EPA-440/1-78-028-b
SUPPLEMENT TO DEVELOPMENT DOCUMENT
FOR EFFLUENT LIMITATIONS GUIDELINES
EXISTING SOURCES
FOR THE
CORN WET MILLING SUBCATEGORY
to the
GRAIN PROCESSING SEGMENT
OF THE
GRAIN MILLS POINT SOURCE CATEGORY
Douglas M. Costle
Admini strator
Thomas C. Jorling
Assistant Administrator
for Water and Hazardous Materials
Robert B. Schaffer
Director, Effluent Guidelines Division
William Sonnett
Project Officer
January 1978
Effluent Guidelines Division
Office of Water and Hazardous Materials
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
ABSTRACT
This document presents the findings of an extensive study of
the corn wet milling segment of the grain milling industry
for the purpose of developing revised effluent limitations
guidelines for the industry, to implement Sections 301 and
304 of the Federal Water Pollution Control Act Amendments of,
1972 (the "Act"). This study is submitted in response to an
order issued by the United States Court of Appeals for the
Eighth Circuit on December 3, 1976.
Effluent limitations guidelines contained in this document
set forth the degree of effluent reduction attainable
through the application of the best practicable control
technology currently available and the degree of effluent
reduction attainable through the application of the best
available technology economically achievable, which must be
achieved by existing point sources by July 1, 1977 and July
1, 1983, respectively.
Treatment technology is recommended to achieve the 1977
limitations. This technology includes equalization and
biological treatment followed by secondary clarification.
In order to attain the 1983 limitations, additional solids
removal techniques will be required.
Updated costs of achieving the limitations are described in
this report. Supportive data and rationale for development
of the proposed effluent limitations guidelines are also
presented.
iii
-------
TABLE OF CONTENTS
Section Page
I Conclusions 1
II Revised Limitations * 3
III Introduction 7
Background 7
Current Study 9
IV Waste Treatment Results 11
Introduction 11
American Maize 13
Clinton Corn 14
CPC - Corpus Christi 16
CPC - Pekin 17
Summary . .18
V Basis for 1977 Limitations 21
Background 21
Revised Limitations 22
VI Basis for 1983 Limitations 25
Background 25
Revised Limitations 25
VII Modified Starch Study 29
Background 29
Recent Data 29
Effluent Allowance 32
VIII Variability and Excursions 35
Variability 35
Excursions 36
IX Costs 39
Background 39
Current Costs 40
X References 43
iv
-------
FIGURES AND TABLES
FIGURES
Number Page
1 Location of Corn Wet Milling Plants 12
TABLES
1 Summary of Current Corn Wet Milling Discharges 19
in kg/kkg
2 Summary of Current Corn Wet Milling Discharges 20
in Ib/MSBu
3 Costs to Meet Recommended 1977 and 1983 Effluent 42
Limitations
4 Metric Conversion Table 44
v
-------
SECTION I
CONCLUSIONS
An extensive review of available data on raw waste loads and
waste treatment within the corn wet milling industry was
conducted. It was concluded that the technologies
originally recommended for the industry are still
appropriate. For 1977, these technologies include certain
in-plant controls, such as elimination of once-through
cooling water, and end-of-pipe biological treatment. The
recommended 1983 technology includes more stringent in-plant
control and deep bed filtration following biological
treatment.
The data reviewed indicate that the original 1977 and 1983
BOD limitations-are still valid. It was determined that
increasing the suspended solids limitations would better
reflect the situation at existing corn wet milling treatment-
facilities. The suspended solids levels in effluents from
these facilities are generally somewhat higher than the BOD
levels.
Recent information submitted by the industry shows that
increased raw waste loads may result from production of
modified starches. Significant production of modified
starches may affect waste treatment plant performance and
cause increased BOD and suspended solids effluent levels.
For these reasons, it was concluded that an allowance in the
corn wet milling effluent limitations for modified starch
production is warranted.
Variability factors for the effluent limitations were
reviewed. It was concluded that a single-day BOD limit
three times the maximum 30-day limit is appropriate for 1977
and 1983. A single-day suspended solids limit four times
the 30-day limit was determined appropriate for the 1977
limitations. A factor of three is recommended for the 1983
TSS limitation, however, reflecting improved solids control
through deep bed filtration and in-plant measures. It was
also concluded that provisions for excursions beyond the 30-
day and single-day limits are not necessary.
The costs required to meet the recommended effluent
limitations were reviewed and updated to January, 1977
dollar values. Costs for treatment technologies,
replacement of barometric condensers with surface
condensers, and recirculation of barometric cooling water
with cooling towers are summarized in this report.
1
-------
SECTION II
REVISED LIMITATIONS
The effluent limitations for the corn wet milling industry
are' summarized below. The limitations for 1977 and 1983
include basic limits for standard product plants, plus
allowances for plants producing modified starches.
The effluent limitations to be achieved with the best
practicable control technology currently available are as
follows. The basic limitations are:
Effluent Limitations
Average of Daily
Effluent Maximum for Any Values for 30 Consecutive
Characteristic One Day Days Shall Not Exceed
Metric Units (kilograms per 1,000
kg of corn)
BOD5 2.67 0.89
TSS 4.32 1.08
pH within the range 6.0 to 9.0
English Units (pounds per 1,000
std bushels of corn)
BOD5 150 50
TSS 240 60
pH within the range 6.0 to 9.0
For those plants producing modified starches at a rate of at
least 15 percent by dry-basis weight of total sweetener and
starch products per month for 12 consecutive months, the
. following limitations should be used to derive an additive
adjustment to the above basic limitations:
Effluent Limitations
Average of Daily
Effluent Maximum for Any Values for 30 Consecutive
Characteristic One Day Days Shall Not Exceed
Metric Units (kilograms per 1,000
kg of corn)
BOD5 0.81 0.27
3
-------
TSS
2.16
0.54
English Units (pounds per 1,000
std bushels of corn)
BOD5 45 15
, TSS 120 30
The effluent limitations to be achieved with the best
available control technology economically achievable are as
follows. The basic limitations are;
Effluent Limitations
Average of Daily
Effluent Maximum for Any Values for 30 Consecutive
Characteristic . One Day Days Shall Not Exceed
Metric Units (kilograms per 1,000
kg of corn) ^
BODS 1.08 0.36
TSS~ 1.62 0.54
within the range 6.0 to 9.0
English Units (pounds per 1,000
std bushels of corn)
BOD5 60 20
TSS~ 90 30
pH within the range 6.0 to 9.0
For plants producing modified starches at a rate of at least
15 percent by dry-basis weight of total sweetener and starch
products per month for 12 consecutive months, the following
limitations should be used to derive an additive adjustment
to the above basic 1983 limitations:
Effluent Limitations
Average of Daily
Effluent Maximum for Any Values for 30 Consecutive
Characteristic One Day Days Shall Not Exceed
Metric Units (kilograms per 1,000
kg of corn)
BOD5 0.43 0.14
TSS~* 0.66 0.22
4
-------
English Units (pounds per 1,000
std bushels of corn)
BOD 5 24 8
TSS~" 36 12
5
-------
SECTION III
INTRODUCTION
BACKGROUND
EPA1s original effluent guidelines study of the corn wet
milling industry was conducted from December, 1972 to June,
1973. A contractor-prepared Draft Development Document was
completed and distributed at that time. After receiving and
reviewing comments from the industry and other interested
parties, EPA published a notice of proposed rulemaking for
the industry in December, 1973. A revised Draft Development
Document and a Draft Economic Report were also issued. Final
effluent limitations and new source standards were
promulgated on March 20, 1974.
In December, 1974, certain members of the corn wet milling
industry filed petitions for review in the U.S. Court of
Appeals for the Eighth Circuit, contesting the final
regulations. In a decision rendered in May, 1975, the Court
remanded the new source standards and claimed no
jurisdiction over existing source (1977 and 1983)
regulations, CPC International, Inc. v. Train, 515 F 2d 1032
(May 5, 1975) (CPCI) .
During the period May to July, 1975, EPA conducted a remand
study to reassess the new source performance standards for
the industry. Of particular concern was the technology of
deep bed filtration as applied to the effluent from corn wet
milling waste treatment facilities. EPA published a
Supplement to the Development Document in August, 1975, and,
after receiving and evaluating comments, issued a notice of
its decision not to revise the new source performance
standards. This notice was published in November, 1975, and
was based on EPA's finding that filtration could be
successfully applied in corn wet milling waste treatment.
