
O f*\
O
m
a
a
s
s
a;
rH 3
CH S
S
3
3
S
8
S
VII-179
-------
JS -H O
*-> JS 4->
C rt O
o -rt rt
CO
ON
CM
CO CO
VO
CM
rH -H O
•H .£) 4J
rt rt o
Q -H rt
(-1 fci
ON
•
CM
CM CM
CO
co
CM
m
CM
CO
CO
O 'H
Oi O
tnCL)
CO
ON
ON ON
ON
ON
CO
ON
vo
ON
CO
ON
CO
1^
ON
ON
CO
vH
CM
CO
^*
CM
CM
CM
ON
m
CM
VO
CM
vo
CM
m
ON
VII-180
-------
J3 rH 0
4-> -H 4-1
S -H
£
S
«J rt O
Q -H 03
03
Pi
&
»
Id
15 .
$
en
CM
*O,
vp
^
iO-
3
m
CO
M-l O
O -H
4-J
(-1 0)
g; >
-Q t-i
3 CO
S »^3
o
*
S
G
O
•H C4 rt X-N
T3 OJ M rH
tt) S 4-» ^v
» rH C bO
M-l QJ B
CO W C
s
a
s
'
s
a a *-;
0» 0) nj ^-s
S 3 l-i i-H
W3 MH C! bO
co M-I cu e
E-i w O -^
C!
O
o
8
8
3
S
S
O nJ
-------
«n
«
to
iH i-H VI
a -H o
•u .a *•>
G cfl o
o -H rt
a n fa
to
>>rH O
r-l 'H -M
•H JS CJ
rt oj crj
Q -H fa
l-l
co
<4-l O
O -H
*J
H rd
X) l-l
S OJ
D W
S-rt
4J 4J
n),
a r-j
<4-l
-
a
OJ
.
bo
o
0)
C (tj
co «w _
co M-I o •
H W C
VI rH
^0
3,
O nS
co
CO
0)
fa
\0
10
in
CO
CO
CO
00
CO
CO
CO
CO
CO
SO
CO
m m
CO
co
vo
VO
CO
VO
o
cd
fa
VO
CO
vo
«J
co
co
H
0)
9
M
vo
CM
VII-182
-------
2. BAT Effluent Limitations
As discussed in Section VI, the Agency has decided to control 63 toxic
pollutants under BAT Subcategory One (End-of-Pipe Biological Plants) and 59
toxic pollutants under BAT Subcategory Two (non-End-of-Pipe Biological
Plants). This section discusses the data editing rules and methodology used
to derive the toxic pollutant long-term averages and variability factors that
provide the basis of the final BAT effluent limitations guidelines for both
subcategories. , .
a. BAT Data Editing Rules
The BAT toxic pollutant data base has basically two sources of data:
1) data collected during EPA sampling studies, and 2) data submitted by
industry either in response to Section 308 Questionnaire requests or as a
result of submissions during the public comment periods for ,the March 21,
1983, Proposal, the July 17, 1985. Federal Register Notice of Availability, or
the December 8, 1986, Federal Register Notice of Availability. Table VII-58
presents a summary of the BAT toxic pollutant data sources as organized into
four sets for review and editing purposes.
In general, the Agency's BAT toxic pollutant data base editing criteria
were as follows:
• Analytical methodology had to be EPA-approved (or equivalent) and have
adequate supporting QA/QC documentation.
• It was not necessary to have influent-effluent data pairs for the same
day, because many treatment systems have a wastewater retention time
of more than 24 hours.
• Since most of the effluent data have values of ND, the average
influent concentration for a compound had to be at least 10 times the
analytical minimum level ;(ML) for the difference to be meaningful and
qualify effluent concentrations for calculation of effluent limits.
For in-plant control effluent data for steam stripping and activated
carbon, the average influent concentration for a compound had to be at
least 1.0 ppm.
• Exclude data for effluent that has been diluted more than 25 percent
after treatment, but before final discharge. NPDES monitoring data
often reflects such dilution, which may be discerned by reference to
the wastewater flow diagram in a plant's response to the 1983 Section
308 Questionnaire. Appendix VII-G characterizes the problems
associated with dilution of NPDES application Form 2C data.
VII-183
-------
TABLE VII-58.
PRIORITY POLLUTANT (PRIPOL) DATA SOURCES FOR THE FINAL OCPSF RULE
EPA Sampling Programs
1.1 37 Plant Verification Study, 1978-80
1.2 Five Plant Study, 1980-81 (EPA/CMA Study)
2.0 Twelve Plant Study, 1983-84
OCPSF Proposal. 48 FR 11828 (March 21. 1983)
3.1 Data attached to 28 public comments
1983 Supplemental "308" Questionnaire*
(sent to selected plants only)
3.2 Data submitted by 74 selected plants
NOA (Proposal Revision 1), 50 FR 29068 (July 17, 1985)
4.1 Data attached to comments, or requested by EPA
as an extension of the attached data**
4.2 Requested from commenters, because the comment
implied that supporting data were available**
NOA (Proposal Revision 2). 51 FR 44082 (Dec. 8, 1986)
4.3 Data attached to comments from 5 commenters
Data Set 1
Data Set 2
Data Set 3
Data Set 4
™f r?f r?f EirV Prlori1ty Pollutant data submitted in response to
questions C13-C16 of the general questionnaire were average concentration
values instead of daily concentration values. This precluded the use of the
data for statistical calculation of effluent limitations.
**Data from a total of 21 plants were reviewed for data sets 4.1 and 4.2.
VII-184
-------
• Cyanide should be considered as having an analytical minimum level of
0.02 mg/1, and subject to the four criteria listed above.
• For data submitted by industry, exclude total phenols data, which
become meaningless with the specific measurement of phenol (priority
pollutant 65). The total phenol parameter represents a colorometric
response to the 4-Aminoantipyrine (4-AAP) reagent, which is non-
specific and characteristic of a host of both phenolic and non-
phenolic organic chemicals.
• Data not representative of BAT technology performance were eliminated
from the data base. Examples of reasons for not being representative
of BAT technology performance include process spills; treatment system
upsets; equipment malfunctions; performance not up to design specifi-
cations; past historical performance; or performance exhibited by
other plants in the data base with BAT technology in place.
• Exclude data for pollutants that could not be validated as present
based on the product/processes and the related process chemistry
associated with each product/process. Examples include phthalate
esters found because of sample contamination by the automatic sampler
tubing and methylene chloride found because of sample contamination in
the laboratory (methylene chloride is a common extraction solvent used
in GC/MS methods).
• Data for pollutants that do not satisfy the 10 times ML editing
criteria at the influent to'the end-of-pipe treatment sampling site,
because their original raw waste concentrations had been reduced
previously by an in-plant control technology, were retained when
sufficient information (i.e., verification, 12-Plant Sampling Reports,
or Section 308 Questionnaire) was available to validate the in-plant
control's presence.
In addition to the detailed editing criteria presented above, more
general editing criteria involved;
• Deletion of presampling grab samples collected prior to the EPA
12 Plant Sampling Study
• Choosing the appropriate sampling sites for the treatment system of
interest( e.g., influent to and effluent from steam stripper for BAT
Subcategory Two data base)
• Deletion of not quantifiable (NQ) values discussed above
• Averaging of replicate and duplicate samples or analyses at a sampling
site by day and, if appropriate, then across multiple laboratories.
All data points in decimal form as a result of replicate and duplicate
averaging were rounded to the nearest whole number (in ppb)
VII-185
-------
SSt I™ ? Z!r° dlscha5&ers a°d plants without appropriate BAT or
PSES treatment systems (e.g., indirect dischargers without appropriate
in-plant controls such as steam stripping, and direct dischargers
without end-of-pipe biological treatment or in-plant controls).
BAT
• Deletion of plants with more than the recommended BAT treatment
technology. [Plant 2680V from the BAT Subcategory Two data base]
• Deletion of plants without a combined raw waste sampling point, or if
430V,P1563V] r°CeSS Sampling data were collected at a plant. [Plants
tOXl P°llutant data f"m six plants for which
vere utnized- [piants
• Deletion of plant/pollutant combinations for which no effluent data
exist
™tHm* combinations when all influent values were
not detected (ND) (except for the overrides discussed above for
pollutants that do not satisfy the 10 times ML editing criteria)
• All values reported by the analytical laboratory at less than the
level minimum level were set equal to the analytical minimum
*nn °f,combined Pollutant analytical results (e.g., anthracene
and phenanthrene reported as a combined total concentration)
»na Xf ora^ory-comPosited volatile grab samples as required by
the analytical protocols instead of individual grab or automatic
composite sample analyses
• Deletion of plant/pollutant combinations based on BAT Option III
and £ni°S ^'*" ^^ Controls, end-of-pipe biological treatment,
and end-of-pipe activated carbon). [Plant 1494V, benzene]
• Deletion of plants which will be regulated under another point source
category. [Plant 1099V under the Petroleum Refining Point Source
Category] .
In addition to the editing criteria mentioned above, the Agency also
established another set of editing criteria in reviewing priority pollutant
metals data:
°!! *riori*y Poll«tant metals from non-process sources,
» non-contact cooling water blowdown and ancillary sources. An
example of an ancillary source is caustic, which commonly assays for
low levels of Cr(119), Cu(120), Ni(124), and sometimes Hg(123)
VII-186
-------
• Excluded end-of-pipe (NPDES) data, as well as data from other sampling
points, that do not represent the direct effluent from technology that
is specifically for the control of metals. In general, NPDES monitor-
ing data do not directly reflect the reduction of.priority, pollutant
'metal concentrations by such technology. Rather, the data reflect
dilution (by process wastewater and non-contact cooling water) and/or
absorption into biomass (if biological treatment of the process waste-
water is employed). Both dilution and biomass absorption of priority
pollutant metals are plant-specific factors that vary widely through-
out OCPSF wastewater collection and treatment systems. -
• Exclude complexed priority pollutant metal data, unless it is .the
direct effluent from technology that is specifically for the control
of complexed priority pollutant metals. This edit is generally appli-
cable to priority pollutant metals (e.g., chromium*3 and copper+2)
that have been very strongly complexed with organic dyes or chelating
compounds, so that the metal remains in solution and is unresponsive
to precipitation with usual reagents (lime or caustic).
• Exclude data that represent the direct effluent from technology
specifically for the control of metals, if there is no corresponding
influent data with which to evaluate the effectiveness of the
technology. .
The Agency's editing procedure differed somewhat for each data, source.
