DRAFT
DEVELOPMENT DOCUMENT FOR
EFFLUENT LIMITATIONS GUIDELINES
AND NEW SOURCE PERFORMANCE STANDARDS
MISCELLANEOUS FOODS AND BEVERAGES
POINT SOURCE CATEGORY
PART II
EFFLUENT GUIDELINES DIVISION
OFFICE OF WATER AND HAZARDOUS MATERIALS
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
MARCH 1975
-------
NOTICE
The attached document is a DRAFT CONTRACTOR'S REPORT. It includes tech-
nical information and recommendations submitted by the Contractor to the
United States Environmental Protection Agency ("EPA") regarding the sub-
ject industry. It is being distributed for review and comment only. The
report is not an official EPA publication and it has not been reviewed by
the Agency.
The report, including the recommendations, will be undergoing extensive
review by EPA, Federal and State agencies, public interest organizations
and other interested groups and persons during the coming weeks. The
report and in particular the contractor's recommended effluent guidelines
and standards of performance is subject to change in any and all respects.
The regulations to be published by EPA under Sections 304(b) and 306 of
the Federal Water Pollution Control Act, as amended, will be based to a
large extent on the report and the comments received on it. However,
pursuant to Sections 304(b) and 306 of the Act, EPA will also consider
additional pertinent technical and economic information which is developed
in the course of review of this report by the public and within EPA. EPA
is currently performing an economic impact analysis regarding the subject
industry, which will be taken into account as part of the review of the
report. Upon completion of the review process, and prior to final pro-
mulgation of regulations, an EPA report will be issued setting forth EPA's
conclusions regarding the subject industry, effluent limitations guide-
lines and standards of performance applicable to such industry. Judgements
necessary to promulgation of regulations under Sections 304(b) and 306 of
the Act, of course, remain the responsibility of EPA. Subject to these
limitations, EPA is making this draft contractor's report available in
order to encourage the widest possible participation of interested per-
sons in the decision making process at the earliest possible time.
The report shall have standing in any EPA proceeding or court proceeding
only to the extent that it represents the views of the Contractor who
studied the subject industry and prepared the information and recommenda-
tions. It cannot be cited, referenced, or represented in any respect in
any such proceedings as a statement of EPA's views regarding the subject
industry.
U. S. Environmental Protection Agency
Office of Hater and Hazardous Materials
Effluent Guidelines Division
Washington, D. C. 20460
Please note: Because of the volume of this report, it has been printed
in the following manner: "Miscellaneous Foods and Beverages."
Part I Pgs. 1-292 Section I-IV
Part II Pgs. 293-500 Section V-VI
Part III Pgs. 501-840 Section VII -
Part IV Pgs. 841-1196 Section VIII (partial)
Part V Pgs. 1197-1548 Section VIII (cont.) - XIV
-------
-------
DRAFT
SECTION V
WATER USE AND WASTE CHARACTERIZATION
The purpose of this section is to identify, for those subcategories
defined in Section IV, the wastewater quantities and constituents which
are characteristic of the subcategory. For each subcategory discussed
herein, a representative model is developed and defined in terms of
wastewater flow and characteristics.
It should be carefully noted that within this document, all pollutant
concentrations and loadings, unless otherwise specified, are in terms
of net units, i.e., do not include pollutants entering the process
in the fresh water supply.
It should also be noted that the raw wastewater flows and character-
istics described for each model plant are intended only to be represent-
ative of the subcategory, primarily as a basis for developing control
and treatment technology and cost analyses to be developed subsequently
in Sections VII and VIII of this document. These values should not under
any conditions be construed as being exemplary nor used as a basis
of pretreatment guidelines for industrial discharges into publicly
owned treatment works.
All pollutant parameters (except pH, color, and temperature) are
ultimately expressed as a ratio of their mass in kilograms to a process
unit. The process unit may be kkg or cu m (or in one case proof gallons)
of product or raw material produced or consumed per day. Table 16
defines the process units used for each subcategory.
VEGETABLE OIL PROCESSING AND REFINING
That segment of the miscellaneous foods and beverages industry involved
in the processing and refining of vegetable oil (including the production
of margarine) has been subcategorized into subcategories A 1 through
A 15 (see Table 13 in Section IV.)
Subcategories A 1 through A 4 cover those installations processing un-
refined vegetable oil from various oilseeds and the production of olive
oil by hydraulic press and solvent extraction in combination, and by
mechanical screw press extraction.