At the same time, certain members of the corn wet milling
industry filed suit in the U.S. District Court for the
Southern District of Iowa to contest the 1977 and 1983
regulations for existing plants. The suit was filed in
June, 1975; arguments were heard in September; and a
decision was issued by the Court in January, 1976. This
decision remanded the existing source regulations back to
EPA, primarily because ranges and factors were not
identified.
7
-------
Trie industry was also not satisfied with EPA* s expanded new
source record and its decision not to change the new source
standards. Therefore, in December, 1975, industry members
challenged the decision by EPA's Administrator in the Eighth
Circuit. Arguments were heard in March, 1976, and a
decision was reached in August, 1976. CPC International,
Inc. v. Train. 540 F 2d 1329 (August 18, 1976) (CPCII) . The
Court upheld EPA*s BOD limit for new sources (0.357 kg/kkg
or 20 lb/MSBu* for a maximum 30-day average) but suggested
that the suspended solids limit of 0.179 kg/kkg (10 lb/MSBu)
be increased to 0.447 kg/kkg (25 lb/MSBu). EPA accepted the
Court*s findings and issued revised new source standards in
November, 1976.
In July, 1976, EPA appealed the District Court decision on
existing source standards to the Eighth Circuit Court 'of
Appeals. Arguments were heard in October, and an order was
issued from the Court in December, 1976. The Court agreed
to postpone a decision until after the Supreme Court's
ruling in the duPont case.** In the interim, however, the
Court instructed the EPA to gather new and updated- data
relating to the 1977 and 1983 effluent limitations.
The Supreme Court issued a decision in the duPont case on
February 23, 1977. The court concluded that EPA has
authority under Sections 301 and 304 of the Federal Water
Pollution Control Act Amendments of 1972 (the "Act") to
limit discharges through industry-wide regulations
establishing uniform effluent limitations for 1977 and 1983.
The court also determined that review of both existing plant
and new source effluent regulations should be held
exclusively in the United States Courts of Appeal.
~The effluent limitations are expressed in terms of quantity
. of pollutant per unit of raw material processed. The units
are kilograms (kg) of pollutant per thousand kilograms (kkg)
of corn processed and pounds (lb) of pollutant per thousand
standard bushels (MSBu) of corn processed. One standard
bushel equals 56 lb or 25.4 kg. To convert lb/MSBu to
kg/kkg, a factor of 0.01787 (0.454 kg/lb ~ 25.4 kg/SBu) is
used.
~*E. I. duPont de Nemours S Co. v. Train, S. Ct. Nos. 75-978
and 75-1473, and Train V. E.I. duPont de Nemours 5 Co., S.
Ct. No. 75-1075. This case was an appeal of a Fourth
Circuit Court of Appeals decision regarding effluent
regulations for the inorganic chemicals industry.
8
-------
CURE EN T" STUDY
In response to the Eighth Circuit's order of December, 1976,
EPA made a request for new data to the industry on January
26, 1977. The industry submitted information on February 28
that included proposals for revised effluent limitations
specifying ranges of numbers and allowances for factors such
as barometric condensers, wet scrubbers, and old plants.
Also included in the submittal were testimonies from the CPC
International - Corpus Christi NPDES adjudicatory hearings,
data on wet scrubber waste loads, and data on product mix
and raw waste loads at CPC«s Pekin and Corpus Christi
plants.
Cost information was submitted by the industry on March 22,
1977. Included were data on industry revenues (prices of
various products) and costs of waste treatment at Hubinger*s
Keokuk plant and CPC's Corpus Christi and «Pekin plants.
Also included were two additional testimonies from the NPDES
permit proceedings for CPC - Corpus Chris-ti.
Information was submitted by the industry on May 2 regarding
the following: waste treatment performance data for CPC's
Corpus Christi and Pekin plants, a summary of effluent
results and variability at CPC - Corpus Christi, and raw
waste load data for Hubinger. In addition, data on raw
waste loads from modified starch production were submitted
on June 3, 1977.
EPA and its contractor visited the American Maize corn wet
mill at Hammond, Indiana, and the CPC corn wet mill at
Pekin, Illinois. The visits were made in May, 1977. Both
of these mills have waste water treatment facilities and
direct discharges to navigable waterways.
Current waste load data were received for several mills,
including American Maize, Clinton Corn, CPC - Corpus
. Christi, and CPC - Pekin. These data were computerized and
evaluated in terms of pollutant discharge (BOD and TSS) per
unit of raw material processed.
9
-------
SECTION IV
WASTE TREATMENT RESULTS
INTRODUCTION
At the time of the original effluent guidelines study for
the corn wet milling industry in 1973, there were 12
companies operating 17 mills in this country. Only four of
these mills have direct discharges of concentrated process
waste waters following on-site treatment. These four mills
are American Maize at Hammond, Indiana; Clinton Corn at
Clinton, Iowa; CPC at Corpus Christi, Texas; and CPC at
Pekin, Illinois. The current situation at each of these
four plants is described in detail below.
Of the other 13 mills operating in 1973, six plants pretreat
their wastes before discharge to municipal treatment
systems, six mills discharge untreated wastes to municipal
systems, and one mill uses land treatment to dispose of its
wastes. Several mills have direct discharges of untreated,
contaminated cooling water to surface waters.
Five mills have been constructed or are under construction
since 1973. One new mill has complete waste treatment
facilities and direct discharge of the treated effluent.
The other four mills will pretreat their wastes before
discharge to municipal systems. Figure 1 shows the
locations of corn wet milling plants in the United States.
The effluent data reviewed generally reflect long-term
performance of treatment plants within the industry. In
most cases, all of the available data were analyzed. Only
in cases where extreme values or anomalies were noted were
data points excluded. For example, treated effluent levels
at one plant were extremely higher than normal during a
labor strike, and these data were omitted. In another case,
a huge spill of corn syrup to a mill*s waste treatment plant
occurred, causing an upset that lasted several weeks.
Because this occurrence was deemed preventable, the
treatment data for that period were not included in the
analysis.
A note should be made concerning use of once-through
barometric condensers for cooling. Several mills use such
systems for cooling in steepwater and syrup evaporation
processes. These systems result in the discharge of large
quantities of mildly contaminated cooling water. Use of
surface condensers rather than barometric condensers, or
11
-------
MC
"wash
MONT
IDAHO
N. OAK
OAK
HE BP-
NEV
H*
to
UTAH
COLO
kCAUF
ARIZ
H MM
t«K
O NEW PLANTS
# PLANTS
OPERATING
IN 1973
rioune i
LOCATION OP CORN WET MILLING PLANTS
-------
cooling towers to recirculate the cooling water, results in
a concentrated waste stream that can be sent to a mill's
waste treatment system. This waste stream, although
concentrated, is readily treatable in a biological system.
Conversion to surface condensers or cooling towers would
increase the waste load to be treated at a mill and would
necessitate increased treatment plant capacity, but would
not create treatment problems or increased effluent
concentrations.
AMERICAN MAIZE
The American Maize corn wet mill at Hammond, Indiana, has
on-site facilities for treatment and discharge of process
wastes. Production at the mill has increased by about 50
percent since 1973, and all of this increase has been in the
manufacture of modified starches. Three main waste streams
are sent to the mill's treatment facility: (a) steepwater
condensates, (b) starch filtrates, and (c) carbon filter
sluice water. Treatment consists of activated sludge
(aeration and clarification) followed by three polishing
lagoons. The lagoon effluent is combined with spent cooling
water from the mill and discharged to Lake Michigan.
American Maize has recently added a chemical coagulation
treatment step for polishing the lagoon effluent. Alum and
polymer are added, followed by settling in a clarifier.
Laboratory tests have indicated that BOD and TSS levels of
20 mg/1 can be expected after chemical treatment.(1)
Since 1973, American Maize has eliminated use of its sludge
lagoon, enlarged its aeration basin, and constructed a
larger secondary clarifier. Waste treatment BOD and TSS
effluent levels are summarized below in terms of kg/kkg and
lb/MSBu. Current treated effluent flow is now about 2,650
cu m/day (0.7 mgd) .