The data from the EPA sampling programs were edited using a combination of
computer analysis and manual analysis by Agency personnel. This was done
because al.l sampling data had previously been encoded. Data submitted by
industry were first reviewed to determine if the data submitted warranted,
encoding for further study, lending itself to manual editing rather than
computer analysis. However, all manual editing that could be validated by
computer analysis (e.g., the 10 x ML/1.0 ppm edit) was performed. Based on
this analysis, data from industry sources for a total of 17 plants were
retained for use in calculation of final BAT effluent limitations. Table
VI1-59 presents a summary of the data retained for each plant and how-, it was
utilized.
Table VII-60 presents a detailed explanation of the data excluded from
the limitations analysis based on the BAT performance editing criterion.
Based on this analysis, data from a total of 36 plants (plus six plant
overlaps due to resampling) for Subcategory One and 10 plants for Subcategory
Two (with nine plant overlaps with Subcategory One) from Agency studies and
public comments were retained for the limitations analysis, and, are presented
in Table VII-61 for BAT Subcategory One and Table VII-62 for BAT Subcategory
Two.
VII-187
-------
TABLE VII-59.
DATA RETAINED FROM DATA SETS 3 AND 4 FOLLOWING
BAT TOXIC POLLUTANT EDITING CRITERIA
Plant ID
63
387
500
682
1012
1650
1753
2227
1617
2445
2693
267
399
415
913
1769
1774
T, -,-, Data
Pollutants set
Zinc 3
Zinc 3
Nitrobenzene 3
Toluene 3
Zinc 3
Benzene, Naphthalene, Phenanthrene, 3
Toluene
Ethylbenzene 3
1,2-4-Trichlorobenzene, 1,2-Dichloroben- 3
zene, Nitrobenzene
Toluene 3
Methylene Chloride, Phenol 3
Chloroform, Methylene Chloride 3
Methylene Chloride 4
Zinc 4
Benzene, Toluene 4
1,2-Dichloroethane, 1,1,1-Trichloroe thane, 4
1,1,2-Trichlorethane, Chloroethane, Chloro-
form, 1,1-Dichloroe thane, 1,2-Trans-
Dichloroethylene, 1,1-Dichloroethylene,
Methylene Chloride, Tetrachloroethylene,
Trichloroethylene, Vinyl Chloride
Chlorobenzene, Chloroethane, 4
1 , 2-Dichlorobenzene, 2 , 4-Dini trotoluene ,
2,6-Dinitrotoluene, Nitrobenzene, Phenol
Zinc 4
BAT Subcategory
Data Base
One and Two
One and Two
Two Only
One Only
One and Two
One Only
One Only
One Only
One Only
One Only
One Only
One Only
One and Two
Two Only
Two Only
One Only
One and Two
VII-188
-------
TABLE VII-60.
F BAT TOXIC POLLUTANT DATA
PERFORMANCE EDITS
0 w
S3
Lj
<**
§
3
?\
P4
W
Explanation
QJ
B
td
*£
|
D
i-H
i — 1
O
a
4-i
§
i-H
P-l
This pollutant should be treated in-plant with activated carbon
prior to discharge to the end-of-pipe biological system. This
plant pretreats its phenolic wastewaters with a trickling filter
that is adequate for phenol but not for 2,4,6-trichlorophenol.
,_,
O
QJ
•G
0
ii
0
3
*£"!
H-
i
T1
"**
CM
CK
1 —
CM
6 —
This pollutant should be (but was not) treated in-plant with s tea
stripping prior to discharge to the end-of-ppe biological system
also, compared to data from plants retained in the data base
treating chlorobenzene with only biological treatment and having
similar raw waste concentrations, this plant's treatment system
performance for this pollutant was considered inadequate.
01
a)
§
fl
o
,—f
0
m
CM
in
This plant experienced a polyols spill during the sampling
period which saturated the end-of-pipe activated carbon
columns. These columns are used as an integral segment of
the treatment system rather than a polishing step. Th«efore,
because the listed pollutants were basically passing through the
activated carbon system untreated, this plant's treatment system
performance was considered inadequate.
01
c
a o -
-------
w
. PL, I
vo O !
^H
HS-S
3!
3'
Hi
-o
, o> +•<
J
co y -H
3 «H rH
S bO
o >> cu
/-MH y 43
0) O E *->
y rH CO -H E -H CU
•rt 4-1 CU y CO 4J *O
43 bo
rH O
O E -43 I
:r 52 •ri •y
E CU E _
•rt O, O) 4->
Cd43cdCOUOCUE
CU i
43 CO 43 bO *-> • 43 C
O I
rH T3 O O
o >> y 3
l-l CO -H O rH
u a) cu
1-4 CU
rH rH rH & 43
E I Cd 43
o s
•H o •-
CO
CO T3 bO'H 4J i
o > e o
Cd CO
en o -w *j
'a; •" >,
"bOCW
bO
rH CU B CU
cd a)
'HOO.CdO.J-'y CU-H
3 to CX-H
O T3 J-1 n) 4)
0) l-l
T3B I bO(UO4JOOCUU
I O M-l O -I-1 4-J CO rH 4^ 4J 4-*
0) B
y s
•H O rH CO
*•" I O -H -O
I T3 -rt rH CU
I B 43 l-l
y i-i s co cu
•H o cr~
•H CU 4J
•H O 0) (U O
u -H y js
OJ 4.J C 4J 4J
a, n) o
vo
tl .. 4J f3 Q, CU
y o v>
cd co
a. w o
rH 4J -
+•> O bOTJ
I O
J3 4->
bo cd o
14 H
cd •« cu
WBCOytO
,- - CO CO „
> 4J 4V .H O CO
O B bo B 4J U
S S E cd T3 CU
*-' OByB-H43CU (UB-H
-O-rtWCU^Ha) CUQ.4-.0 rH04J
y co 4->
rH O
cu y cd
CO 4-> O X -H
CU Cd B rH (U 43 CO
l-l > Q) 43 4J 4J
cu y y co
cd co
C/l
CO CO
&
U
rH 43
43 O
I O 3
CM C rH
•• cu o
rH 43 H
CU
~ o ••
cu u cu
BOB
CU rH -H
N 43 T3
B y -H
CD -H N
43 l-t B
OHO)
M I 43
O VO O
rH - M
CN 43
•• y
0) -^.H
B CU Q
O) B I
N Cd -
43 CO
4) 01 '^
T5 4-> CO
•H t/3 00
O O CM
rH & O
' >> O en
43 boeo
*j cu ^
0) *J O>
» S?s
-.a in
CU 3
o cu
rH -T3 a
43 -H 4->
U i-i rd
O Q
rH Q,
>,rH O,
« B M
CU -rt 4J
E > CO
CU CXi
cu a>
3 3
43 43
CO CO
en
tH
CO
VII-190
-------
TABLE VII-61.
PLANT AND POLLUTANT DATA RETAINED IN BAT ORGANIC TOXIC POLLUTANT
DATA BASE FOR BAT SUBCATEGORY ONE LIMITATIONS
Plant ID
2394
2536
725
3033
384
415
Data Set Pollutant #
1 7
25 .
27
38
57
58
59
65
86
1 3
38
65
1 6
9
12
23
44
45
52
85
88
1 10
32
34
55
65
85
1 4.
38
55
65
76
86
1 10
14
16
23
' • ' "•• - 29
30
32
44
87
Pollutant Name
Chlorobenzene
1 , 2-Dichlorobenzene
1 , 4-Dichlorobenzene
Ethylbenzene
2-Nitrophenol
4-Nitirophenol
2 , 4-Dini trophenol
Phenol
Toluene
Acrylonitrile
Ethylbenzene
Phenol
Carbon Tetrachloride
Hexachlorobenzene
Hexachloroe thane
Chloroform
Me thylene Chloride
Chloromethane
Hexachlorobu tad i ene
Tefrachloroe thylene
Vinyl Chloride
1,2-Dichloroethane
1,2-Dichloropropane
2 , 4-Dimethylphenol
Naphthalene
Phenol
Tetrachloroe thylene
Benzene
Ethylbenzene
Naphthalene
Phenol
Chrysene
Toluene
1,2-Dichloroethane
1,1, 2-Trichloroethane
Chloroe thane
Chloroform
1,1, -Dichloroethylene
1 , 2-Trans-dichloroe thylene
1 , 2-Dichloropropane
Methylene Chloride
Trichloroe thylene
VII-191
-------
TABLE VII-61.
PLANT AND POLLUTANT DATA RETAINED IN BAT ORGANIC TOXIC POLLUTANT
DATA BASE FOR BAT SUBCATEGORY ONE LIMITATIONS
(Continued)
2313
2631
2481
1
4
34
39
55
65
72
73
74
75
76
77
78
80
81
84
86
8
24
25
26
31
58
81
4
10
14
16
23
29
30
32
33
38
44
86
87
4
56
59
Acenaphthene
Benzene
2,4-Dimethylphenol
Fluoranthene
Naphthalene
Phenol
Benzo(a)Anthracene
Benzo(a)Pyrene
3,4-Benzofluoranthene
Benzo(k)Fluoranthene
Chrysene
Acenaphthylene
Anthracene
Fluorene
Phenanthrene
Pyrene
Toluene
1,2,4-Trichlorobenzene
2-Chlorophenol
1,2-Dichlorobenzene
1,3-Dichlorobenzene
2,4-Dichlorophenol
4-Nitrophenol
Phenanthrene
Benzene
1,2-Dichloroethane
1,1,2-Trichloroethane
Chloroethane
Chloroform
1,1-Dichloroethylene
1,2-Trans-dichloroethylene
1,2-Dichloropropane
1,3-Dichloropropene
Ethylbenzene
Methylene Chloride
Toluene
Trichloroethylene
Benzene
Nitrobenzene
2,4-Dini trophenol
VII-192
-------
TABLE VII-61.
PLANT AND POLLUTANT DATA RETAINED IN BAT ORGANIC TOXIC POLLUTANT
DATA BASE FOR BAT SUBCATEGORY ONE LIMITATIONS
(Continued)
Plant ID
948
267
12
2221
2711
725
444
Data Set Pollutant #
2 3
4
10
29
38
65
66
68
70
71
86
2 8
25
31
65
2 1
4
34
38
55
65
86
3 38
65
86
3 65
86
3 6
10
12
23
30
52
85
88
3 4
86 ,
Pollutant Name
Acrylonitrile
Benzene
1 , 2-Dichloroe thane
1 , 1-Dichloroethylene
Ethylbenzene
Phenol
Bis-(2-Ethylhexyl)Phthalate
Di-N-Butyl Phthalate
Diethyl Phthalate
Dimethyl Phthalate
Toluene
1 , 2-4-Trichlorobenzene
1 , 2-Dichlorobenzene
2 , 4-Dichlorophenol
Phenol
Acenaphthene
Benzene
2 , 4-Dimethylphenol
Ethylbenzene
Naphthalene
Phenol
Toluene
Ethylbenzene
Phenol
Toluene
Phenol
Toluene
Carbon Tetrachloride
1 , 2-Dichloroethane
Hexachloroe thane
Chloroform
1 , 2-Trans-dichloroethylene
Hexachchlorobutadiene
Tetrachloroethylene
Vinyl Chloride
Benzene
Toluene
VII-193
-------
TABLE VII-61.