Subcategories A 5 through A 15 include those installations engaged in
what can generally be called edible oil refining. The historical data
complied for this study by the Institute of Shortening and Edible Oils
(ISEO) in conjunction with contractor plant visitations and verification
sampling of ten plants represents the most current information available
on the wastewater characteristics of edible oil refineries. Wastewater
293
-------
DRAFT
TABLE 16
PROCESS UNITS EMPLOYED FOR
THE MISCELLANEOUS FOODS AND BEVERAGES
SUBCATEGORY
POINT SOURCE CATEGORY
VEGETABLE OIL PROCESSING AND REFINING
Al, A2,
A3, A4
A5 - A12
A13, A14, A15
A16, A17, A18
A19
A20, A21
A22, A23
A24
A25
A26, A27
BEVERAGES
PROCESS UNIT
kkg of oilseed
crushed/day.
kkg of raw olives
crushed/day.
kkg of crude oil
processed/day.
kkg of finished
product.
cu m of beer
produced/day.
kkg of barley
processed/day.
during crushing, kkg
of grapes crushed/
day; during process-
ing, cu m of wine
produced/day.
kkg of grain
mashed/day.
proof gallons of
spirits produced/day.
None.
cu m of beverage
produced/day.
294
-------
DRAFT
TABLE 16 (CONT'D)
SUBCATEGORY
A28
A30
C8, C9, CIO
F1
BEVERAGES
BAKERY AND CONFECTIONERY PRODUCTS
All Subcategories
All Subcategories
PET FOODS
MISCELLANEOUS AND SPECIALITY ITEMS
A29
A31
A32
A33
PROCESS UNIT
cu m of syrup or
concentrate
produced/day.
kkg of instant
tea produced/day.
kkg of green
coffee beans.
None.
kkg of finished
product/day
kkg of finished
product/day.
cu m of finished
product/day.
kkg of granular
bouillon
produced/day.
kkg of solid pro-
duct produced/day.
(1) kkg of yeast
packaged/day.
295
-------
DRAFT
TABLE 16 (CONT'D)
SUBCATEGORY PROCESS UNIT
A34, A35 kkg of peanut
butter produced/
day.
A36 kkg of dry pectin
produced/day.
A37, B1-B4, C6, C12, D4 kkg of finished
product/day.
C4, C5 kkg of raw eggs
processed/day.
E1-E6, F2-F4 None.
296
-------
DRAFT
characteristics within the industry vary widely from plant to plant due
to differences in degrees of process variation, plant size, and the types
of oils processed daily. However, for the most part, process variation
is the single most important factor in determining the total waste load
for a particular refining operation. The total waste loading for an
edible oils refinery is dependent upon the individual waste load contri-
butions from the various integrated process units within the refinery.
In general terms, large, integrated, full scale refineries produce sign-
nificantly higher wasteloads than small, less integrated operations.
The principle sources of wastewater discharge within the industry are
from the following process units: acidulation; caustic refining;
contact cooling tower blowdown from barometric condensers; tank car
cleaning; storage and handling facilities; margarine; shortening; and
table oils packaging; and general cleanup from oil processing procedures
such as hydrogenation, bleaching, deodorization, and winterization.
Figure 42 in Section III presents a schematic diagram of the various
wastewater flows from individual process units for a typical full
scale, integrated, edible oils refining operation. Table 17 presents
a summary of the waste loading characteristics of individual unit processes
commonly associated with edible oil refineries as described in Section III.
Due to the high degree of variability in refinery plant size and process
integration, it was necessary to adopt a building block approach for
the formulation of the model plant and its associated unit process
waste streams. Model plants were developed for subcategories A 5
through A 14 by combining the waste load for the various unit processes
making up a subcategory. For example the Subcategory A 5 model plant,
includes the unit processes of caustic refining, tank car cleaning and
storage and handling. A total waste load for Subcategory A 5 was
derived by converting all unit process waste loads to a 454 kkg (500 ton)
per day plant and then, by summation of the unit waste loads, a total
waste load was assumed for each parameter. The hypothetical model
plants developed utilizing this procedure are intended to be representa-
tive of the subcategory as it presently exists, but cannot be expected
to be identical to any particular plant. In some cases the model may
be representative of an actual refinery only to a limited extent, but
in all cases the model is considered adequate for the purpose of devel-
oping control and treatment technology (Section VII) and for cost
analyses (Section VIII).
SUBCATEGORY A 1, OILSEED CRUSHING, EXCEPT OLIVE OIL. FOR DIRECT SOLVENT
EXTRACTION AND PREPRESS SOLVENT EXTRACTION OPERATIONS
A total of six direct solvent extraction plants and two prepress solvent
extraction facilities were visited and verification sampling was con-
ducted at four direct solvent extraction plants.