AMERICAN MAIZE TREATMENT PLANT
kq/kkq lb/MSBu kq/kkq lb/MSBu
BOD TSS
1972:
average
0.04
2.1
0.19
10.5
maximum
month
0.06
3.3
0.37
20.8
maximum
day
0.11
6.0
1.23
69.0
1975-77:
average
—
—
0.12
6.6
maximum
month
-
-
0.36
20.4
maximum
day
-
-
0.96
54.0
13
-------
Total discharge from the American Maize corn vet mill
includes spent cooling water- This water is primarily from
surface condensers, although the plant still has three
barometric condensers in the syrup refinery. Replacement of
these units is planned before 1982. The cooling water flow
of about 41,635 cu m/day (11 mgd) is combined with treated
process waste water before discharge. BOD and TSS levels in
the mill's total discharge are summarized below:
AMERICAN MAIZE TOTAL PLANT*
BOD TSS
kq/kkq lb/MSBu kq/kkq lb/MSBu
1972:
average
0.25
14.1
0.19
10.8
maximum
month
0.41
22.8
0.39
21.8
maximum
day
3.40
190.0
1.17
65.4
1975-77:
average
-
0.31
17.5
0.51
28.7
maximum
month
0.64
35.6
0.83
46.5
maximum
day
3.66
204.7
3.18
178.0
The American Maize NPDES permit sets limits on BOD and TSS
discharges in terms of lb/day. Using current estimated
production figures for the mill, the NPDES permit levels
convert to the following kg/kkg and lb/MSBu figures:
AMERICAN MAIZE NPDES LIMITS
Maximum
30-Day
kq/kkq lb/MSBu
kq/kkq
Maximum
Day
lb/MSBu
BOD
TSS
0.40
0.37
22.2
20.7
1.19
1.11
66.7
62.0
CLINTON CORN
The Clinton Corn Processing Company corn wet mill has on-
site biological treatment facilities for process waste
waters. Treatment includes limited equalization, trickling
filtration, aeration, secondary clarification, and deep bed
filtration. Treated effluent is recycled back to the mill's
raw water supply. The treatment plant began operating in
November, 1974, and operations continued until December,
*TSS values for 1975-1977 are gross values; figures for 1972
are net values. The plant*s NPDES permit levels are based
on net discharge. Maximum day BOD and TSS values reflect
apparent upsets; normal daily maximums are 1.39 kg/kkg (78
lb/MSBu) BOD and 1.79 kg/kkg (100 lb/MSBu)
14
-------
1975, when an explosion in the mill destroyed the plant's
waste water collection and equalization basin. Treatment
operations resumed in late 1976. Treatment results for the
two periods of operation, in terms of kg/kkg and lb/MSBu BOD
and TSS, are summarized below. Average effluent flow is
9,084 cu m/day {2.4 mgd) .
CLINTON CORN TREATMENT PLANT
BOD TSS
ka/kkg
lb/MSBu
kq/kka
lb/MSBu
1975:
average
0.11
6.0
0.30
17.0
maximum
month
0.19
10.8
0.55
30.5
maximum
day
1.19
66.7
3.12
174.5
1976-77:
average
0.26
' 14.8
0.34
19.3
maximum
month
0.62
34.5
0.77
43.1
maximum
day
3.23
180.5
3.46
193.5
Clinton Corn discharges spent cooling water and certain
other waste waters to the Mississippi River. Surface
condensers have replaced barometric condensers for
steepwater evaporation, but barometric condensers are still
used in other areas of the mill. Limited data are available
on total discharges from the mill and are summarized below
in terms of kg/kkg and lb/MSBu BOD and TSS:
CLINTON CORN TOTAL PLANT
BOD TSS
kg/kkg lb/MSBu kg/kkg lb/MSBu
Oct.—Dec.* 1975: average 0.81 45.4 0.36 20.4
maximum month 1.20 64.4 0.43 23.9
Nov., 1976-
Apr., 1977: average 2.12 118.7 0.76 42.8
maximum month 3.63 202.9 1.26 70.7
Clinton Corn's NPDES permit limits are based on EPA*s
original effluent limitations promulgated in 1974 (0.89
kg/kkg or 50 lb/MSBu maximum 30day values for BOD and TSS,
with maximum daily values of 2.68 kg/kkg or 150 lb/MSBu).
15
-------
CPC - CORPUS CHRISTI
CPCfs Corpus Christi corn wet mill treats its process waste
waters prior to discharge into the Corpus Christi Ship
Channel. Treatment includes tilted plate separation of
starch wastes, cooling and equalization of all wastes,
aeration, and secondary clarification. A polishing lagoon
and final clarifier have recently been added after the
secondary clarifier. The waste treatment plant has rarely
operated satisfactorily for any length of time and is
subject to frequent upsets. During the period 1975 to 1976,
for example, average effluent BOD was 141 mg/1 and average
effluent TSS was 498 mg/1. Average effluent flow from the
waste treatment plant was 1,136 cu m/day (0.3 mgd) .
Treatment plant results for the period in terms of kg/kkg
and lb/MSBu are summarized below:(2)
CPC- CORPUS CHRISTI TREATMENT PLANT
BOD TSS
kq/kkq lb/MSBu kq/kkq lb/MSBu
1975 - 1976: average 0.36 20.1 1.27 70.9
maximum month 2.13 119.2 3.49 195.1
maximum day 5.21 291.4 16.29 911.5
The above results do not reflect the polishing lagoon and
clarifier that were recently added.
The Corpus Christi mill also has a separate discharge that
includes spent cooling water, boiler blowdown, cooling tower
blowdown, and storm runoff. Surface condensers are
primarily used in the plant, but there are two barometric
condensers in the syrup refinery. Net BOD discharge in the
cooling water stream averaged 7 mg/1 or 84 kg/day (186
lb/day) in 1975. The maximum monthly value was 17 mg/1 (264
kg/day or 582 lb/day), and the maximum daily value was 52
mg/1 (887 kg/day or 1,909 lb/day). These discharges convert
to the following kg/kkg and lb/MSBu figures:(3)
CPC - CORPUS CHRISTI COOLING WATER DISCHARGE
BOD
kq/kkq lb/MSBu
1975: average 0.20 11.3
maximum month 0.63 35.3
maximum day 2.07 115.7
16
-------
CPC - PEKIN
CPC*s Pekin, Illinois, corn wet mill has operated waste
treatment facilities for concentrated process wastes since
1970. The present treatment scheme includes equalization,
cooling, activated sludge (aeration and clarification)*
chemical settling after alum addition, and dissolved air
flotation. Treated effluent is combined with spent cooling'
water and discharged to the Illinois River. The present
volume of concentrated waste water treated is about 2,270 cu
m/day (0.6 mgd). This volume is considerably less than the
flow of 3,030 cu m/day (0.8 mgd) being treated in 1972, and
reflects significant waste reduction through recent in-plant
controls.(4,5) Current treatment plant data, reflecting
biological treatment followed by chemical clarification and
dissolved air flotation, show an average effluent BOD of 6
mg/1 and an average TSS level of 35 mg/1. Treatment plant
effluent data, in kg/kkg and lb/MSBu, are summarized below
for several periods of operation:(2)
CPC - PEKIN TREATMENT PLANT
BOD TSS
kq/kkcr
lb/MSBu
kq/kkq
lb/MSBu
Jan.-June, 1972:
average
0.48
27.0
0.62
34.7
maximum
month
0.73
41.1
1.15
64.2
maximum
day
3.16
176.7
3.51
196.2
Jul.1972-Jan.1973:
average
0.14
7.6
0.24
13.7
maximum
month
0.28
15.9
0.78
43.6
maximum
day
1.85
103.4
2.31
129.1
Jun.1973-Jun.1974:
average
0.15
8.5
0.29
16.1
maximum
month
0.34
19.3
0.61
34.0
maximum
day
1.58
88.3
3.88
217.2
Sep.1976-Mar.1977:
average
0.007
0.4
0.04
2.4
maximum
month
0.01
0.6
0.09
5.0
maximum
day
0.10
5.4
0.85
47.8
The Pekin mill makes extensive use of barometric condensers
in the steepwater evaporation and syrup refining operations.