PLANT AND POLLUTANT DATA RETAINED IN BAT ORGANIC TOXIC POLLUTANT
DATA BASE FOR BAT SUBCATEGORY ONE LIMITATIONS
(Continued)
Plant ID Data Set Pollutant
695 3 4
6
10
23
24
25
29
32
38
42
44
55
65
86
L650 3 4
38
55
65
77
80
81
86
•48 3 3
65
66
68
70
71
430 3 4
55
65
86
349 3 3
88
# Pollutant Name
Benzene
Carbon Tetrachloride
1 , 2-Dichloroethane
Choloroform
2-Chlorophenol
1 , 2-Dichlorobenzene
1,1-Dichloroethylene
1 , 2-Dichloropropane
Ethylbenzene
Bis-(2-Chloroisopropyl) Ether
Methylene Chloride
Naphthalene
Phenol
Toluene
Benzene
Ethylbenzene
Naphthalene
Phenol
Acenaphthylene
Fluorene
Phenanthrene
Toluene
Acrylonitrile
Phenol
Bis-(2-^Ethylhexyl) Phthalate
Di-N-Butyl Phthalate
Diethyl Phthalate
Dimethyl Phthalate
Benzene
Naphthalene
Phenol
Toluene
Acrylonitrile
Vinyl Chloride
VII-194
-------
TABLE VII-61.
PLANT AND POLLUTANT DATA RETAINED IN BAT ORGANIC TOXIC POLLUTANT
DATA BASE FOR BAT,SUBCATEGORY ONE LIMITATIONS
(Continued)
Plant ID
1494
883
659
1609
851
1890
1890*
Data Set Pollutant #
3 25
35
36
44
56
57
58
59
65
86
3 3
38
3 38
3 4
23
24
- 31
65
86
87
3 4
38
39
55
78
80
81
84
86
3 86
3 65
86
Pollutant Name
1 , 2-Dichlorobenzene
2 , 4-Dini t ro toluene
2,6-Dinitrotoluene
Methylene Chloride
Nitrobenzene
2-Nitrophenol
4-Nitrophenol
2 , 4-Dini trophenol
Phenol
Toluene
Acrylonitrile
Ethylbenzene
Ethylbenzene
Benzene
Chloroform
2-Chlorophenol
2 , 4-Dichlorophenol
Phenol
Toluene
Trichloroethylene
Benzene
Ethylbenzene
Fluoranthene
Naphthalene
Anthracene
Fluorene
Phenanthrene
Pyrene
Toluene
Toluene
Phenol
Toluene
VII-195
-------
TABLE VII-61.
PLANT AND POLLUTANT DATA RETAINED IN BAT ORGANIC TOXIC POLLUTANT
DATA BASE FOR BAT SUBCATEGORY ONE LIMITATIONS
(Continued)
Plant ID
Data Set Pollutant #
Pollutant Name
2631 3 4
10
11
14
16
23
29
32
33
38
55
65
86
i051 3 4
10
32
33
86
87
196 3 4
10
11
65
86
06 3 1
4
34
39
65
72
76
77
78
81
84
86
57 4 44
'2 4 86
Benzene
1 , 2-Dichloroethane
1,1, 1-Trichloroethane
1,1,2-Trichloroethane
Chloroethane
Chloroform
1,1-Dichloroethylene
1 , 2-Dichloropropane
1 , 3-Dichloropropene
Ethylbenzene
Naphthalene
Phenol
Toluene
Benzene
1 , 2-Dichloroethane
1, 2-Dichloropropane
1, 3-Dichloropropene
Toluene
Trichloroethylene
Benzene ,
1 , 2-Dichloroethane
1,1, 1-Trichloroethane
Phenol
Toluene
Acenaphthene
Benzene
2 , 4-Dimethylphenol
Fluoranthene
Phenol
Benzo(a)Anthracene
Chrysene
Acenaphthylene
Anthracene
Phenanthrene
Pyrene
Toluene
Methylene Chloride
Toluene
VII-196
-------
TABLE VII-61.
PLANT AND POLLUTANT DATA RETAINED IN BAT ORGANIC TOXIC POLLUTANT
DATA BASE FOR BAT SUBCATEGORY ONE LIMITATIONS
(Continued)
Plant ID
Data Set Pollutant #
Pollutant Name
1617
1650
1753
4
4
86
4
55
81
86
38
Toluene
Benzene
Naphthalene
Phenanthrene
Toluene
Ethylbenzene
1769
2227
2445
2693
4 7
16
25
35
36
56
65
4 8
25
56
4 44
65
4 23
44
Chlorobenzene
Chloroe thane
1 , 2-Dichlorobenzene
2 , 4-Dini trotoluene
2 , 6-Dini trotoluene
Nitrobenzene
Phenol
1 , 2-4-Trichlorobenzene
1 , 2-Dichlorobenzene
Nitrobenzene
Methylene Chloride
Phenol
Chloroform
Methylene Cloride
Note: * denotes a plant which had tvo different treatment systems in the data
base
Data Set 1 denotes 12-Plant Study.
Data Set 2 denotes 5-Plant Study.
Data Set 3 denotes Verification Study.
Data Set 4 denotes public comments and supplemental questionnaire data.
VII-197
-------
TABLE VII-62.
PLANT AND POLLUTANT DATA RETAINED IN BAT ORGANIC TOXIC POLLUTANT
DATA BASE FOR BAT SUBCATEGORY TWO LIMITATIONS
Plant ID
725
1494
415
2680
415
913
2680
500
)48
Data Set Pollutant #
1 44
45
88
1 4
1 10
14
16
23
' 29
30
44
87
1 4
3 4
86
3 10
11
13
14
16
23
29
30
44
85
87
88
1 56
57
58
59
60
3 56
2 66
68
70
71
Pollutant Name
Methylene Chloride
Chlorome thane
Vinyl Chloride
Benzene
1 , 2-Dichloroetheane
1, 12-Trichloroethane
Chloroe thane
Chloroform
1 , 1-Dichloroethylene
1 , 2-Trans-Dichloroethylene
Methylene Chloride
Trichloroethylene
Benzene
Benzene
Toluene
1 , 2-Dichloroethane
1 , 1 , 1-Trichloroethane
1 , 1-Dichloroethane
1,1, 2-Trichloroethane
Chloroe thane
Chloroform
1 , 1-Dichloroethylene
1 , 2-Trans-Dichloroethylene
Methylene Chloride
Te t rachloroe thylene
Trichloroethylene
Vinyl Chloride
Nitrobenzene
2-Nitrophenol
4-Nitrophenol
2 , 4-Dini trophenol
4 , 6-Dini tro-o-Cresol
Nitrobenzene
Bis-(2-Ethylhexyl) Phthalate
Di-n-Butyl Phthalate
Diethyl Phthalate
Dimethyl Phthalate
VII-199
-------
TABLE VII-62.
PLANT AND POLLUTANT DATA RETAINED IN BAT ORGANIC:TOXIC POLLUTANT
DATA BASE FOR BAT SUBCATEGORY TWO LIMITATIONS
(Continued)
Plant ID
Data Set
Pollutant #
Pollutant Name
2536 1 3
1293 1 1
34
39 '
55
65
72
73 "
74 , '
75
76
77
78
80
81
84
Acrylonitrile
Acenaphthene
2 , 4-Dimethylphenol
Fluoranthene
Naphthalene
Phenol
Benzo(a) Anthracene
Benzo(a)Pyrene
3 , 4-Benzof luoranthene
Benzo(k) Fluoranthene
Chrysene
Acenaphthylene
Anthracene
Fluorene
Phenanthrene
Pyrene
Note: Data Set 1 denotes 12-Plant Study.
Data Set 2 denotes 5-Plant Study.
Data Set 3 denotes public comments and supplemental questionnaire data.
VII-200
-------
One industry commenter questioned the validity of treating pollutant data
from one plant in two different sampling projects independently. It should be
noted that the six plant overlaps occur because these plants were either
sampled in separate Agency studies or the Agency received data submitted by
commenters in addition to its sampling studies. EPA has treated these over-
lapping plant data sets separately for limitations calculations purposes
because of general changes in a plant's production levels and product mix, and
changes in a plant's treatment system or treatment system operation in the
time period between sampling studies. Using the plant data in this manner did
not significantly affect most of the pollutants being regulated.
EPA reviewed its files on these six plants relating to circumstances at
the plants during the sampling episodes. Plant 725 upgraded a steam bath to a
steam stripper by adding trays between sampling episodes. Plant 2631 had two
processes in operation during the first sampling event and three on the
second. EPA, accordingly, maintains that the 4 data sets associated with
these 2 plants be treated separately because of the referent known changes.
For the remaining four plants, EPA combined the corresponding eight data
subsets into four to yield a single data set for each of the four plants. EPA
then recomputed all of the end-of-pipe BAT toxic limitations to perform a
comparative analysis of these results to those for the EPA methodology for
calculating daily maximum limitations for all of the 55 organic pollutants
derived by this analysis.
The findings were that 11 of the 55 daily limitations changed value, but
for seven of the 11 changes the shifts were only 5 percent or less. For the
four limitations that showed larger changes, two increased and two decreased.
EPA maintains that the general rationale for treating these six plants as
12 separate entities is appropriate and that there is no bias introduced by
this approach.
VII-200
-------
b. Derivation of BAT Toxic Pollutant LTAs
Table VII-63 presents a .summary of the plants retained in the BAT toxic
pollutant data base for BAT Subcategory One and Two, and the in-plant and
end-of-pipe technologies in-place at each plant based on the 1983 Section 308
Questionnaire for industry-supplied data and on field sampling reports for,EPA
data. The table shows that the technology basis for the data to be used for
BAT Subcategory One is mainly end-of-pipe biological treatment (in the form of
activated sludge) preceded in many cases by some form of in-plant control.