297
-------
TABLE 17
SUMMARY OF UNIT PROCESS RAW DATA ON EDIBLE OIL REFINERY WASTEWATER
CHARACTERISTICS
Unit
Process
Caustic
Ref i ni ng
Acidulation
Contact Cooling
Tower Slowdown
Oil Processing*
ro
VO
00
Tank Car
Cleaning
Storage and
Handling
Shortening and
Table Oil
Production
Margarine
Production
Production
KKG/DAY
Ave. 320
Std, Dev. 221
Ave.
Std. Dev.
Ave.
Std. Dev.
Ave.
Std. Dev.
Ave.
Std. Dev.
Ave.
Std. Dev.
Ave.
Std. Dev.
Ave.
Std. Dev.
486
459
348
264
389
212
167
112
285.
80
195
103
112
61
Flow BOD COD
MVDAY kg/kkg kq/kkg
72
145
225
148
178
135
25
22
38
32
83
159
75
113
169
139
1.01
1.58
4.69
5.08
2.21
3.51
0.09
0.23
0.49
0.84
1.36
4.33
0.48
0.75
1.93
4.06
1.8
1.7
14.97
23.44
4.24
5.73
0.22
0.47
1.38
2.41
3.83
14.60
0.19
0.15
4.23
5.63
SS
kg/kkg
0.51
1.13
1.66
3.84
0.31
0.37
0.05
0.08
0.19
0.24
0.87
2.54
0.18
0.24
1.34
2.41
0 & G
kg/kkg
0.61
O.C6
1.20
3.06
0.30
0.34
0.02
0.03
0.20
0.31
0.69
2.47
0.19
0.38
2.86
5.6
PH
Range
7.3 - 11.9
0.6 -
3.3 -
7.3 -
5.5 -
2.5 -
6.1 -
6.0 -
3
7.3
13.0
8.9
11.1
11.5
8.0
BOD/COD
Ratio
0.46
0.19
0.58
0.20
0.53
0.16
0.48
0.22
0.42
0.25
0.51
0.11
0.52
0.10
0.53
0.23
BOD/0 & G
Ratio
3.4
3.9
77.69
153.56
15.91
25.78
13.99
34.33
4.36
4.87
53.89
201.79
5.08
6.68
4.14
4.29
* Includes floor wash and general cleanup of the following unit processes:
Hydrogenation, deodorization, winterization, and Bleaching.
-------
DRAFT
The principal sources of contact process wastewater generated from
solvent extraction operations include wastewater generated from
soybean oil degumming operations to remove and recover phosphatides
(lecithin); wastewater generated from wet scrubber systems to reduce
air particulate emmissions from mill preparation areas; wastewater con-
taining oil, grease, and solvents, resulting from the extraction of oil-
seeds; steam condensates contaminated by oils, fatty acids, or hexane
solvent; and periodic in-plant floor washing and equipment cleanup
represented by oil or miscellaneous spillage, valve or pump leakage,
etc. In addition to these process wastes, a large number of processors
were observed to combine their process wastewaters with non-contact cool-
ing water from cooling tower and boiler blowdown.
Historical data supplied by the NSPA and the National Cottonseed Producer's
Association (NCPA) for 18 solvent extraction facilities in combination
with four verification surveys found the following averages for
Subcategory A 1 plants:
Production
Flow
BOD
COD
SS
O&G
PH
BOD/COD Ratio
BOD/O&G Ratio
780 kkg/day (860 ton/day)
140 cu m/day (0.037 MGD)
0.115 Ib/ton)
0.281 Ib/ton)
0.07 Ib/ton)
311 mg/1; 0.058 kg/kkg
619 mg/1; 0.140 kg/kkg
140 mg/1; 0.035 kg/kkg
253 mg/1; 0.064 kg/kkg (0.128 Ib/ton)
6.2 to 10.4
0.50
19.8
Table 18 presents a statistical description of the process wastewater
characteristics compiled during the study including mean, standard
deviations, minimum and maximum values.
There was a significant correlation observed in the industry between
the volumes of process wastewater discharged per day and total daily
production as is evidenced in Figure 107. However, there was no corre-
lation indicated between production, BOD, COD, or oil and grease concen-
trations. These data are summarized by the scatter diagrams presented in
Figures 108, 109, and 110.