About 87,000 cu m/day (23 mgd) of spent, contaminated
cooling water are discharged. Recent data indicate an
average BOD concentration of 34 mg/1 in the cooling water,
with a maximum monthly value of 62 mg/1 and maximum daily
value of 231 mg/1. Discharge quantities of BOD in kg/kkg
and lb/MSBu are summarized below for the Pekin mill's
cooling water discharge:
17
-------
CPC - PEKIN COOLING WATER
BOD
kq/kkq lb/MSBu
Sept., 1976 - Mar.,
Mar.r 1977
average
maximum month
maximum day
I.70
2.82
II.85
95.2
157.6
663
The NPDES permit for CPC-Pekin sets BOD and TSS
concentration limits for treatment plant effluent and
cooling water discharge. The BOD limits for both streams
are 20 mg/1 daily average (30-day average) and 50 mg/1 daily
maximum. The TSS limits for the treatment plant effluent
are .25 mg/1 daily average and 62-5 mg/1 daily maximum. An
average net TSS increase of 5 mg/1 over intake water levels
is allowed in the cooling water discharge.
Tables 1 and 2 summarize present effluent discharge levels
at the four corn wet mills discussed above. The data in
Table 1 are presented in kg/kkg; lb/MSBu figures are used in
Table 2.
SUMMARY
18
-------
TABLE 1
SUMMARY
OF CURRENT CORN WET MILLING DISCHARGES
IN KG/KKG
BOD
TSS
Max 30-day
Max Max 30-day
Max
Value
Value
M
nerican Maize
WTP
0.05-0.07
o.n
0.38
0.96
(Waste Treatment Plant)
Total Plant
0.41-0.64
1.43
0.39-0.82*
1.79*
NPDES Allowance
0.40
1.19
0.37
1.11
linton Corn
WTP
0.20
1.20 (0.61)**
0.55
3.13 (1.79)
Total Plant
1.20-3.63
-
0.43-1.27
-
NPDES Allowance
0.89
2.68
0.89
2.68
PC-Corpus Christi
WTP
2.13 (0.82)
5.21 (2.81)
3.49 (2.32)"
16.29 (5.52)
Cooling Water
0.63
2.07
Negligible
Total Plant
2.76 (1.45)
7.27 (4.88)
3.49 (2.32)
16.29 (5.52)
PC-Pekin
WTP (1976-77)
0.01
0.10
0.09
0.86
WTP (1973-74)
0.34
1.57-1.79
0.61
. 2.32-3.57
Cooling Water
2.82
11.85
5 mg/1 increase over raw-
(1976-77)
water***
Total Plant
2.82
11.94
* Gross discharge; permit limits are based on net TSS discharged.
** Numbers in parentheses are adjusted values; adjustment made by
eliminating values during preventable upsets or anomalies.
*** A 5 mg/1 TSS increase at CPC - Pekin represents about 0.21 kg/kkg.
19
-------
TABLE 2
SUMMARY OF
CURRENT CORN WET MILLING DISCHARGES
IN LB/MSBU
BOD
TSS
Max 30-day
Value
Max
Day
Max 30-day
Value
Max
nerican Maize
KTP
Total Plant
NPDES Allowance
3-4
23-36
22.2
6
80
66.7
21
22-46*
20.7
54
100*
62
linton Corn
WTP
Total Plant
NPDES Allowance
11
67-203
50
67 (34)**
150
31
24-71
50
175 (100)
150
PC-Corpus Christi
WTP
Cooling Water
Total Plant
119.2 (46)
154*.5 (81.3)
291.4 (157)
115.7
407 (273)
195.1 (130)
Negligible
195.1 (130)
911.5 (309)
911.5 (309)
PC-Pekin
WTP (1976-77)
WTP (1973-74)
Cooling Water
(1976-77)
Total Plant
0.6
19
157.6
158
5 4
88-100'
663
668
5.0
34
5 nig/1 increase
water***
48
130-200
over raw
* Gross discharge; permit limits are based on net TSS discharged.
** Numbers in parentheses are adjusted values; adjustment made by
j eliminating values during preventable upsets or anomalies.
*** A 5 mg/1 TSS increase at CPC - Pekin represents about 12 lb/MSBu.
20
-------
SECTION V
BASIS FOR 1977 LIMITATIONS
BACKGROUND
The' original effluent guidelines study for the corn wet
milling industry defined the best practicable control
technology currently available (BPT). This technology,
which is the basis for the recommended 1977 effluent
limitations, includes the following (as discussed in the
Development Document, pages 108-109}: (6)
1. Isolating and collecting the major waste streams
for treatment.
2. Eliminating once-through barometric cooling waters,
especially from the steepwater and syrup evaporators. This
change can be accomplished by recirculating these cooling
waters over cooling towers or replacing the barometric
condensers with surface condensers.
3. Isolating once-through noncontact (uncontaminated)
cooling waters for discharge directly to the receiving
waters or provision of recirculating cooling tower systems
with the blowdown directed to the treatment plant.
4. Diking of all process areas subject to frequent
spills in order to retain lost product for possible reuse or
by-product recovery.
5. Installing and maintaining modern entrainment
separators in steepwater and syrup evaporators.
6. Monitoring the major waste streams to identify and
control sources of heavy product losses.
7. For the resulting process waste waters, providing
extensive waste treatment consisting of: flow and quality
equalization, neutralization, biological treatment, and
solids separation. The biological treatment methods
available include activated sludge, pure oxygen activated
sludge, bio-discs, and possible combinations of other
biological systems.
The original 1977 effluent limitations for the corn wet
milling subcategory were as follows:
21
-------
Effluent
Characteristic
Maximum for Any
One Day
BOD 5
TSS~
pH
kq/kkq lb/MSBu
2.68 150
2.68 150
within the range
Average of Daily Values for
30 Consecutive Days Shall Not Exceed
kq/kkq lb/MSBu
0.89 50
0.89 50
6.0 to 9.0
The current study of the industry has shown that the
technology forming the basis for the original 1977 effluent
limitations is still valid. The revised limitations
discussed below are based on the technology outlined above.
REVISEP LIMITATIONS
The revised 1977 effluent limitations for corn wet mills are
as follows:
Effluent
Characteristic
BOD5
TSS
pH ,
Maximum for Any
One Day
kq/kkq
2.67
4.32
lb/MSBu
150
240
Average of Daily
Values for 30 Consecutive
Days Shall.Not Exceed,
kq/kkq lb/MSBU
0.89
1.08
50
60
within the range 6.0 to 9.0
The BOD limits are the same as the limits originally promul-
gated in 1974. The 30-day TSS limit has been increased
somewhat to reflect data from existing treatment plants
indicating normal TSS:BOD ratios greater than 1.0. The
factor used to establish the single-day TSS limit has been
increased to four, reflecting the fact that effluent TSS
levels may vary more than BOD levels.
The limits " are supported by data from existing corn wet
milling treatment plants, as discussed in Section IV.
Maximum monthly effluent values for three mills are
summarized below:
Plant
American Maize (total discharge)
Clinton Corn (waste treatment plant)
CPC - Pekin (waste treatment plant)
BOD
kq/kkq lb/MSBu
TSS
kq/kkq lb/MSBu
0.64 35.6 0.83 46.5
0.62 34.5 0.77 43.1
0.28-0.34 15.9-19.3 0.61-0.78 34-43.6
The American Maize data are for the year 197 2 and reflect
activated sludge and lagoon treatment. The total discharge
includes some once- through barometric condenser water. The
Clinton Corn figures are from 1975 and show effluent levels
22
-------
after biological treatment and filtration. Equalization of
the raw wastes was limited, however, and the filters were
bypassed on occasion. Effluent levels for CPC - Pekin
reflect treatment plant operation during the period 1972 to
1974, when treatment included activated sludge and dissolved
air flotation. American Maize and Clinton are presently
meeting NPDES requirements that are more stringent than the
recommended effluent limitations- It should be noted that
none of the above three mills employs all of the recommended
1977 technology. All three mills discharge contaminated
cooling water from barometric condensers without treatment.
In fact, all of the evaporators at CPC's Pekin plant use
once-through barometric condensers. American Maize and
Clinton Corn have made significant replacements of
barometric condensers with surface condensers.
The 0.89 kg/kkg (50 lb/MSBu) BOD limit is supported by data
from three existing corn wet mills. Support for the limit
was also given by the Eighth Circuit Court of Appeals in its
decisions in CPC I and CPC II. In CPC I the Court
determined that the 1977 technology, if installed in a new.
plant, would enable the facility to meet the 1977 effluent
limitations.(7) In CPC II the Court reaffirmed its
position, and also supported a new source BOD standard of
0.36 kg/kkg (20 lb/MSBu), based on 1983 technology. (8) The
1983 technology basically consists of deep bed filtration
following biological treatment.