These in-plant controls are sometimes in the form of highly efficient tech-
nologies such as activated carbon or steam stripping, or are a more gross form
of control used more for product recovery (e.g., distillation), but nonethe-
less contributing to a reduction.or equalization of raw waste concentrations
discharged to the end-of-pipe biological treatment system. The technology
basis for the BAT Subcategory Two toxic pollutant data base is based on
performance data from in-plant controls such as steam stripping, activated
carbon, and in-plant biological treatment.
For each pollutant at each plant from each of the four data sets, an
estimated long-term average (LTA) effluent concentration .was calculated. The
nondetected values at a plant were assigned an analytical minimum level value
using the minimum levels associated with EPA analytical methods 1624 and 1625.
The estimated long-term average was computed using a method that assigned
nondetected values a relative weight in accordance with the frequency with
which nondetected values for the pollutant were found in the daily data plants
as defined in Appendix VIII-C.
The estimated long-term average, m, for a plant-pollutant combination is
as follows: .
X.
= pD
(1 - p)
n
VII-201
-------
TABLE VII-63.
TREATMENT TECHNOLOGIES FOR PLANTS IN THE
FINAL BAT TOXIC POLLUTANT DATA BASE
Plant I.D.
Treatment Technology
2394
2536
725
3033
384
415
1293
2313
2680
2481
948
267
12
2221
2711
444
Steam stripping, distillation, chemical oxidation, thio-
sulfate waste reuse, sewer segregation, phase separation,
EQ, NEU, GRSP, ASL, SCLAR, POL, PAER
Gravity separation, EQ, NEU, SCR, CLAR, ASL, SCLAR, FILT
Steam stripping, API separator, EQ, NEU, FLOCC, CLAR, ASL,
SCLAR, FILT, CHLOR, SLDTH, SLDFILT
NEU, SCSP, NUDADD, ALA, SSIBS, SETTLING LAGOON, POL, FILT,
CAD, SSITS, POLISH BAGFILTERS
EQ, NEU, API, ASL, SCLAR, POL
Air stripping, steam stripping, carbon adsorption, distil-
lation, retention impoundment, oil separation, API
separation, EQ, NEU, CLAR, NUDADD,1 MULTISTAGE POASL, SCLAR
Primary settling, oil removal, EQ, BIOLOGICAL DIGESTION,
CLAR
Chemical precipitation, steam stripping, solvent
extraction, distillation, chemical oxidation, filtration,
equalization, EQ, NEU, CLAR, NUDADD, ASL, PACA, SCLAR
Decant sump, EQ, NEU, SS, CAD
Carbon adsorption, EQ, NE, SCR, CLAR, FLOCC, ASL, SCLAR
NEU, ASL, SCLAR, POL
Steam stripping, NEU, SCR, OLSK, OLS, CLAR, NUDADD, TF,
ASL, SCLAR, POL . ..
Solvent extraction, decantation, EQ, NEU, OLS, API, NUDADD,
ASL, SCLAR
Solvent extraction, carbon adsorption, distillation, EQ,
GR, ASL, SCLAR
EQ, ARL, ANL, SCLAR
EQ, NEU, ASL, SCLAR, DAF
VII-202
-------
TABLE VII-63.
TREATMENT TECHNOLOGIES FOR PLANTS IN THE
FINAL BAT TOXIC POLLUTANT DATA BASE
(Continued)
Plant I.D.
Treatment Technology
695
2430
1349
1494
883
659
1609
851
1890
1890*
2631
4051
« '296
306
63
387
Chemical precipitation, steam stripping, chemical
oxidation, filtration, separation, catalyst recovery, EQ,
NEU, OLSK, OLS, DAF, CLAR, FLOCC, NUDADD, ALA, SCLAR
EQ, NEU, OLS, DAF, FLOCC, NUDADD, TF, POASL, SCLAR
Steam stripping, EQ, NEU, CLAR, COAG, FLOCC, NUDADD, ASL,
SCLAR, POL
Steam stripping, solvent extraction, EQ, NEU, CLAR, ASL,
SCLAR, CAD
EQ, ASL, SCLAR, POL, FILT
EQ, NEU, SCR, DAF, COAG, FLOCC, ALA, SCLAR
EQ, NEU, CLAR, ASL, SCLAR
EQ, API, NUDADD, ASL, TF, SCLAR
Septic tank, API separator, gravity separation, ion
exchange, steam stripping, GR, API, EQ, NEU, API, NUDADD,
ALA, TF, FSA, SCLAR, FILT, CHLORINE ADDITION
Septic tank, API separator, EQ, NEU, NUDADD, ASL, SCLAR,
FILT, AERATION
Steam stripping, solvent extraction, EQ, NEU, API, CLAR,
ASL, SCLAR
API, ALA, DAF
Steam stripping, ion exchange, distillation, decantation,
org. recovery^ EQ, NEU, GR, OLSK, CLAR, ALA, POASL, SCLAR
Steam stripping, EQ, NEU, OLS, FLOCC, NUDADD, ASL, SCLAR,
FILT
Distillation, chemical precipitation, evaporation, EQ,
CLAR, ARL, ASL, SCLAR, CHLOR
Filtration, crystallization, evaporation, EQ, NEU, SCR,
CLAR, NUDADD, POLISHING BASIN, ASL, SCLAR
VII-203
-------
TABLE VII-63.
TREATMENT TECHNOLOGIES FOR PLANTS IN THE
FINAL BAT TOXIC POLLUTANT DATA BASE
(Continued)
Plant I.D.
Treatment Technology
500
682
913
1012
1617
1650
1753
1769
1774
2227
2445
2693
Steam stripping, carbon adsorption, spill containment, NEU,
CLAR, ASL, SCLAR, POL, pH ADJUSTMENT
Settling, flotation, EQ, NEU, SCR, CLAR, COAG, SETTLING,
FLOTATION, MIXING, SURFACE BAFFLES, ASL, SCLAR, DEAERATION
Steam stripping, chemical oxidation, phase separation, EQ,
NEU
EQ, SEDIM, CP, RBC, TF, SCLAR, SEDIM
Distillation, EQ, COAG, SAND BED FILTRATION, TF, SCLAR, POL
NEU, SCR, OLSK, OLS, API, ARL1, ARL2, ARL3, ARL4, ARL5,
ARL6, ANL
EQ, NEU, CLAR, NUDADD, POLADD, CP, POASL, SCLAR
Chemical precipitation, NEU, CLAR, NUDADD, FLOCC, ASL,
PACA, SCLAR, POL
EQ, NEU, CLAR, FLOCC, FILT
EQ, NEU, CLAR, FLOCC, NUDADD, ASL, SCLAR
Dissolved air flotation, EQ, NEU, SCR, API, CLAR, NUDADD,
POASL, SCLAR
Chemical precipitation, steam stripping filtration, EQ,
NEU, NUDADD, ASL, SCLAR
Note: The order in which these treatment technologies are listed does not
necessarily indicate that they are in series, since certain plants
employ multiple treatment systems to treat segregated waste streams.
*Two separate treatment systems were sampled at the same plant during the same
sampling study.
VII-204
-------
TABLE VII-63.
TREATMENT TECHNOLOGIES FOR PLANTS IN THE
FINAL BAT TOXIC POLLUTANT DATA BASE
(Continued)
Key:
CND - Cyanide Destruction
CP - Chemical Precipitation
CHRRED - Chromium Reduction
AS - Air Stripping
SS - Steam Stripping
DISTL - Distillation
EQ - Equalization
NEU - Neutralization
SCR - Screening
GR - Grit Removal
OLSK - Oil Skimming
OLS - Oil Separation
API - API Separation
DAF - Dissolved Air Flotation
CLAR - Primary Clarification
COAG - Coagulation
FLOCC - Flocculation
NUDADD - Nutrient Addition
ASL - Activated Sludge
ALA - Aerated Lagoon
ARL - Aerobic Lagoon
ANL - Anaerobic Lagoon
RBC - Rotating Biological Contractor
TF - Trickling Filters
POASL - Pure Oxygen Activated Sludge
SSIBS - Second Stage of Indicated Biological System
PACA - Powdered Activated Carbon Addition
SCLAR - Secondary Clarification
POL - Polishing Pond
FILT - Filtration
CAD - Carbon Adsorption
SSITS - Second Stage of Indicated Tertiary System
GRSP - Gravity Separation
PAER - Post Aeration
CHLOR - Chlorination
FSA - Ferrus Sulfide Addition
SLDTH - Sludge Thickening
SLDFILT - Sludge Filtering
AER - Aeration
SEDIM - Sedimentation
POLADD - Polymer Addition
Notes:
Upper Case:
Lower Case:
End-of-Pipe Treatment
In-Plant Control
VII-205
-------
where H^ is the estimated long-term average at plant j; D is the analytical
minimum level; n is the number of concentration values where Xi is detected at
or above the minimum level at plant j; and p is the proportion of nondetected
values reported from all the daily data base plants. That is, p equals the
total number of reported nondetected values from all daily data plants for a
particular pollutant divided by the total number of values reported from all
daily data plants for a particular pollutant. For plant-pollutant combina-
tions with all nondetected values, the long-term average, m, equals the
analytical minimum level. For plant-pollutant combinations where all values
are detected, the long-term average is the arithmetic mean of all values.
Pollutant group values for p were used when pollutant-specific estimates were
not available.
c. Steam Stripping Long-Term Averages
EPA is regulating 28 volatile organic pollutants based on steam stripping
technology. EPA had data on 15 of these pollutants, which were used to deter-
mine limitations using the same methodology used to determine other BAT
organic pollutant limitations. For 13 volatile organic pollutants controlled
by steam stripping, EPA lacked sufficient data to calculate estimated long-
terra averages directly from data relating to these pollutants. Instead, EPA
concluded that these pollutants may be treated to levels equivalent, based
upon Henry's Law Constants, to those achieved for the 15 pollutants for which
there were data. Dividing the 15 pollutants into "high" and "medium"
strippability subgroups, EPA developed a long-term average for each subgroup
and applied these to the 13 pollutants for which data were lacking (six
pollutants in the high subgroup and seven in the medium subgroup). The
long-term average for pollutants with no data in each subgroup was determined
by the highest of the long-term averages within each subgroup based upon the
15 pollutants for which the Agency had data. This approach tends to be
somewhat conservative but in the Agency's judgment not unreasonable in light
of the uncertainty that would be associated with achieving a lower long-term
average for the pollutants for which data are unavailable. The high
strippability long-term average thus derived is 64.5 jMg/1, while the medium
strippability long-term average is slightly higher, 64.7 ug/1.