Total Process Effluent
As indicated in the data presented above, the pollutant concentrations
and waste loadings for solvent extraction plants are highly variable due
to the following in-plant variations: (1) the amount of wet cleanup
and general housekeeping practices utilized by each plant, (2) the
quality of seed being crushed, and (3) plants that perform soybean oil
degumming periodically in combination with solvent extraction processes.
299
-------
TABLE 18
A STATISTICAL DESCRIPTION OF THE WASTEWATER CHARACTERISTICS
FOR SOLVENT EXTRACTION PROCESS WASTEWATER
o
73
co
o
o
VARIABLE
Flow (MGD)
Prod, (ton/day)
BOD (mg/1)
SS (mg/1)
COD (mg/1)
*FOG (mg/1)
BOD (Ib/day)
COD (Ib/day)
SS (Ib/day)
FOG (Ib/day)
Lb/Ton-BOD
kg/kkg-BOD
Lb/Ton-COD
kg/kkg-COD
Lb/Ton-SS
kg/kkg-SS
Lb/Ton-FOG
kg/kkg-FOG
BOD/COD Ratio
BOD/FOG Ratio
Flow Ratio
N
58
58
49
52
48
US
49
46
52
45
49
49
46
48
52
52
45
45
40
37
58
MEAN
0.037069
859.996552
311.367347
139.8PU615
619.333333
252.786667
78.002401
176.221113
38.317027
68.104652
0.115201
0.057601
0.2S1114
0. 1UG557
0.070004
0.035002
0.127937
0.063968
0.503020
19.806625
46.204019
STANDARD
DEVIATION
0.019112
428.025181
403.986381
198.246297
572.186450
759.765755
56.897093
154.727087
53.240701
205.259085
0.121087
0.060544
0.350194
0.175097
0.116903
0.059451
0.421886
0.210943
0.203246
32.029051
19.213248
VARIANCE
0.000365
183205.555777
163206.612245
39301.594268
327397.333333
577244.002545
3237.279?
-------
LEGENDt 1 = ONE OBSERVATION, 2 = TWO O5ERSVATIONS. ETC.
0.090
0.072
0.054
0.036 -
0.018 -
0.0 _
LINEAR REGRESSION
COEFFICIENT = +0.70
_L
110 470 830 1190 1550 1910
PRODUCTION (TON/DAY)
FIGURE 107
A LINEAR REGRESSION PLOT OF FLOW (MGD) VERSUS PRODUCTION (TON/DAY)
FOR PROCESS WASTEWATERS DISCHARGED FROM OILSEED SOLVENT EXTRACTION PLANTS. SUBCATEGORY Al
o
JO
-------
3000
2400
1800
GO
O
ro
aaoo
600
LEGEND: 1 = ONE OBSERVATION, 2 = TWO OBSERVATION, ETC.
4 i
I
I >
I
2 4
' ft
i i
* < i
7 ,
* 4 i
4 • >
it a
110 470 830 1190 155O 1910
PRODUCTION (TON/DAY).
FIGURE 108
A SCATTER DIAGRAM PLOTTING BOD CONCENTRATION VERSUS PRODUCTION (TON/DAY) FOR THE PROCESS WASTEWATERS
GENERATED FROM OILSEED SOLVENT EXTRACTION PLANTS, SUBCATEGORY A l
-------
co
3000
2000 '
1600
sS
u o
1200
600
0
1 i_EGEND« 1 = ONE OBSERVATION, 2 = TWO OBSERVATI
•..'•'•' 'l • ' • •
• ,
1
-
/ 1
a
1 3 2 1 , ,
3 i i 1 1 II
. ' ,
3 i i 1 •
« 1
I I 1 I 1 1
110
470
830 1190
PRODUCTION (TON/DAY)
FIGURE 109
1550
1910
A SCATTER DIAGRAM PLOTTING COD CONCENTRATIONS VERSUS PRODUCTION (TON/DAY)
FOR THE PROCESS WASTES FROM OILSEED SOLVENT EXTRACTION PLANTS, SUBCATEGORY Al
-------
CO
o
J\J V^
4000
£ 3000
i ^
< s
1-4
C
2000
1000
0
LEGEND: 1 = ONE OBSERVATION, 2 = TWO OBSERVATIC
t '
—
_
i
i
— ~
2 1
3 HI 1
3315 *
2531 1 5 SI 1 11 3 12 till
" 1 1 1 . 1 . .. i —
110
470
830
1190
73
1550
PRODUCTION (TON/DAY)
FIGURE 110
A SCATTER DIAGRAM PLOTTING CONCENTRATIONS OF OIL AND GREASE VERSUS DAILY PRODUCTION (TON/DAY)
FOR THE PROCESS WASTEWATERS DISCHARGED FROM OILSEED SOLVENT EXTRACTION PLANTS. SUBCATEGORY Al
-------
DRAFT
Model Plant
The model plant for subcategory A 1 is based on the following
assumptions:
1. The model plant is assumed to have a daily production of
816 kkg (900 ton).