The 1977 TSS limit is to be increased because the evidence
shows that effluent TSS levels are usually somewhat higher
than BOD levels. A TSS:BOD ratio of 1.2 is approximated in
the effluent data for American Maize and Clinton Corn. For
example, TSS levels in the Clinton Corn treatment effluent
are generally about 0.09 kg/kkg (5 lb/MSBu) higher than the
BOD levels. A similar situation exists at American Maize,
although the polishing lagoons tend to generate additional
suspended solids because of algae growth. Lagoons are not
part of EPA* s recommended 1977 technology.
To set single-day effluent limits, factors of three for BOD
and four for TSS were applied to the 30-day values. The TSS
factor was increased to accomodate the higher variation in
effluent TSS values experienced at several corn wet mills.
These factors find extensive support in data from existing
treatment plants:
23
-------
Ratio of Single-day Value to
30-day Value
Plant
BOD
TSS
American Maize
Clinton Corn
CPC - Corpus Christi
CPC - Pekin (treatment plant)
CPC - Pekin (total plant)
1.8
3.1
2.4
4.6
2.6-3.1
2.6-3.3
3.3
4.7
3.0-6.4
2.5-3.6
The above factors are ratios of single-day maximum values to
30-day maximum values. The factors are based on the
effluent data presented in Section IV. It should be noted
that not all of the above treatment operations represent
stable operation and maximum efficiency. A discussion of
the shortcomings of the various treatment plants in the
industry can be found in the 1975 Supplement to the
Development Document.(9)
24
-------
SECTION VI
BASIS FOR 1983 LIMITATIONS
BACKGROUND
For corn wet milling plants, the best available technology
economically achievable (BAT) was defined during the
original effluent guidelines study. The main element of
this technology is improved solids separation following
biological treatment. Improved solids separation is best
represented by deep bed filtration, although other means are
available and are in use for solids removal after biological
treatment. Other components of BAT for corn wet mills
include:(6)
1. Isolation and treatment of all process waste
waters. No process wastes should be discharged without
treatment.
2. Institution of maximum water reuse at all plants
over and above the current levels of practice.
3. Provisions for improved solids recovery at
individual waste sources. >
The original 1983 effluent limitations
milling subcategory were as follows:
: Effluent
Characteristic
B0D5
TSS
PH
Maximum for Any
One Day
kg/)ckq
1.07
0.54
lb/MSBu
60
30
for the corn wet
Average of Daily
Values for 30 Consecutive
Days Shall Not Exceed
kg/kkq lb/MSBu
0.36
0.18
within the range 6.0 to 9.0
The current study of the industry has shown that the
technology forming the basis for the original 1983 effluent
limitations is still valid- The revised limitations
discussed below are based on the technology outlined above.
20
10
REVISED LIMITATIONS
The revised 1983 effluent limitations are as follows:
Average of Daily
Effluent Maximum for Any Values for 30 Consecutive
Characteristic One Day Days Shall Not Exceed
25
-------
kq/kkq
lb/MSBu
kg/kkg
lb/MSBu
B0D5
TSS
1.08
1.62
60
90
0.36
0. 54
20
30
pH
within the range 6.0 to 9.0
The' BOD limit for 1983 is identical to the original limita-
tion promulgated in 1974. The TSS limit has been increased
from 0.18 kg/kkg (10 lb/MSBu) to 0.54 kg/kkg (30 lb/MSBu) to
reflect recent data on waste treatment performance within
the industry. These data have shown that effluent TSS
levels are generally somewhat higher than BOD levels.
Deep bed filtration, the recommended 1983 technology, is
currently in use at one corn wet milling plant, Clinton Corn
in Clinton, Iowa. (Clinton has also installed filters as
part of the waste treatment facilities at its new plant in
Montezuma, New York.) The applicability of deep bed
filtration to corn wet milling waste waters was at issue in
CPC II. EPA made an extensive analysis of filtration as
part of the new source remand study conducted in 1975.(9)
It was concluded that filtration was applicable to the
industry. In its decision in CPC II, the Eighth Circuit
upheld the applicability of deep bed filtration, although it
recommended that the new source TSS standard be increased
from 0.18 kg/kkg (10 lb/MSBu) to 0.45 kg/kkg (25 lb/MSBu).
The Court based its recommendation on treatment results from
Clinton Corn. (8)
Although filtration is the recommended step to be added to
biological treatment, there are other available and
demonstrated technologies that can be used to improve
effluents from corn wet milling treatment plants- Chemical
coagulation with alum and polymers has been successfully
demonstrated at CPC* s Pekin plant. The chemical coagulation
step follows a conventional activated sludge treatment
system. Chemical coagulation has also been successfully
tested at the American Maize plant in Hammond, Indiana. A
full-scale system has recently been installed to polish the
effluent from American Maize's activated sludge and lagoon
system- Pilot tests demonstrated that BOD and TSS levels of
20 mg/1 can be achieved with chemical coagulation.
Dissolved air flotation is also an available effluent
polishing technology that is in use at CPC*s Pekin plant
following the chemical coagulation step- The present
treatment system at CPC - Pekin is producing an effluent
with BOD levels less than 10 mg/1 and TSS levels generally
less than 40 mg/1- A third available technology is effluent
polishing with lagoons. Lagoons have been successfully used
at the American Maize plant for several years following
26
-------
activated sludge treatment, and CPC«s Corpus Christi plant
has recently installed a secondary lagoon and clarifier
following its biological treatment plant.
The 1983 BOD limit of 0.36 kg/kkg <20 lb/MSBu) is supported
by data from several corn wet milling plants. The American
Maize treatment system (activated sludge and lagoons)
achieved a maximum 30-day BOD level of 0.06 kg/kkg (3.3,
lb/MSBu). Tests with chemical coagulation indicate that BOD
values will be even lower after treatment with alum and
polymer. The Clinton Corn treatment facility (activated
sludge and filtration) produced an effluent with an average
BOD load of 0.11 kg/kkg (6 lb/MSBu) and maximum monthly
value of 0.19 kg/kkg (10.8 lb/MSBu). Recent data from CPC*s
Pekin plant indicate a 30-day BOD level of 0.01 kg/kkg (0.6
lb/MSBu) after biological treatment, chemical coagulation,
and dissolve^ air flotation. Even CPC's Corpus Christi
plant, with a poorly operating activated sludge system and
no effluent polishing, achieved an average effluent BOD
level of 0.36 kg/kkg (20.1 lb/MSBu). This value, will
improve when the polishing lagoon and clarifier are in
operation. It should be noted that none of the above plants
has installed all of the recommended 1983 technology.
Specifically, the plants rely to varying degrees on once-
through barometric condensers for cooling.
A BOD limit of 0.36 kg/kkg (20 lb/MSBu) for new plants was
also upheld by the Eighth Circuit in CPC II. (8)
The revised 1983 TSS limit of 0.54 kg/kkg (30 lb/MSBu) is
also amply supported. For example, current maximum 30-day
TSS values at American Maize are less than 0.38 kg/kkg (21
lb/MSBu), and effluent TSS levels will decrease when
chemical coagulation is added. The current maximum 30-day
TSS level at CPC Pekin plant is 0.09 kg/kkg (5.0 lb/MSBu)
following activated sludge, chemical coagulation, and
dissolved air flotation. The maximum 30-day TSS value at
Clinton Corn's treatment plant in 1975 was 0.55 (30.5
lb/MSBu). These last two figures reflect only treatment
plant discharges.
To determine single-day limits for 1983, a factor of three
was applied to the 30-day limits. The rationale for the
factors used in the 1977 limits is discussed in Section V.
The factor of three used for the 1983 TSS limit reflects
improved effluent control through a polishing step (such as
filtration) and more stringent in-plant control.
27
-------
28
-------
SECTION VII
MODIFIED STARCH STUDY
BACKGROUND
The original Development Document noted that waste waters
from modified starch production represent the largest single
source of organic load from corn wet mills. These waste
waters were described as high strength and highly variable
in terms of composition, flow, and biodegradability. It was
also noted that production of modified starches varies not
only from plant to plant, but also from day to day and week
to week at a given plant. Also, the organic strength of the
waste waters depends on the degree of starch modification.