VII-206
-------
While it may appear anomalous that the high strippable subgroup yields
just a slightly lower long-term average effluent concentration, EPA believes
that this is not the case. First, in the context of the maximum levels
entering the steam strippers within the two subgroups (12,000 ug/1 to over
23 million ug/1), the differences between these two long-term averages is
negligible and essentially reflect the same level of long-term control from an
engineering viewpoint. Second, the "high" and "medium" strippable compounds
behave comparably in steam strippers, in the sense that roughly the same low
effluent levels can be achieved with properly designed and operated steam
strippers. In other words, it is possible to mitigate small differences in
theoretical strippability among compounds in these groups with different
design and operating techniques. The small differences in long-term average
performance seen in the data reflect, in EPA's judgment, no real differences
in strippability among pollutants but rather the difference in steam stripper
operations among the plants from which the data were taken. Indeed, one could
reasonably collapse the two subgroups into one group and develop a single
long-term average for the 13 pollutants for which EPA lacks data. While such
an approach might be technically defensible, EPA decided it would be.most
reasonable to retain the distinction between "high" and "medium" subgroups,
which remains a valid and important distinction for the purpose of transfer-
ring variability factors, as discussed below.
Table VII-64 presents the long-term average values for each organic
pollutant, calculated by taking the median of the plant estimated averages for
those pollutants regulated under BAT Subcategory One and Two. The BAT
Subcategory One median of long-term average values for 1,1-dichloroethane and
4,6-dinitro-o-cresol have been transferred from BAT Subcategory Two. Since
the in-plant steam stripping and activated carbon units attain effluent levels
equal to the analytical minimum level, the addition of end-of-pipe biological
treatment for BAT Subcategory Two will not produce a measurable lower effluent
concentration.
d. Calculation of Daily Maximum and Maximum Monthly Average
Variability Factors>
After determining estimated long-term average values for each pollutant,
EPA developed two variability factors for each pollutant-^-a 99th percentile
VII-207
-------
TABLE VII-64.
BAT TOXIC POLLUTANT MEDIAN OF ESTIMATED LONG-TERM
AVERAGES FOR BAT SUBCATEGORY ONE AND TWO
Subcategory One Subcategory Two
Median of Median of
Pollutant- »• • , Estimated Estimated
MnmK^ D n „ fc „ Minimum Number Long-Term Number Long-Term
Number Pollutant Name Level of Plants Means of Plants Means
1 Acenaphthene
3 Acrylonitrile
4 Benzene
6 Carbon Tetrachloride
7 Chlorobenzene
8 1,2,4-Trichlorobenzene
9 Hexachlorobenzene
10 1,2-Dichloroe thane
11 1,1,1-Trichloroethane
12 Hexachloroe thane
13 1 , 1-Dichloroethane
14 1,1,2-Trichloroethane
16 Chloroethane
23 Chloroform
24 2-Chlorophenol
25 1 , 2-Di chlorobenzene
26 1,3-Dichlorobenzene
27 1,4-Dichlorobenzene
29 1,1-Dichloroethylene
30 Trans-l,2-Dichloroethylene
31 2,4-Dichlorophenol
32 1,2-Dichloropropane
33 1,3-Dichloropropene
34 2,4-Dimethyl Phenol
35 2, 4-Dinitro toluene
36 2,6-Dinitrotoluene
10
50
10
10
10
10
10
10
10
10
10
10
50
10
10
10
10
10
10
10
10
10
10
10
10
10
3
5
17
3
2
3
1
9
2
2
-
3
4
8
3
7
1
1
5
3
3
6
3
4
2
2
10.0 1
50.0 1
10.0 4
10.0
10.0
42.909
10.0
25.625 2
10.0 1
10.0
(10.0)** 1
10.0 2
50.0 2
12.208 2
10.0
47.946
24.80
10.0
10.0 2
10.0 2
17.429
121.50
23.00
10.794 1
58.833
132.667
10.00
50.00
28.5761
64.5000*
64.5000*
64.7218*
64.7218*
64.7218
10.0
64.7218*
10.00
10.2931
50.00
44.1081
64.7218*
64.5000*
64.5000*
10.0517
11.0517
64.7218*
64.7218*
10.00
VII-208
-------
TABLE VII-64.
BAT TOXIC POLLUTANT MEDIAN OF ESTIMATED LONG-TERM
AVERAGES FOR BAT SUBCATEGORY ONE AND TWO
(Continued)
Subcategory One Subcategory Two
Pollutant Minimum
Number Pollutant Name Level
38
39
42
44
45
52
55
56
57
58
59
60
65
66
68
70
71
72
73
74
75
76
77
78
80
81
Ethyl benzene
Fluoranthene
Bis-(2-Chloroisopropyl)
Ether
Methylene Chloride
Methyl Chloride
Hexachlorobutadiene
Naphthalene
Nitrobenzene
2-Nitrophenol
4-Nitrophenol
2 , 4-Dini trophenol
4 , 6-Dini tro-0-Cresol
Phenol
Bis(2-Ethylhexyl)Phthalate
Di-n-Butyl Phthalate
Diethyl Phthalate
Dimethyl 'Phthalate
Benzo (a) Anthracene
Benzo(a)Pyrene
3 , 4-Benzof luoranthene
Benzo (k) . Fluoranthene
Chyrsene
Acenaphthylene
Anthracene
Fluorene
Phenanthrene
10 i
10
10
10
50
10
10
14
20
50
50
24 ;
10
10
10
10
10
10 ,
10 ;
10 '
10
10
10
10
10
10
Median of Median ot
Estimated Estimated
Number Long-Term Number Long-Term
of Plants Means of Plants Means
14
3
1
8
1
2
10
4
2
3
3
.. -
22
2
2
2
2
2
1
1
1
3
3
-3
3
. 6
10.0
11.533
156.667
22.956
50.0
10.0
10.0
14.0
27.525
50.00
50.0
(24.0)**
10.363
47.133
17.606
42.50
10.0
10.0
10.333
10.267
10.00 '
10.0
10.0
10.0
10.0
1QJ).
-
1
-
3
1
-
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
64.5000*
11.5333
64.7218*
10.800
50.00
64.5000*
10.0
948.675
20.00
50.00
373.00
24.00
10.0
43.4545
13.0909
23.6667
10 ..00
10.00
10.333
10.2667
10.00
10.00
10.00
10.00
10.00
10.00
VII-209
-------
TABLE VII-64.
BAT TOXIC POLLUTANT MEDIAN OF ESTIMATED LONG-TERM
AVERAGES FOR BAT SUBCATEGORY ONE AND TWO
(Continued)
Subcategory One
Median of
Subcategory Two
Median of
Pollutant
Number Pollutant Name
Estimated Estimated
Minimum Number Long-Term Number Long-Term
Level of Plants Means of Plants Means
84
85
86
87
88
Pyrene
Te t rachloroe thy lene
Toluene
Trichloroethylene
Vinyl Chloride
10
10
10
10
50
3
3
24
4
3
11
10
10
10
50
.333
.4231
.00
.00
.0
1
1
2
2
2
10.
18.
12.
11.
64.
3333
4286
4177
5862
5000
Note: All units in ug/1 or ppb.
transferred median of long-term means by strippability groupings.
**Transferred from BAT Subcategory Two.
VII-210
-------
Maximum for Any One Day variability factor (VF1) and a 95th percentile Maximum
for Monthly Average variability factor ,(VF4). These .were developed by fitting
a statistical distribution to the daily data for each pollutant at each plant;
estimating a 99th percentile and a mean of the daily data distributions for
each pollutant at each plant; estimating a 95th percentile and a mean of the,
distribution of 4-day monthly averages for each pollutant at each plant;
dividing the 99th and 95th percentiles by the respective means of daily and
4-day average distributions to determine.plant-specific variability factors;
and averaging variability factors across all plants to determine a VF1-and VF4
for each pollutant. All plant-pollutant combinations for which variability -
factors were calculated have at least seven effluent concentration values
(including NDs) with at least three values at or above the minimum level.
For certain pollutants, the amount of daily data was limited and
individual pollutant variability factors could not be calculated. For such
pollutants regulated in BAT Subcategory One, variability factors were Imputed
from the variability factors for groups of pollutants expected to exhibit
comparable treatment variability based upon comparison of chemical structure
and characteristics. The priority pollutants'were^grouped, as shown in Table
VII-65, by generic classification based on a similarity of functional group or.
structure (isomers, homologs, analogs, etc.). As a consequence of these
similarities, members of each group share precursors, and/or have a common
response to generic process chemistry (7-27) and, in the Agency's judgment,
would be expected to exhibit similar characteristics in wastewater treatment
unit operations. Each pollutant in each chemical group without a variability
factor was then assigned a VF1 and VF4 equal to the average of the VFls and
VF4s of any pollutants in the same group. However, there are six pollutants
without individual variability factors that are also in pollutant variability
groups without an average variability factor. An overall average variability
factor based on all individual pollutant variability factors was transferred
to these pollutants [acrylonitrile, 2,4-dinitrotoluene, 2,6-dinitrotoluene,
~ bis(2-chloroisopropyl) ether, hexachlorobutadiene, and nitrobenzene]. In the
case of acrylonitrile and hexachlorobutadiene, the reason for not having
individual variability factors was not lack of sufficient daily data but that
all or nearly all values for these pollutants were not detected.
VII-211
-------
TABLE VII-65.
PRIORITY POLLUTANTS BY CHEMICAL GROUPS
Halogenated Methanes (Cls)
46
45
44
47
23
48
51
50
49
6
Methyl bromide
Methyl chloride
Methylene chloride (dichloromethane)
Broraoform (tribromomethane)
Chloroform (trichloromethane)
Bromod i chlorome thane
Dibromochloromethane
Dichlorodifluoromethane
Tri chlorofluorme thane
Carbon tetrachloride (tetrachloromethane)
2. Chlorinated C2s
16
88
10
13
30
29
14
11
87
85
15
12
Chloroethane (ethyl chloride)
Chloroethylene (vinyl chloride)
1,2-Dichloroethane (ethylene dichloride)
1,1-Dichloroethane
1,2-1 rans-Di chloroe thylene
1,1-Dichloroehtylene (vinylidene chloride)
1,1,2-Trichloroethane
1,1,1-Trichlorethane (methyl chloroform)
Trichloroethylene
Tetrachloroethylene
1,1,2,2-Tetrachloroethane
Hexachloroe thane
3. Chlorinated C3s
32 1,2-Dichloropropane
33 1,3-Dichloropropylene
4. Chlorinated C4
52 Hexachlorobutadiene
5. Chlorinated C5
53 Hexachlorocylopentadiene
6. Chloroalkyl Ethers
17 bis(chloromethyl)ether
18 bis(2-chloroethyl)ether
42 bis(2-chloroisopropyl)ether
19 2-chloroethylvinyl ether
43 bis(2-chloroethoxy) methane
VII-212
-------
TABLE VHr-65.