2. The model plant has a flow volume of 0.144 cu m/day
(0.039 MGD).
3. The model plant may or may not have the unit process
of degumming.
4. The model plant may or may not have a wet scrubber
system for removing air particulates.
By converting the data base compiled for this study to a model plant
production of 816 kkg (900 ton) per day by multiplying by a factor of
1.05 (i.e., 816 kkg/780kkg = 1.05), the following wastewater
characteristics were derived for the Subcategory A 1 model plant.
Production
Flow
BOD
COD
SS
O&G
BOD/COD Ratio
pH Range
816 kkg/day
148 cu m/day (0.039 MGD)
340 mg/1; 0.061 kg/kkg (0.122 Ib/ton)
815 mg/1; 0.147 kg/kkg (0.244 Ib/ton)
210 mg/1; 0.038 kg/kkg (0.076 Ib/ton)
380 mg/1; 0.069 kg/kkg (0.138 Ib/ton)
0.50
6 to 8
SUBCATEGORY A 2 - OILSEED CRUSHING, EXCEPT OLIVE OIL, BY MECHANICAL
SCREW PRESS OPERATIONS
Seven typical mechanical screwpress extraction plants (three cotton-
seed crushers and four peanut crushers) were visited in conjunction
with information from the National Cottonseed Producer's Association
and the Southeastern Peanut Association. Only two sources of contact
wastewater were observed. These consisting of 1) contaminated steam
condensate from steam cooker operations and 2) wastewaters generated
from infrequent floor and equipment cleanup. Four sources of non-
contact wastewater were observed from the following unit processes:
1) non-contact cooling water circulated through the hollow expeller
worm shaft to keep the oilseed cakes from burning, 2) boiler blow-
down, 3) non-contact cooling tower blowdown (only during the winter
months), and 4) storm water runoff. In general, the resultant con-
tact wastewater generated from screw press operation is less than
4,000 liters (1000 gallons) per day. Screw press operations near to
or in conjunction with an edible oils refinery dispose of wastewater
by trucking it to the refinery where the oil is recovered in the
acidulation process. Three plants were also observed to recycle their
wastewater into the boiler feed water. Due to the small volume of waste-
water discharged, it is not necessary to develop a model plant for Sub-
category A 2.
305
-------
DRAFT
SUBCATEGORY A 3 - OLIVE OIL EXTRACTION BY HYDRAULIC PRESSING AND
SOLVENT EXTRACTION
The process descriptions of the extraction of olive oil by hydraulic
pressing and solvent extraction were presented in Section ill. At
present, there is only one plant which utilizes either the hydraulic
press or solvent extraction processes for the recovery of olive oil.
The only source of wastewater generated by the extraction of olive oil
by hydraulic pressing is centrifuge fruit water. Wastewater attribu-
table to solvent extraction consists of a small amount of water which
drains from pits and culls during storage, and an equally small non-
contact condenser water flow. Equipment is wiped clean.
The wastewater from the hydraulic pressing process was determined to
have the following characteristics:
Flow 10.9 cu m/day (0.0029 MGD)
BOD 63,000 mg/1
SS 14,000 mg/1
FOG 3,220 mg/1
pH 5.1
Model Plant
The model plant for this subcategory is plant 79102. Between the months
of October and June, the plant generally operates 24 hours per day, seven
days per week with the operating schedule dependent on olive crop yield
and availability of harvesters.
The total plant effluent consists of centrifuged fruit water with the
characteristics listed above.
SUBCATEGORY A 4 - OLIVE OIL EXTRACTION BY MECHANICAL SCREW PRESSING
At present there is only one olive oil manufacturer in the United States
which extracts olive oil by the mechanical screw press process. The ex-
traction of olive oil by screw press operations produces wastewater from
the following sources:
1. Washing of whole ripe olives prior to pulverizing
2. Centrifuged fruit water
3. Centrifuged sludge
4. General plant cleanup
Fruit Wash Water
Prior to grinding in the hammer mill the fruit is washed by pump
and air percolation washers. These wash tanks are filled and discharged
306
-------
DRAFT
at least once per day or more depending upon fruit condition. The quantity
of wastewater discharged from the washers varies between 19 cu m/day (0.005
MGD) and 38 cu m/day (0.010 MGD).