Although EPA was aware that higher raw waste loads could
result from modified starch production, no correlation could
be. established between the types and amounts of starches
beihg produced and the resulting waste loads. EPA had
requested data on product mix and raw waste loads so that
such a correlation could be attempted. The variability of
the raw waste load could not be quantitatively defined in
terms of product mix. Furthermore, there was no evidence
indicating that waste waters from any specific process (such
as modified starch production) so affected the total plant
waste stream as to reduce the ability of a mill to implement
the best practicable control technology currently available.
For this reason, no additional allowances for modified
starch production were included in the promulgated effluent
regulations.(6)
RECENT DATA
During the new source standards remand study conducted in
1975, EPA solicited additional information from the corn wet
milling industry regarding effects of modified starch
production. On June 25, 1975, CPC submitted data on raw
waste loads and treatability for modified starch wastes.
During the design phase of the waste treatment facility at
CPC's Pekin plant, batch treatability tests were run on
various components of the mill's raw waste load. The tests
were run in 1969, using a biomass that had been developed on
mixed waste. The COD removal rates for modified starch
filtrate were considerably lower than the rates for other
waste components such as ethyl alcohol, dextrose, and
steepwater. The results would indicate that modified starch
wastes have a lower treatability (or reaction rate in the
aeration basin) than other components of a mill's waste
load. At the Pekin plant during the period June, 1973 to
29
-------
June, 1974r a particular modified starch was produced on
about 40 different occasions. On about half of these
occasions, there was a significant increase in the suspended
solids level of the waste treatment plant effluent, despite
the fact that there was no significant increase in the
organic loading (measured as COD) to the aeration basins in
the treatment system. There were about 20 other occasions
when production of modified starch did not affect
performance of the waste treatment plant.
CPC*s data submittal also included information on modified
starch raw waste loads at its Argo, Illinois, plant. Data
were presented for long- term average waste loads at the 50
percent probability level. Data were gathered during
production of regular starch, a starch ether, and a cationic
starch ether. The data are summarized below, and are
expressed per 100 lb of starch produced:
Background Plus Background Plus Background Plus
Unmodified Starch Starch Łther - Cationic Starch Ether
Flow, gal/100 lb 470 760 740
BOD, lb/100 lb 1.8 18 5.0
TSS, lb/100 lb 1.2 3.2 2.0
The information was gathered from an isolated waste stream,'
and appears to indicate significant increases in raw waste
flow, BOD, and suspended solids.
At CPC1s Corpus Christi plant, production of a particular
starch ether on two occasions caused a large increase in the
waste treatment plant effluent suspended solids. The solids
became extremely difficult to settle. On one occasion,
effluent TSS increased from 300 mg/1 to over 4,000 mg/1; at
the other time the starch ether was produced, effluent TSS
increased from 90 mg/1 to over 1,500 mg/1. CPC also
submitted data on laboratory-scale biodegradability studies
of seven simulated wastes. The studies were done as part of
the development work for the Corpus Christi waste treatment
plant. The following wastes were studied: (a) unmodified
starch filtrate, (b) hydrolyzed starch filtrate, (c)
starch ether filtrate, (d) starch ester filtrate, (e)
cationic starch ether filtrate, (f) wet starch and refinery
waste mixture. Although CPC admits that the experiments
were done on only one or two samples of waste and are
probably not truly representative, the tests do indicate
differences in observed treatability, even with acclimated
sludge.(10)
30
-------
More recently, the corn wet milling industry has submitted
additional information on modified starch raw waste loads.
In their February 28, 1977, submittal, the industry
estimated that the total raw waste load for a mill would
increase as the percentage of modified starch production
increased. The estimates were as follows:
Plant
Basic
Standard
Complex
Percentage of
Modified Starch
Less than 25
25 - 40
Total Plant
Raw Waste BOD
kq/kkq lb/MSBu
4.20 - 6.70
6.25 - 8.94
7.77 -10.28
235 - 375
350 - 500
435 - 575
These estimates were based on revisions to Table 8 in the
original Development Document. A figure of 10.72 kg/kkg
(600 lb/MSBu) BOD was used for the raw waste load
attributable to- modified starch production, although the
industry stated that the amount could be as high as 41.1
kg/kkg (2300 lb/MSBu) .
In their testimony during the Corpus Christi NPDES permit
hearings, CPC indicated that filtrates from production of
acid modified and derivatized starches contain up to 5 lb
per hundredweight of soluble and insoluble starches.
Filtrates from modified starch production were claimed to
contain up to ten times the amount of waste (per unit of
production) produced by regular starch production. CPC
developed a product mix analysis and presented the following
waste load factors for different modified starches:
Waste Load (lb BOD per
Product hundredweight of product)
Acid Modified 3.27
Cross Bonded 3.08
Double Derivatized 7.10
The above data were based on actual operating experience and
actual product loss measurements. The BOD load from acid
modified and cross bonded products results from soluble and
insoluble starch losses because of the chemical
modifications. The major part of the double derivatized
waste load also results from such losses, but there is an
additional load from the treating chemical used. Using
Corpus Christi production data and the above factors, CPC
calculated that modified starches (specials) accounted for
less than 10 percent of the grind by weight, yet contributed
about 23 percent of the total equalized waste load BOD.
31
-------
Additional information on modified starch waste loads was
submitted by Hubinger and CPC. As part of an industry data
submittal on May 2, Hubinger provided the following data on
waste loads that would not be generated if only regular
starch were produced (Hubinger*s actual submittal was dated
April 29, 1977):
Flow
cu m/kkg qal/SBu
BOD
kq/kkg lb/MSBu
TS'S
kq/kkg lb/MSBu
Before stripping
centrifuge
After stripping
centrifuge
3. 23
3.17
21.7
21.3
14.68
24.57
2,500
1,375
11.92
2.54
The above figures are based on the amount of grind converted
to modified starch. Hubinger also provided raw waste loads
for the total 80,000 bu/day mill. Waste loads for starch
modifying were: 0.25 cu m/kkg (1.7 gal/SBu), 3.43 kg/kkg
(192 lb/MSBu) BOD, 0.93 kg/kkg (52 lb/MSBu) TSS.(12)
In a data submittal dated June 3, 1977, CPC provided the
following raw waste loads for production of various modified
starches:
667
142
BOD
Type of Starch
Mill starch (unmodified)
Hydrolyzed starch
Starch ether
Starch ester
Oxidized
kq/kkg
3.04
12.15- 27.34
20.73-101.5
6.08- 46.82
3.04- 16.08
lb/MSBu
170
680-1,530
1,160-5,680
340-2,620
170- 900
CPC indicated that production of more highly modified
starches will result in even higher raw waste loads.(13)
EFFLUENT ALLOWANCE
On the basis of the recently supplied industry data, it was
determined that an effluent allowance for modified starch
production might be required for certain plants. An
allowance is to be allowed for plants producing modified
starches at a rate of at least 15 percent by dry-basis weight
of total sweetener and starch products. This figure of 15
percent was based on industry comments and on waste
treatment experience at CPC*s Corpus Christi corn wet mill.
Production of modified starches at Corpus Christi have
apparently had adverse effects on treatment plant
performance on certain occasions. In the past the mill has
32
-------
operated below a 15 percent modified starch production
level, and CPC projects production greater than 15 percent
by 1983. A plant such as CPC-Pekin would not be eligible
for a modified starch allowance at current production
levels. This determination of a modified starch allowance
for existing plants does not affect EPA* s previous
determination that such an allowance is not required for new
plants. This determination for new plants is discussed in
the 1975 Supplement to the Development Document.
The modified starch allowances for 1977 and 1983 are as
follows:
1977:
1983:
BOD
TSS
BOD
TSS
Maximum for Any
One Day
kg/kkg
0.27
0.54
0.14
0.22
lb/MSBu
15
30
8
12
Average of Daily
Values for 30 Consecutive
Days Shall Not Exceed
kg/kkg lb/MSBu
0.81
2.16
0.43
0.66
45
120
24
36
To arrive at the allowances, an additional raw waste BOD
figure of 14.3 kg/kkg (800 lb/MSBu) for modified starch
production was used. This was based on an average of the
data submitted by industry. Assuming a plant produced 15
percent modified starch, the additional raw waste load
(above production of sweeteners and regular starch) would be
14.3 kg/kkg x 0.15 = 2.15 kg/kkg (800 x 0.15 = 120 lb/MSBu).