PRIORITY,POLLUTANTS BY CHEMICAL GROUPS
(Continued)
7. Metals
114 Antimony
115 Arsenic
117 Beryllium
118 Cadmium
119 Chromium
120 Copper
122 Lead
123 Mercury
124 Nickel
125 Selenium
126 Silver
127 Thallium
128 Zinc
.' ' HI .1 , _ f '4. ' :
8. Pesticides
89 Aldrin
90 Dieldrin
91 Chlordane
95 alpha-Endosulfan
98 Endrin
99 Endrin aldehyde
100 Heptachlor
101 Heptachlor epoxide
102 alpha-BHC
103 beta-BHC
104 gamma-BHC (Lindane)
105 delta-BHC
92 4,4'-DDT
93 4,4'-DDE (p,p'-DDx)
94 4,4'-DDD (p,p'-TDE)
113 Toxaphene
9. Nitrosamines
61 N-Nitrosodimethyl amine
62 N-Nitrosodiphenyl amine
63 N-Nitrosodi-n-propyl amin.e
10. Miscellaneous
2 Acrolein
3 Acrylonitrile
54 Isophorone
121 Cyanide
VII-213
-------
TABLE VII-65.
PRIORITY POLLUTANTS BY CHEMICAL GROUPS
(Continued)
11. Aromatics
4 Benzene
86 Toluene
38 Ethylbenzene
12. Polyaromatics
14.
15.
16.
55
1
77
78
72
73
74
75
76
79
82
80
39
83
81
84
Naphthalene
Acenanaphthene
Acenaphthylene
Anthracene
Benzo(a)anthracene (1,2-benzantharacene)
Benzo(a)pyrene (e,4-benzopyrene)
3,4-Benzofluorantehne
Benzo(k)fluorantehene (11,12-benzofluoranthene)
Chrysene
Benzo(ghi)perylene (1,1,2-benzoperylene)
Dibenzo(a,h)anthracene (1,2,5,6-dibenzanthracene)
Fluorene
Fluoranthene
Indeno(l,2,3-cd)pyrene (2,3-o-Phenylene pyrene)
Phenanthrene
Pyrene
13. Chloroaromatics
7 Chlorobenzene
25 o-Dichlorobenzene
27 p-Dichlorobenzene
26 m-Dichlorobenzene
8 1,2,4-Trichlorobenzene
9 Hexachlorobenzene
Chlorinated Polyaromatic
20 2-Chloronaphthalene
Polychlorinated Biphenyls
106-112 Seven listed
Phthalate Esters
66
67
68
69
70
71
bis(2-Ethylhexyl)
Butylbenzyl, .,
Di-n-butyl
Di-n-octyl
Diethyl
Dimethyl
VII-214
-------
TABLE VII-65.
PRIORITY POLLUTANTS BY CHEMICAL GROUPS
(Continued)
17. Nitroaromatics
56 Nitrobenzene
35 2,4-Dinitrotoluene
36 2,6-Dinitrotoluene
18. Benzidines
5 Benzidine
28 3,3'-Dichlorobenzidine
37 1,2-Diphenylhydrazine
19. Phenols _
65 Phenol
34 2,4-Dimethylphenol
20. Nitrophenols
57 2-Nitrophenol
58 4-Nitrophenol
59 2,4-Dinitrophenol
60 4,6-Dinitro-o-cresol
21. Chlorophenols
24 2-Chlorophenol
22 4-Chloro-m-cresol
31 2,4-Dichlorophenol
21 2,4s,6-Trichlorophenol
64 Pentachlorophenol
22. 144 TCDD (2,3,7,8-Tetrachloro-dibenzo-p-dioxin)
23. Haloaryl Ethers
40 4-Chlorophenylphenyl ether
41 4-Broraophynylphynyl ether
Priority pollutant numbers refer to a published alphabetical listing of the
priority pollutants.
Source: Wise, H.E., and P.O. Fahrenthold (1981). Occurrence and
Predictability of Priority Pollutants in ffastewaters of the Organic
Chemicals and Plastics/Synthetic Fibers Industrial Categories, USEPA 1981.
VII-215
-------
For pollutants regulated in Subcategory Two (non-end-of-pipe biological),
a different methodology was employed to transfer variability factors to
pollutants without individual variability factors. In this case, transfer was
accomplished not by pollutant group, but instead by the in-plant control
technology. Therefore, variability factors were transferred among the
pollutants treated by steam stripping, activated carbon, and in-plant
biological treatment. The Agency further subdivided the pollutants controlled
by steam stripping into high and medium strippability groups (based on Henry's
Law Constants). As discussed previously in this section, Henry's Law Constant
is an important criterion in the design of steam strippers and is therefore an
appropriate factor for the transfer of variability factors. Further sub-
division of the pollutants controlled by in-plant biological treatment was not
considered necessary since all pollutants were determined to be effectively
biodegraded,- transfer of variability factors by adsorpability groups for
pollutants controlled by activated carbon was based on using the variability
factor for 2,4-dinitrophenol (low adsorpability) for the other three pollu-
tants controlled by activated carbon.
For certain pollutants controlled by in-plant biological treatment, the
transferred variability factors for in-plant biological treatment systems are
lower than the variability factors used for end-of-pipe BAT Subcategory One.
This results because BAT Subcategory One variability factors ares 1) in
general, calculated using a different data base; and 2) transferred using the
pollutant variability groups (presented in Table VII-65) rather than across
the technology (as BAT Subcategory Two variability factors are transferred).
Based on these differences, pollutants controlled by in-plant biological
systems which require transferred variability factors will receive variability
factors based on data from three phthalate esters [bis(2-ethylhexyl)
phthalate, di-n-butyl phthalate, and diethylphthalate]: this occurs because
all other pollutants controlled by in-plant biological systems have all daily
data equal to the analytical minimum level. The Agency believes that, in
addition to the reasons mentioned above, the larger end-of-pipe biological
systems have higher variability factors because they receive more commingled
waste streams with a larger number of organic pollutants; thus, they may be
more susceptible to daily fluctuations in performance.
VII-216
-------
Based on the reasons mentioned above, the Agency has decided to retain
the methodology used to transfer in-plant biological system variability
factors. EPA feels that it would be inconsistent to transfer a higher
variability factor to pollutants whose in-plant biological system reduces high
raw waste concentrations (higher than end-of-pipe biological raw waste
concentrations) to the analytical minimum level solely on the basis of
chemical structure. (It should be noted that the transferred end-of-pipe
biological system variability factor for all polynuclear aromatics would be
based on one plant-pollutant combination.)
In response to comments on the statistical aspects of the proposed
limitations development, several statistical techniques were investigated for
deriving limitations. This investigation found that a modification of the
delta-lognormal procedures provides a reasonable approximation of the under-
lying empirical toxic pollutant data. The delta-lognormal distribution
assumes that data are a mixture of positive lognormally distributed values and
zero values. Consequently, zero concentration values are modeled by a point
distribution; positive concentration values follow a lognormal distribution;
and the mixture of these values forms the delta-lognormal distribution. The
statistical methodology used for testing the assumption of lognormality is
found in Appendix VII-E, previously referenced in the BPT Section; the results
of these hypothesis tests are also included in this Appendix.
This method provides a reasonable approach for combining quantitative
concentration values with information expressed only as a nondetect, which is
more qualitative in nature. For the determination of variability factors, the
delta-lognormal procedure was modified by placing the point distribution at
the' analytical minimum level. The details of this modification of the delta
distribution are presented in Appendix VII-F. This approach, is somewhat
conservative since values reported as nondetect may actually be any value
between zero and the minimum level. The detection limit used for each pollu-
tant was the analytical minimum level in EPA analytical methods 1624 and 1625.
Assigning a minimum level to nondetected values in calculating both variability
factors and long-term averages for this data base tends to result in slightly
higher limitations than would be derived if lower values were assumed. If the
point distribution were set to a value below the analytical minimum level,
VII-217
-------
then the variability component of the limitation would increase and the
component corresponding to the mean would decrease. The net effect (mean
times variability factor) would generally result in lower limitations. In the
absence of establishing a firm estimate of the distribution of data below the
analytical minimum levels, the Agency concluded that it would be more
equitable to use the analytical minimum level to model the point distribution
in the modification to the delta-lognormal statistical procedures.
Comments were also received regarding the use of the average variability
factor for transfer to pollutants without individual variability factors for
BAT Subcategory One within each of the 23 pollutant groups. Commenters stated
that the source of data for many of the pollutants was the 3-day Verification
sampling program, and that transfer of an average variability factor to an LTA
based only on data from a 3-day sampling program did not adequately address
the effluent variability of a pollutant. To address this comment, the Agency
examined its edited BAT toxic pollutant data base and determined that the
predominant reason for a pollutant not having an individual variability factor
was not lack of sufficient daily data but that all or nearly all values for
that pollutant were not detected. Therefore, the Agency has decided to retain
the use of an overall average variability factor for each pollutant group to
transfer variability factors to all pollutants within the group without an
individual variability factor.
The Agency also notes the exclusion of two plants (2227P and 500P) from
the variability factor calculations even though they were retained for
calculation of long-term averages. For plant 2227P, EPA examined the
end-of-pipe biological treatment performance data submitted by the plant
(which consisted of data for 1,2,4-trichlorobenzene, 1,2-dichlorobenzene, and
nitrobenzene over a 1-year period) and observed a 2-month period when effluent
concentrations of these pollutants were considerably higher than the remaining
10-month period; during this period of higher effluent concentrations, the
corresponding raw waste concentrations were consistent with the remaining
10 months of raw waste concentration data. Based on this inconsistent
performance, the Agency has concluded that this plant did not have good enough
control of variability to be used to develop variability factors. Thus, the
Agency has excluded this plant from variability factor calculations. However,
VII-218
-------
the overall long-term performance is good and consistent with that achieved by
other, good performers. Therefore, this plant's data has been retained for
long-term average calculations.
For plant 500P, the Agency examined the steam stripping and carbon
adsorption performance data submitted by the plant (which consisted of data
for nitrobenzene over a 3-month period) and believes the data exhibit both
competitive adsorption effects and column breakthrough. Competitive
adsorption exists when a matrix contains adsorbable compounds in solution
which are being selectively adsorbed and desorbed. A review of the data
indicates that while the plant's long-term performance demonstrates
significant removals' of pollutants, it is not consistent, thus much more
variable than that of another plant using similar treatment and achieving
comparable long-term average concentrations. Therefore, the Agency has
excluded this plant from variability factor calculations but has retained the
data for long-term average calculations.