Centrifuged Fruit Water
The quantity of fruit water generated by the centrifuge is approximately
38 1/min (10 gal/min) for a total centrifuge effluent of 54.5 cu m
(0.0144 MGD). The constituents of the centrifuged fruit water indicate
a BOD concentration of 60,000 mg/1 and a fat content of 25 percent.
Centrifuged Sludge
Approximately 38 cu m/day (0.010 MGD) of centrifuged sludge is generated
from the initial centrifuge following pressing. The pollutant concentra-
tions of the centrifuged sludge were determined to be as follows:
BOD 48,000 mg/1
SS 57,000 mg/1
FOG 34,000 mg/1
General Plant Cleanup
Cleanup of equipment is done on an irregular basis with little generation
of wastewater. Due to the irregular nature and inherent variability of
the cleaning operation, representation of waste flow cannot be reliably
determined. It is, however, reflected in the total waste discharge.
Total Plant Effluent
The total effluent from the plant would amount to approximately 114 cu
m/day (0.03 MGD) and would have the following characteristics:
BOD 30,000 mg/1
SS 57,000 mg/1
FOG 20,000 mg/1
pH 5.5
Selection of Model Plant
The model plant, illustrated in Figure 39 in Section III, processes
44 kkg/day (48 ton/day) of pljves. The total plant effluent consists
of the combined waste streams as previously presented. The plant operates
24 hours per day, seven days per week between the months of September
and April except during unpredictable harvesting lulls. The plant's
wastewater has the following characteristics:
Flow 114 cu m/day (0.03 MGD)
BOD 30,000 mg/1
SS 57,000 mg/1
FOG 20,000 mg/1
pH 5.5
307
-------
DRAFT
SUBCATEGORY A 5 - PROCESSING OF EDIBLE OIL BY THE USE OF CAUSTIC REFINING
METHODS ONLY
The individual unit processes characteristic of Subcategory A 5 plants
include: (1) caustic refining operations; (2) general cleanup of
storage and handling facilities; and (3) tank car cleaning operations.
Caustic Refining
A principle source of wastewater generation from Subcategory A 5 refineries
results from the caustic refining of crude vegetable or animal oils.
Wastewater discharged from the washing of refined edible oils will
vary considerably from day to day depending upon the nature of the
crude oil being refined. Seng (53) reported an Illinois caustic refining
operation to have the following average pollutant concentrations:
BOD, 1240 mg/1; COD, 5000 mg/1; suspended solids, 690 mg/1I; and ether
solubles, 1800 mg/1. The average waste loads for the Illinois plant
were: BOD, 0.27 kg/kkg (0.55 Ib/ton) COD, 1.1 kg/kkg (2.2 Ib/ton);
suspended solids, 0.15 kg/kkg (0.30 Ib/ton); and ether solubles 0.4
kg/kkg (0.8 Ib/ton). The average flow was recorded as 0.054 cubic
meters per day (0.0144 MGD).
Historical and verification survey data compiled for this report from
six edible oil caustic refining operations found significantly higher
concentrations of BOD, COD, suspended solids, and oil and grease. Mean
concentrations and wasteload values from all data collected were:
Production 353 kkg
Flow 75.7 cu m/day (0.02 MGD)
BOD 6,900 mg/1; 1.01 kg/kkg (2.02 Ib/ton)
COD 14,800 mg/1; 1.8 kg/kkg (3.6 Ib/ton)
SS 3,700 mg/1; 0.5 kg/kkg (1.0 Ib/ton)
O&G 5,000 mg/1; 0.6 kg/kkg (1.2 Ib/ton)
pH Range 7.3 to 11.9
BOD/COD Ratio 0.46
Table 19 provides a statistical description of the data compiled from
six refineries including mean, sample size, standard deviations, minimum^
and maximum values. Table 20 presents a summary of caustic refining
waste loadings from the six plants visited and sampled during the course
of this study. As would be expected, calculated correlation coefficient
statistics show a significant correlation between the concentrations
of BOD and COD in the caustic refining wastewater with a calculated BOD/COD
ratio of 0.46. A significant correlation also exists between the kg/kkg
of BOD, suspended solids, and oil and grease. These data indicate that
much of the hexane extractable material exists as oil attached to suspended
solids particles with a specific gravity close to that of water.