The recommended 1977 30-day BOD effluent limitation for
plants producing less than 15 percent modified starch is
0.89 kg/kkg (50 lb/MSBu), which requires a reduction of 88
percent from the standard raw waste load of 7.15 kg/kkg (400
lb/MSBu) BOD. If the same' reduction is. applied to the
modified starch waste load, the necessary allowance would be
2.15 x 0.12 = 0.26 kg/kkg (120 x 0.12 = 14.4 lb/MSBu). The
allowance recommended is 0.27 kg/kkg (15 lb/MSBu).
The allowance for suspended solids is higher than the BOD
figure, since the data indicate that modified starch wastes
cause more problems with effluent TSS than BOD. The
recommended 1977 TSS allowance is twice the BOD allowance,
reflecting TSS:BOD ratios in effluents from existing corn
wet milling waste treatment plants.
To determine the modified starch allowance for the 1983
effluent limitations, a BOD reduction factor of 95 percent
was used, based on the reduction required by the basic 1983
33
-------
BOD limitation. A TSS;BOD ratio of 1.5 was used to
establish the TSS allowance, reflecting better solids
control through use of deep bed filtration, part of the
recommended 1983 control technology.
The factors used to convert 30-day limits to single-day
maximums are identical to the factors used for the basic
1977 and 1983 limitations. For 1977, the single-day BOD
allowance is three times the 30-day value. A factor of four
was used for the TSS limits- For both the 1983 BOD and TSS
single-day limits, a factor of three was used. The basis
for these factors is explained in Sections V and VI.
34
-------
SECTION VIII
VARIABILITY AND EXCURSIONS
VARIABILITY
The factors used to convert 30-day limitations to single-day
limitations were 4.0 for the 1977 TSS limit and 3.0 for the
1977 BOD limit and the 1983 BOD and TSS limits. Support for
these factors can be found in existing waste treatment data,
and a discussion of the rationale is included in Sections V
and VI. Variability factors at existing treatment plants
are summarized below:
BOD TSS
Max. Month Max. Day Max. Month Max. Day
Average Max. Month Average Max. Month
American Maize 1.6 1.8 2.0 3.3
Clinton Corn 1-8 3.1 1.8 3.3
CPC-Corpus Christi 5.9 2.4 2.8 4.7
CPC-Pekin (1972-74) 2.1-2.3 4.6-6.5 2.1-3.2 3.0-6.4
(1975-76) 1.5 9.5* 2.1 9.5*
~These values occurred during a period when the maximum 30-day
value was well below that required by the revised
effluent limitations.
The above factors were calculated from the data presented
and discussed in Sections IV, V, and VI. It should be noted
that these figures do not represent a strict statistical
analysis that led to determination of appropriate effluent
limitations. The figures are presented merely to show that
the variability factors in the effluent limitations are
reasonable and supported by existing data.
EPA,s new source standards remand study, completed in 1975,
indicated that the data base from existing corn wet milling
treatment plants was deficient and was not suitable for
strict statistical variability calculations. It was pointed
out that suitable data should reflect good, stable operation
of a plant that has implemented all of the recommended
technology, including end-of-pipe treatment and in-plant
controls. This was and still is not the case at any mill
within the industry, and the 1975 Supplement to the
Development Document (pp. 19-23, 43-45) makes this clear in
its evaluation of present treatment systems.(9)
35
-------
The Eighth Circuit upheld EPA*s approach to variability in
its decisions in CPC I and again in CPC II. In CPC I, the
Court stated that ~~
. "there are an infinite number of ways in which the
Administrator may approach the related questions of basic
standards, variability, and excursions". The Court felt
* that the critical issue was whether the recommended
technology would enable a plant to meet the required
effluent limitations.(7> The Court in CPC II agreed with
EPA that "strict statistical computation of the variability
factor for new plants based upon existing plant results
creates an artificially high number".(8) Although the Court
was considering new plants, the same reasoning applies to
use of data to determine limitations for existing plants.
It is inappropriate to use a data base from plants that have
not incorporated the recommended technology. The Court also
upheld EPA's use of a single-day limit three times the 30-
day limit.
EXCURSIONS
The corn wet milling industry has continually asserted that
provisions for excursions should be included in the effluent
limitations. They argue, for example, that if variability
factors based on 99 percent probability are used to
determine effluent limitations, then an additional allowance
should be made to take care of excursions occurring the
remaining one percent of the time.
The industry has asserted that even if the effluent
limitations accurately reflect daily and 30-day fluctuations
in effluent quality from a well- run treatment facility,
there will be times when the limitations are exceeded. They
claim that even the best-operated facility will occasionally
be subject to excessive discharges because of factors such
as (a) influent quality ,changes, (b) plant start-up or
shut-down conditions, (c) equipment malfunction, (d)
. catastrophic conditions, or (e) other circumstances.
Factors beyond the control of a plant that may cause upsets
or excursions, such as mechanical failure, accidental
spills, or catastrophic conditions, can be handled
informally by the permit-issuer. EPA1s "Guidelines for
Water Pollution Enforcement", 8 (July 23, 1974), provide
direction. Such uncontrollable events can be handled either
through a force majeure clause in a permit or by enforcement
discretion.
Industry claims that certain other factors causing
excursions are unique to corn wet milling plants and exist
36
-------
in the course of normal operations. The following were
cited as examples of such factors;
1. Influent quality changes reflecting intermittent
production of specialty products and changes in raw
materials used.
2. Extraordinary additions to the raw waste load.
3. Exceptional production of finished products at
plants where finishing capacity exceeds normal grind
capacity.
4. Difficulty at older plants of continuously
monitoring and controlling raw waste flows.
Industry proposes that permit-issuers be given the authority
to include an "excursion provision" in a permit for a corn
wet milling plant. Such a provision would be based on the
plant's demonstrating that one or more factors such as those
listed above do indeed cause upsets or excursions beyond the
uniform effluent limitations.
EPA has determined, that such an "excursion provision" is not
a necessary component of a corn wet milling discharge
permit. Regarding events that may occur in the course of
normal operations in the industry, EPA is bound by the
rationale of the Decision of the General Counsel on Matters
of Law Pursuant to «0 C.F.R. sec. 125.36(m) , No. 57 (March
16, 1977). That decision essentially states that a permit-
issuer may consider including an excursion or upset
provision in a permit only if the particular events giving
rise to the upset condition have not been taken into account
in establishing the effluent limitations guidelines. After
reexamining all available information, EPA has determined
that the kind of events that the industry proposes should be
accorded exceptional consideration by the permit-issuer (for
example, influent quality changes reflecting intermittent
production of certain special corn products and changes in
raw materials used) have already been incorporated into the
data base from which the guidelines were derived. Since
these types of factors were considered a part of the normal
operation of corn wet milling plants and were incorporated
into the data base and reflected in the guidelines, EPA
cannot authorize permit-issuers to reconsider these same
factors in the context of special excursion or upset
provisions.
The only factors used to exclude certain data points from
the effluent guidelines data base were occurrences such as
37
-------
labor strikes and preventable spills. Such factors can be
adequately handled either through a force majeure clause or
by enforcement discretion, as discussed above. EPA has
determined that most excursions or upsets occurring at corn
wet milling treatment plants are caused by controllable
factors or deficiencies in the treatment system. These
factors and deficiencies can be corrected through diligent
use of in-plant controls and improved design and operation'
of treatment facilities.
Provision for variability of treatment plant performance can
also be found in EPA's recommended effluent limitations and
modified starch allowance. The 30-day and daily TSS limits
for both 1977 and 1983 have been increased to more
accurately reflect data from existing treatment plants. In
addition, the modified starch allowance provides additional
effluent discharge for plants with significant production of
specialty modified starches.
In CPC II, the Eighth Circuit agreed with EPA*s position on
excursion. The Court agreed that excursions within this
industry could be controlled. The Court also pointed to the
conclusions of Dr. Raymond C. Loehr, . who stated that
industry could decrease variability by at least one percent
and more likely by five percent, through use of better
treatment technology, attention to treatment system
operation and maintenance, and improved in-plant control.(8)
It should also be noted that, if a plant is found to be
affected by factors "fundamentally different" from those
considered in establishing the guidelines, the. current
regulations allow for a "variance provision" relative to the
limitations based on application of the "best practicable
control technology currently available", 40 C.F.R. sec.