Table VII-66 presents the individual pollutant variability factors for
BAT Subcategory One summarized by pollutant group including the pollutants for
which the overall average variability factor has been transferred. Table
VII-67 presents the individual pollutant variability factors for BAT Sub-
category Two summarized by in-plant control technology and strippability and
adsorpability groups for steam stripping, and activated carbon, respectively.
3. BAT and PSES Metals and Cyanide Limitations
Raw wastewaters generated by certain 6.CPSF facilities contain relatively
high concentrations of.metals and total cyanide. Based on a detailed analysis
(as discussed in Sections V and VI .of this document), the Agency has. decided
to regulate the following six pollutants under BAT and PSES:
• Total chromium
• Total copper , • . .
• Total lead
• Total nickel
• Total zinc
• Total cyanide. ....-.-
VII-219
-------
TABLE VII-66.
INDIVIDUAL TOXIC POLLUTANT VARIABILITY FACTORS
FOR BAT SUBCATEGORY
-------
TABLE VII-66.
INDIVIDUAL TOXIC POLLUTANT VARIABILITY FACTORS
FOR BAT SUBCATEGORY ONE
(Continued)
Pollutant
Number Pollutant Name
Daily VF
Monthly VF
Imputed Varia-
bility Factor?
Pollutant Class = 12
1 Acenaphthene
39 Fluoranthene
55 Naphthalene
72 Benzo(a)Anthracene
73 Benzo(a)Pyrene
74 3,4-Benzofluoranthene
75 Benzo(k)Fluoranthene
76 Chrysene
77 Acenaphthylene
78 Anthracene
80 Fluorene
81 Phenanthrene
84 Pyrene
7 Chlorobenzene
8 1,2,4-Trichlorobenzene
9 Hexachlorobenzene
25 1,2-Dichlorobenzene
26 1,3-Dichlorobenzene
27 1,4-Dichlorobenzene
5.89125
5.89125
5.89125
5.89125
5.89125
5.89125
5.89125
5.89125
5.89125
5.89125
5.89125
5.89125
- 5.89125
Pollutant Class = 13
2.79155
3.25317
2.79155
3.38091
1.74057
2.79155
Pollutant Class = 16
66 Bis-(2-Ethylhexyl) Phthalate 5.91768
68 Di-n-Butyl Phthalate 3.23768
70 Diethyl Phthalate 4.75961
71 Dimethyl Phthalate 4.63833
35 2,4-Dinitrotoluene
36 2,6-Dinitrotoluene
56 Nitrobenzene
Pollutant Class = 17
4.83045
4.83045
4.83045
2.1563
2.1563
2.1563
2.1563
2.1563
2.1563
2.1563
2.1563
2.1563
2.1563
2.1563
2.1563
2.1563
1.46787
1.58318
1.46787
1.59720
1.22323
1.46787
2.17027
1.51824
1.89895
1.86249
1.91724
1.91724
1.91724
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
•Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
VII-221
-------
TABLE VII-66.
INDIVIDUAL TOXIC POLLUTANT VARIABILITY FACTORS
FOR BAT SUBCATEGORY ONE
(Continued)
Pollutant
Number Pollutant Name
34 2,4-Dimethylphenol
65 Phenol
57 2-Nitrophenol
58 4-Nitrophenol
59 2,4-Dinitrophenol
24 2-Chlorophenol
31 2,4-Dichlorophenol
Daily VF
Imputed Varia-
Monthly VF bility Factor?
Pollutant Class =19
3.25650 1.59976
2.49705 1.40602
Pollutant Class = 20
2.49725 1.4643
2.47783 1.4331
2.45842 1.4019
Pollutant Class = 21
9.70575 3.05490
6.37097 2.22674
Yes
VII-222
-------
TABLE VII-67.
INDIVIDUAL TOXIC POLLUTANT VARIABILITY FACTORS
FOR BAT SUBCATEGORY TWO
Pollutant
Number Pollutant Name
Daily VF
Monthly VF
Imputed Varia-
bility Factor?
4
11
13
16
23
29
30
45
85
86
87
88
10
14
44
4.65485
5.88383
5.88383
5.88383
7.36230
5.88383
5.88383
5.88383
8.85657
5.88383
5.88383
2.66160
8.8604
12.2662
15.6720
1.97430
2.18759
2.18759
2.18759
2.49394
2.18759
2.18759
2.18759
2.78458
2.18759
2.18759
1.49754
2.77681
3.02524
3.27366
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Benzene
1,1,1-Trichloroe thane
1,1-Dichloroe thane
Chloroe thane
Chloroform
1,1-Dichloroe thy lene
1,2-Trans-dichloroethylene
Methyl Chloride
Tetrachloroethylene
Toluene
Trichloroethylene
Vinyl Chloride
1,2-Dichloroe thane
1,1,2-Trichloroethane
Methylene Chloride
variability factors when no variability factors available.
Pollutant
Number Pollutant Name
Daily VF
Monthly VF
Imputed Varia-
bility Factor?
56
57
58
59
60
Nitrobezene
2-Nitrophenol
4-Nitrophenol
2 , 4-Dini t rophenol
4 , 6-Dini t ro-o-Cresol
6.7477
11.5023
11.5023
11.5023
11.5023
2.35797
3.23479
3.23479
3.23479
3.23479
Yes
Yes
Yes
Note: Pollutant variability factors-variability factors for pollutant 59
used to impute variability factors for 57, 58, and 60.
VII-223
-------
TABLE VII-67.
INDIVIDUAL TOXIC POLLUTANT VARIABILITY FACTORS
FOR BAT SUBCATEGORY TWO
(Continued)
Pollutant
Number Pollutant Name
1 Acenaphthene
3 Aery lonit rile
34 2,4-Dimethylphenol
39 Fluorantehene
55 Naphthalene
65 Phenol
66 Bis-(2-Ethylhexyl) Phthalate
68 Di-n-Butyl Phthalate
70 Diethyl Phthalate
71 Dimethyl Phthalate
72 Benzo(a)Anthracene
73 Benzo(a)Pyrene
74 3,4-Benzofluoranthene
75 Benzo(k)Fluoranthene
76 Chrysene
77 Acenaphthylene
78 Anthracene
80 Fluorene
81 Phenanthrene
84 Pyrene
Daily VF
—
4.63833
4.63833
4.63833
4.63833
4.63833
4.63833
5.91768
3.23768
4.75961
4.63833
4.63833
4.63833
4.63833
4.63833
4.63833
4.63833
4.63833
4.63833
4.63833
4.63833
1 . —
Monthly VF
1.86249
1.86249
1.86249
1.86249
1.86249
1.86249
2.17027
1.51824
1.89895
1.86249
1.86249
1.86249
1.86249
1.86249
1.86249
1.86249
1.86249
1.86249
1.86249
1.86249
Imputed Varia-
bility Factor?
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
VII-224
-------
The technology basis for control of these pollutants is hydroxide precipita-
tion for the metals and alkaline chlorination for cyanide. Although sulfide
precipitation was the basis for BAT and PSES compliance cost estimates, it was
not used as the technology basis for the limitations because the Agency's
final regulation does not include control of complexed sources of these
metals. This results in a slight overestimation of costs for compliance with
the metals limits for ,BAT and PSES levels of control.
Although the concentrations of these pollutants in certain samples of
untreated OCPSF wastewater are relatively high, the metals fall within the
range of concentrations found in untreated wastewaters from metal processing
and finishing, such as those for the metal finishing and battery manufacturing
industries. Because no metals treatment performance data for OCPSF waste-
waters generated by the validated product/processes listed in Section V were
available, the Agency decided to transfer limitations from the metal finishing
point source category. Cyanide is found at levels in certain OCPSF waste
streams at higher concentrations than in metal finishing. Destruction; of
cyanide by alkaline chlorination is demonstrated in the OCPSF industry; this
technology uses .excess oxidizer (chlorine) and excess alkaline conditions, and
should be able to treat cyanide by adding sufficient detention time which has
been costed. Table VII-68 presents the long-term averages and daily and
monthly maximum variability factors for each pollutant.
The monthly maximum limitations for the metal finishing industry are
based on an assumed monitoring requirement of 10 samples per month and employ
the 99th percentile as a basis for the monthly maximum standard. For the
OCPSF standard, however, the monthly maximum standards are based on an assumed
monitoring requirement of four samples per month and they use the 95 percen-
tile as a basis. The above limitations have been adjusted accordingly to be
consistent with the other OCPSF BAT limitations by deriving 4-day variability
factors from the distributional parameters determined from the 10-day metal
finishing variability factors (see Appendix VII-F). The OCPSF daily and
monthly maximum limitations for each pollutant is the product of the respec-
tive long-term averages and respective 1-day and 4-day variability factors.
VII-225
-------
TABLE VII-68.
u °NE AND TWO LONG-TERM AVERAGES AND
VARIABILITY FACTORS FOR METALS AND TOTAL CYANIDE
Pollutant
Number
119
120
121
122
124
128
Pollutant
Name
Total Chromium
Total Copper
Total Cyanide
Total Lead
Total Nickel
Total Zinc
Long-Term
Average
(mg/1)
0.572
0.815
0.180
0.197
0.942
0.549
Maximum Monthly
Average VF
1.934
1.781
2.343
1.642
1.796
1.912
Maximum Daily
. VF
4.85
4.15
6.68
3.52
4.22
4.75
VII-226
-------
4 BAT Zinc Limitations for Plants Manufacturing Rayon by the Viscose
Process and Acrylic Fibers by the Zinc Chloride/Solvent Process
Raw wastewaters generated by the manufacture of rayon by the viscose
process and acrylic fibers by the'zinc chloride/solvent process exhibit high
concentrations of zinc with levels generally exceeding 100 mg/1. Accordingly,
the Agency has decided to control zinc in the process wastewaters from these
product/processes by establishing separate BAT effluent limitations. Since
these wastewaters do not contain-complexed sources of zinc that could inhibit
treatment by conventional methods, the Agency has selected hydroxide precipi-
tation as the basis for these process-specific BAT effluent limitations.
During the public comment periods on the March 21, 1983, proposal and
July 17, 1985, NOA, industry commenters submitted hydroxide precipitation
performance data for four rayon plants and one acrylic fibers plant. These
data sets contained influent and effluent data for four plants (three rayon,
one acrylic fiber) with over 200 influent/effluent data pairs for each data
set. One rayon plant (399) was eliminated because only effluent data were
submitted. Following a quality assurance review, the effluent concentrations
that exceeded 10 mg/1 or that exhibited less than 90 percent removal of zinc
were deleted from these three data sets. For the performance data from the
acrylic fibers plant (1012), 174-percent of the effluent zinc concentrations
were deleted, while for the three remaining rayon plants (63, 387, and 1774),
5.9, 0.8, and 63.6 percent of the respective effluent zinc concentrations were
deleted.