Tank Car Cleaning
The cleaning of tank cars to remove residual oil constitutes a major
waste stream associated with all Subcategory A 5 through A 12 edible
308
-------
TABLE 19
A STATISTICAL DESCRIPTION OF THE WASTEWATER CHARACTERISTICS FOR
THE EDIBLE OIL CAUSTIC REFINERY PROCESS
CO
o
IO
MEAN
0.019002
352.902439
6933.410256
3715.780186
14781. 675676
SOUS. 658537
552.153229
1162.017060
233.036323
396.015192
2.02G632
1.010316
3.620173
1.81C237
1.027^17
0.513708
1.229996
0.614998
0.466169
3.046033
55.133023
STANDARD
DEVIATION
0,038381
243.959882
7747.790968
7573.994869
14655,901695
8905.132516
636.396225
1469.154629
390.209634
649. 220997
3.170595
1.585298
3.405427
1.702713
2.259675
1.129837
1.712138
0.856069
0,191399
3.909349
94.653221
VARIANCE
0.001
59516.424
60028264.880
57365398.276
214795454.503
79301385.130
405000.155
2159296.908
152263.558
421487.902
10.053
2.513
11.597
2.899
5.106
1.277
2.931
0.733
0.037
15.283
8959.232
MINIMUM
0,001000
38,300000
•35.0000CO
20.000000
100.000000
8.000000
3,067650
20,078070
1.268440
6.375560
0.056093
0.026046
0.074529
0.037264
0.030129
0,015065
0.018893
0.009447
0,153782
0,059710
9.H5993
MAXIMUM
0.216000
1016.000000
36300.000000
41660.000000
61352.000000
49456,000000
2461.023950
7167.754160
1657,817700
3619.159740
18.887367
9.443684
13.408466
6.704234
12.723083
6.361541
8,491982
4.245991
0.878690
20.277778
496.551724
COEFFICIENT
COVARIENCE
201.982
69.130
111.746
203.833
99.149
176.491
115.249
124.318
167.446
163.926
156.911
156.911
94.060
94.060
219.938
219.938
139,199
139.199
41.056
113,379
171.682
OF
(*)
VARIABLE N
Flow (MGD) «1
Prod, (ton/day) «1
BOD (mg/1) 1*
SS'(mg/l) "1
COD (mg/1) 37
*FOG (mg/1) "1
BOD (Ib/day) 39
COD (Ib/day) 37
SS (Ib/day) al
FOG (Ib/day) "1
Lb/Ton-BOD 39
kg/kkg-BOD 39
Lb/Ton-COD 37
kg/kkg-COD 37
Lb/Ton-SS «1
kg/kkg-SS «»
Lb/Ton-FOG «1
kg/kkg-FOG "»
BOD/COD Ratio 35
BOD/FOG Ratio 39
Flow Ratio "1
* FOG = Fats, Oils, and greases.
N ° Number of data points
Note: Computer calculations for this table show no regard for significant figures.
-------
CO
o
Edible
Oils
Refinery
(Process Code)
75R08
75R09
75R15
75RT7
75R05
75R06
TABLE 20
POLLUTANT LOADINGS FOR CAUSTIC REFINING WASH WATERS
Production
(kkg/day)
424
388
310
245
227
276
Vol ume
Wastewater
Discharged
(cu m/day)
29.2
331.6
36.6
54.5
31.5
56.1
BOD
(kg/kkg)
0.39
0.88
0.49
0.28
1.43
2.15
COD
(kg/kkg)
0.99
2.18
2.53
1.11
2.37
0.90*
SS
(kg/kkg)
0.55
0.11
0.38
0.15
0.36
1.46
Oil and
Grease
(kg/kkg)
0.86
0.64
0.28
0.40
0.54
0.75
COD sample size was less than BOD sample size.
-------
DRAFT
oil refining facilities. Average concentrations and waste loading of
pollutants from six plants were as follows:
Production 184 kkg
Flow 37.8 cu m/day (0.01 MGD)
BOD 2950 mg/1; 0.49 kg/kkg (0.98 Ib/ton)
COD 5850 mg/1; 1.38 kg/kkg (2.76 Ib/ton)
SS 900 mg/1; 0.19 kg/kkg (0.38 Ib/ton)
O&G 920 mg/1; 0.20 kg/kkg (0.40 Ib/ton)
BOD/COD 0.42
pH Range 5.5 to 11.9
Table 21 presents a summary table of means, minimums, maximums, sample
size, standard deviations and coefficients of covariance for tank car
cleaning operations from six edible oil refining operations. Table 22
presents a summary table of tank car cleaning wastewater characteristics
for each of the six plants investigated during this study.