406.12 (a). Similarly, according to Section 301(c) of the
Act, the Administrator is authorized to modify the
requirements pertaining to the "best available technology
economically achievable" for an individual point source upon
a sufficient showing by the point source operator, under the
conditions specifically set forth in 301(c).
38
-------
SECTION IX
COSTS
BACKGROUND
In the original Development Document (March, 1974}, costs to
meet EPA's effluent limitations were developed for a
hypothetical corn wet mill. The hypothetical plant was a
medium-sized mill with a daily grind of 1,524 kkg (60,000
SBu) and was assumed to practice good in-plant control and
to use recirculated cooling water. The waste water
characteristics for this hypothetical mill were as follows:
The costs developed for waste treatment at the hypothetical
mill were based on August, 1971 dollar values. Capital
costs for 1977 and 1983 treatment levels were as follows:
The cost for BPT included equalization and activated sludge.
The BAT cost included the addition of deep bed filtration.
In addition, a cost of $288,000 was estimated for a
recirculating cooling tower system for a plant utilizing
once-through barometric condensers.
As part of the new source standards remand study conducted
in 1975, EPA prepared new cost estimates for three different
sized hypothetical corn wet mills. The costs were reported
in terms of January, 1975 dollars. These updated costs are
summarized below (all costs are in $1,000):
Flow
BOD 5
TSS~
11,355 cu m/day (3 mgd)
7.14 Jcg/kkg (400 lb/MSBu)
3.57 kg/kkg (200 lb/MSBu)
Treatment Levels
Capital Costs
BPT (1977)
BAT (1983)
$2,544,000
$ 286,000
30,000
bu/dav
60,000
bu/day
90,000
bu/dav
Capital Costs
Waste Treatment (1977)
Filters (1983)
Recirculating Barometric
3,017
160
215
4,768
240
335
6,407
320
430
Cooling
Operation and Maintenance
39
-------
Waste Treatment 260 378 483
The waste treatment facilities costed in the above figures
include 24- hour equalization, raw waste cooling, nutrient
addition, activated sludge treatment, and chlorination.
Deep bed filters are costed separately.
In 1975, the industry provided data and estimates on the
costs of surface condensers versus barometric condensers.
Grain Processing Corporation provided the following data on
costs of replacing barometric condensers with surface
condensers:(14)
Date Equipment Cost Installed Cost
July, 1974 $49,354 $ 89,626
Nov, 1974 $62,800 $215,093
May, 1975 . $29,684 $ 68,000
CPC provided estimates for costs of barometric condensers
versus surface condensers in new plants. Estimates were .
included for three different new plant sizes:(10)
30,000 60,000 . 90,000
Condensers bu/day bu/day bu/day
Barometric $ 660,000 $1,000,000 $1,300,000
Surface 1,600,000 2,300,000 3,000,000
CURRENT COSTS
In the current study, the technologies comprising BPT and
BAT have not changed from those discussed in the original
1974 Development Document. These technologies are
summarized in Sections V and VI of this report. To develop
up-to-date costs for BPT and BAT, the figures presented in
the original 1974 Development Document and in the 1975
Supplement to the Development Document were reviewed and
updated to 1977 dollar*values. The EPA-STP cost index was
used for updating. The following index values were used:
August, 1971 - 162.5; January, 1975 - 225.7; January, 1977 -
270.9.
The updated costs for 1977 and 1983 are presented in Table
3. It should be noted that these figures represent costs
for a plant that has not yet installed any waste treatment
facilities and that relies on once-through barometric
condensers for cooling. This is not the case for any plant
now discharging directly to a waterway. The treatment
facilities now in place for the four direct dischargers in
40
-------
the industry approach BAT technology. Of the four plants,
one was designed with primarily surface condensers, and two
others have made significant replacements of barometric
condensers with surface condensers.
41
-------
TABLE 3
COSTS TO MEET 1977 AND 1983
EFFLUENT LIMITATIONS
Capital Costs
($1,000)
30,000 60,000 90,000
bu/day
bu/dav
bu/dav
1977 (Equalization and
7,688
Activated Sludge)
3,620
5,722
Cooling Tower (Recirculating
646
Barometric Condenser Water)
305
481
Surface Condensers
2,820
3,900
5,100
1983 (Filters)
305
481
646
Annual costs
(OEM)
($1,000)
30,000
60,000
90,000
bu/dav
bu/dav
bu/dav
1977 (Equalization and
580
Activated Sludge)
312
454
Cooling Tower
19-1
27.7
34.6
Surface Condensers
48
72
108
1983 (Filters)
46
73
98
42
-------
SECTION X
REFERENCES
1- Memorandum - Job 4857E, Subject: Visit to American Maize Com Met
Milling Plant# June 6, 1977.
2. Cleary, Gottlieb Data Submittal to EPA, May 2, 1977.
3. "Wastewater Sampling Program for Outfall 001", CPC International
Inc., Corpus Christi, Texas, Prepared for the Texas Water Quality
Board, March 16, 1976.
4. Memorandum - Job 4857E, Subjects Visit to CPC - Pekin Corn Wet
Milling Plant, August 30, 1976.
5. Memorandum - Job 4857E, Subject: Visit to CPC - Pekin, June 7,
1977.
6. D. S. Environmental Protection Agency, rDevelopment Document for
the Grain Processing Segment of the Grain Mills Point Source
Category", EPA-440/1-74-028-a, March, 1974. .
7. CPC International Inc. v. Train. 515 F.2d 1032 (May 5, 1975) (CPC_I) .
8. CPC International Inc. v. Train. 540 F.2d 1329 (August 18, 1976)(CPC II)
9. U. S. Environmental Protection Agency, "Supplement to Development
Document for the Corn Wet Milling Subcategory, Grain Processing
Segment of the Grain Mills Point Source Category", EPA-440/1-74-
028-b, August, 1975.
-10. CPC Data Submittal to SSP, June 25, 1975.
11. Cleary, Gottlieb Data Submittal to EPA, February 28, 1977.
12. Cleary, Gottlieb Data Submittal to EPA, May 2, 1977.
13. Cleary, Gottlieb Data Submittal to EPA, June 3, 1977.
14. Grain Processing Corporation Data Submittal to SSP, June 3, 1975.
43
-------
***4
TABLE 4
METRIC TABLE
CONVERSION TABLE
MULTIPLY (ENGLISH UNITS)
by
TO OBTAIN (METRIC UNITS)
ENGLISH UNIT
ABBREVIATION CONVERSION
ABBREVIATION
METRIC UNIT
ere
ac
0.405
ha
hectares
acre - feet
ac ft
1233.5
cu m
cubic meters
British Thermal
Unit
BTU
0.252
kg cal
kilogram - calories
British Thermal
Uni t/pound
BTU/lb
0.555
kg cal/kg
kilogram calories/kilogram
cubic feet/minute
cfm
0.028
cu m/min
cubic meters/minute
cubic feet/second
cfs
1.7
cu m/min
cubic meters/minute
cubic feet
cu ft
0.028
cu m
cubic meters
cubic feet
cu ft
28.32
• 1
1i ters
cubic inches
cu in
16.39
cu cm
cubic centimeters
degree Fahrenheit
•F
0.555(°F-32)
* *C
degree Centigrade
feet
ft
0.3043
m
meters
gallon
gal
3.785 .
1
1i ters
gallon/minute
gpm
0.0631
1/sec
liters/second
horsepower
hp
0.7457
kw
killowatts
inches
in
2.54
cm
centimeters
inches of mercury
in Hg
0.03342
atm
atmospheres
pounds
lb
0.454
kg
kilograms
million gallons/day
mgd
3,785
cu m/day
cubic meters/day
mile
mi
1.609
km
kilometer
pound/square
(0.06805 psig +1)
atmospheres (absolute)
inch (gauge)
psig
* atm
square feet
sq ft
0.0929
sq m
square meters
square inches
sq in
6.452
sq cm
square centimeters
ton (short)
ton
0.907
kkg
metric ton (1000 kilograms)
yard
yd
0.9144
m
meter
* Actual conversion,
not a multiplier
44
------- |