The Agency then investigated the data set for plant 1774 because of the
failure of 63.6 percent of the data to pass the editing criteria. Analysis of
the data revealed that the majority of the data failed the 90 percent removal
criteria. Further investigation revealed that the failure to achieve 90 per-
cent removal was not because of high effluent zinc concentrations but due to
low influent concentrations that are the result of a zinc recovery unit
upstream of the influent sampling point. Based on these findings, the entire
performance data set for plant 1774 was deleted from further limitations
calculations.
VII-227
-------
The data sets for the two remaining rayon plants and the one acrylic
fibers plant were analyzed according to the methodology for deriving BAT
effluent limitations described in Appendix VII-F. Table VII-69 presents the
resulting long-term averages and variability factors for the remaining plants.
5- PSES Effluent Limitations
As presented earlier in Section VI, the Agency has determined that
47 toxic pollutants pass through POTtfs and will be controlled by PSES effluent
limitations. For these 47 toxic pollutants, PSES effluent limitations are
equal to BAT Subcategory Two effluent limitations.
VII-228
-------
•—•
Plant
Number
63
387
1012
Long-Term
Average
(mg/1)
1.739
2.114
2.190
Maximum Monthly
VF
1.79
1.41
1.52
Maximum Daily
VF
4.19
2.50
2.95
Median of LTA 2.114
Average VF
1.572
3.214
VII-229
-------
7-1
7-2
7-3
7-4
7-5
7-6
7-7
7-8
7-9
7-10
7-11
7-12
SECTION VII
REFERENCES
Cited References
U.S. Environmental Protection Agency (USEPA). 1981 NPDF R0<=t
Management Practices Guidance Document. OfLe'^atefKrcLnt and
U.S. Environmental Protection Airencv
ic Ch- esa ndar
Organic Chemicals and Plastics and Synthetic Fibers Point Qm,
Category Volume I (BPT), Volume II \EAT), and Volume Il'l ?B!T)
Effluent Guidelines Division. EPA 440/1-83/00% February ^983
Carrio, L. A. , A.R. Lopez, P.J. Krashnoff, and J.J. Donnellon
Sludge reduction by in-plant process modification NerYork'
experiences. ' J. Water Poll. Control Fed. 57(2)?il6-121
»
Percy, Rove, and Tchobanoglous , Environmental Engineering, p. 42-43.
1 ' Existing Wastevater Treatment
Series,
sr-
A'D-' G* Tchobanoglous, and T. Asano.
carbon
VII-230
-------
SECTION VII
REFERENCES (Continued)
Cited References (Continued)
7-13 Aire-02 News, Volume 4, No, 3, Summer 1987. Aeration Industries, Inc.,
Chaska, Minnesota.
7-14 Filtration and Chemically Assisted Clarification of Biologically Treated
Pulp and Paper Mill Industry Wastewaters. .Draft Report to the U.S.
Environmental Protection Agency, Edward C. Jordan Co., Inc., 19/y.. ,
7-15 Rice, N., A.A. Kalinske, and W.I. Arnold. 1979. A Pilot Study of;_ -
Advanced Wastewater Treatment for the Ticonderoga Mill, August 1977.
7-16 Personal Communication withsDana Dolloff, International Paper Co., June
26, 1980. ..•.••--'••
7-17 Ambere, H.R., I. Gellman, and R.H. Scott. "The Status of. Water ,-.-,•
Pollufion Control in the Soviet Union," TAPPI, Vol....58, No. 11, November
1975. .-:••••-.-•-
7-18 Scott, R.H. 1978. "Sophisticated Treatment of Baikal Pulp Mill in
USSR," Pulp and Paper, Vol. 48;, No. 4, April 1978.
7-19 Smith, O.D., R.M, Stein,,and C.E. Adams, Jr, ^975. "How Mills Cope _
With Effluent Suspended Solids," Paper Trade Journal, VoK 159, No. I/,
July 1975.
7-20 Peterson, R.R. and J.L. Graham. 1979. CH2M Hill, Inc., Post Biological
Solids Characterization and Removal from Pulp Mill Effluents.
EPA 600/2-79-037.
7-21 "Preliminary Data Base'for Review of BATEA Effluent Limitations
Guidelines, NSPS, and Pretreatment Standards for 'the Pulp, Paper,, and
Paperboard Point Source Category," Edward C. Jordan, Co., Inc., June
1979. . - • .. . , ... . ^ - • : ' ^ ' ," : "'
7-22 Ambers, H.R.. 1979. Crown Zellerbach Corp., Comments on B.C. Jordan's
Draft Report, "Preliminary Data Base for Review of BATEA Effluent
Limitations Guidelines, NSPS, ,and Pretreatment Standards for .the Pulp,
Paper, and Paperboar4 Point Source Category," September 197:9.
7-23 Cashen, R.P. 1979. St. Regis Paper Co., Comments on B.C. Jordan's
Draft Report, "Preliminary Data Base for Review of BATEA Effluent
Limitations Guidelines, NSPS, and Pretreatment Standards for the Pulp,
Paper, and Paperboard Point Source Category," September 1979.
7-24 Button, E.F. 1979. ITT Rayonier .Inc., Comments on E.G. Jordan's Draft
Report. "Preliminary Data Base for Review of BATEA Effluent Limitations
Guidelines, NSPS, and Pretreatment Standards for the Pupp, Paper, and
Paperboard Point Source Category." September 1979.
VII-231
-------
SECTION VII
REFERENCES (Continued)
Cited References (Continued)
7-25 Barton, Curtis A.. 1979. The Proctor & Gamble Co., Comments on B.C.
Jordan's Draft Report, "Preliminary Data Base for Review of BATEA
Effluent Limitations Guidelines, NSPS, and Pretreatment Standards for
tne Pulp, Paper, and Paperboard Point Source Category," September 1979.
7-26 National Council of the Paper Industry for Air and Stream Improvement,
of BiS^Sil ? °', I8""™"7 °f Data-chemically Assisted Clarification
SDA/ £J g S i yTTr!ated Wastewaters,» Presented in Meeting with U.S.
EPA/ Edward C. Jordan Co., Portland, Maine, February 7, 1980.
7-27 Wise, H.E. and P.O. Fahrenthold. 1981. Occurrence and Predictability
2? L /F P°llutan^ :ln tfastewaters of the Organic Chemicals and
2£3 ^y^hetic Fibers Industrial Categories, Presented at the 181st
April"? 198l! Society National Meeting, Atlanta, Georgia, March 29 -
Other References
Beardsley, M.L. and J.M. Coffey. 1985. Bioaugmentation: Optimizing
biological wastewater treatment. Pollution Engineering 17(12): 30.
Berger, B.B. 1983. Control of organic substances in water and
wastewater. EPA-600/8-83-011, NTIS PB86 184744/AS.
TK'S\ 19i!5i'i Contract operations: Private contracts for public
work. J. Water Poll. Control Fed. 57(7):750-755. FUUX.IC
Eckenfelder, Jr., J.J., j. Patoczka, and A.T. Watkin. 1985. Wastewater
treatment. Chem. Eng. 92:60-74. wasiewater
Eischen, G.W. and J.D. Keenan. 1985. Monitoring aerated lagoon
performance. J. Water Poll. Control Fed. 57(8):876-881.
Foess, G.W. and W.A. Ericson. 1980., Toxic control - the trend of the
tuture. Water & Wastes Engineering (February 1980): 21-27.
Gaudy, Jr., Anthony F. and E.T. Gaudy. 1980. Microbiology for
environmental engineers and scientists. New York: McGraw-Hill Book
Company .
Johnson, T., F. Lenzo, and K. Sullivan. 1985. Raising Stripper
Temperature Raises MEK Removal. Pollution Engineering 17(9): 34.
R'S\ 19o5< Waftevater treatment plant instrumentation
NTISPB86-108636/AS?ear Cincinnati, Ohio. EPA/600/8-85-026,
VII-232
-------
SECTION VII
REFERENCES (Continued)
Other References (Continued)
Metcalf & Eddy, Inc. 1979. Wastewater engineering: Treatment/
disposal/reuse. New York: McGraw-Hill Book Company.
Municipal Environmental Research Laboratory (MERL). 1980. Carbon
Adsorption Isotherms for Toxic Organics. EPA-600/8-80-023.
PB 80197320. April 1980.
Peters, R.W., Y/ Ku, and D. Bhattacharyya. Evaluation of recent
treatment techniques for removal of heavy metals from industrial
wastewaters. AIChE Symposium Series. 243:18.
Richards, S.R., C. Hastwell, and M. Davies. 1984. The comparative
examination of 14 activated sludge-plants using enzymatic techniques.
Water Poll. Control Fed, (G^B.) 83:300.
SAIC 1985. Costing documentation and notice of new information
report, Prepared for? Industrial Technology Division of the USEPA.
June 12, 1985.
Sekizawa, T., K, Fujie, H. Kubota, T. Kasakura, and A. Mizuno. 1985.
Air diffuser performance in activated sludge aeration tanks. J. water
Poll. Control Fed. 57(l):53-59.
U.S. Environmental Protection Agency (USEPA). 1974. Processing Design
Manual for Upgrading Existing Wastewater Treatment Plants. PB-259 148,
EPA 625/l-71-004a. October 1974.
U.S. Environmental Protection Agency (USEPA). 1979. Summary Report,
Control and Treatment Technology for the Metal Finishing Industry:
Sulfide Precipitation. EPA 625/8-80-003.
U.S. Environmental Protection Agency (USEPA). 1980. Sources and
Treatment of Wastewater in the Nonferrous Metals Industry. EPA
600/2-80d-074.
U.S. Environmental Protection Agency (USEPA). 1982. Development
Document for Effluent Limitations, Guidelines, and Standards for the
Battery Manufacturing Point Source Category. EPA 440/l-82-067b.
U.S. Environmental Protection Agency (USEPA). ,1983a. Development
Document for Effluent Limitations Guidelines New Source Performance
Standards for the Metal Finishing Source Category.
U.S. Environmental Protection Agency (USEPA). 1983c. Treatability
Manual. Office of Research and Development. EPA-600/2-82-001a.
February 1983.
VII-233
-------
SECTION VII
REFERENCES (Continued)
Other References (Continued)
d
of a Combined
*U,S GOVERNMENT PRINTING OFFICE:1993 -715 -i
VII-234
003/87033
-------