Storage and Transfer
Another typical unit process waste load associated with all edible oil
refinery Subcategories A 5 through A 12 is that of wastewaters generated
during cleanup from storage, handling, and transfer areas within the re-
fining plant. Waste loads from these areas are highly variable and are
dependent on general daily cleanup necessitated by accidental spills,
leakage, or pump failures. Averaged waste load data from three plants
resulted in the following pollutant concentrations.
Production 314 kkg
Flow 75.7 cu m/day (0.02 MGD)
BOD 8,000 mg/1; 1.4 kg/kkg (2.7 Ib/ton)
COD 21,000 mg/1; 3.8 kg/kkg
SS 5,400 mg/1; 0.87 kg/kkg
O&G 4,200 mg/1; 0.69 kg/kkg
BOD/COD 0.51
pH Range 2.5 to 11.1
7.7 Ib/ton)
1.7 Ib/ton)
1.4 Ib/ton)
Table 23 presents a statistical description of the data compiled for this
study including mean, standard deviations, minimum, maximum, sample size,
and coefficients of covariance for the three plants investigated.
Refinery Floor Wash
Pollutant waste loadings result from general floor washing operations
necessitated by accidental oil spills and pump seal leakages. In general
these cleanup procedures are intermittent and represent a relatively minor
contribution to the total waste load of Subcategory A 5 plants.
311
-------
TABLE 21
A STATISTICAL DESCRIPTION OF THE WASTEWATER CHARACTERISTICS FOR
EDIBLE OIL REFINERY TANK CAR CLEANING OPERATIONS
to
VARIABLE N
Flow (MGD) 36
Prod, (ton/day) 36
BOD (mg/1) 30
SS (mg/1) 35
CCD (mg/1) 36
*FOG (mg/1) 36
BOD (Ib/day) 30
COD (Ib/day) 36
SS (Ib/day) 35
FOG (Ib/day) 36
Lb/Ton-BOD 30
kg/kkg-BOD 30
Lb/Ton-COD 36
kg/kkg-COD 3fc
Lb/Ton-SS 35
kg/kkg-SS 35
Lb/Ton-FOG 36
kg/kkg-FOG 36
BOD/COD Ratio 30
BOD/FOG Ratio 30
Flow Ratio 36
* FOG • Fats, Oils, and greases.
N = Number of data points
Note: Computer calculations for this table show no regard for significant figures.
MEAN
0.010124
163.552778
29U8. 033333
902.428571
5848.166667
923.111111
129.9fl796u
105.489276
66.91U197
67.941679
0.983145
0.491573
2.756511
1.379?56
0.360021
0.190011
0.406728
0.203364
0.4236P3
4.355555
57.827483
STANDARD
DEVIATION
0.008411
123.245906
4666.352162
941.755629
7336.449060
1164.397006
159.11^827
656.5210"!
129.190111
1 19.049493
1.676638
0.839319
4.820171
2.410066
0.494842
0.247421
0.61170U
C. 305852
0.247794
4.870516
52.679299
VARIANCE
0.0001
15169.5534
21961896.5651
666903.6639
53852834.6000
1355820.3873
25317.5281
431019.8776
16690.0648
14172.7817
2.8176
0.7045
23.2341
5.8085
0.2449
0.0612
0.3742
0.0935
0.061U
23.7219
2775.1085
MINIMUM
0.000600
75.000000
70.000000
e.ooooco
60.000000
3.000000
6.717725
5.756050
0.767740
0.2679C3
0.029657
0.014928
0.025591
0.012796
0,003412
0.001706
0.001260
0.000640
0.027057
0.362538
10.666667
MAXIMUM
0.048000
750.000000
16275.000000
3920.000000
31510.000000
4846,000000
643.960312
3552.967200
752.385200
536.249700
8.586404
4.293202
25.378337
12.689169
2.351204
1.175602
2.600914
1.400457
1.166667
23.333333
. 342.857143
COEFFICIENT
COVARIENCE
63.077
67.145 •
158.965
104.356
125.463
126.138
122.407
161.906
193.068
175.223
170.742
170,742
174,736
174.738
130.214
130.214
150.396
150.396
56.466
111.823
, .'J«P"
OF
(X)
-------
TABLE 22
POLLUTANT WASTE LOADINGS FOR EDIBLE OIL REFINERY TANK CAR CLEANING
CO
CO
Edible Oil
Refinery by
Process Code
75T05
75T06
75T08
75T09
75T10
Production
(kkg/day)
68
170
191
127
187
Volume of
Wastewater
Discharged
(cu m/day)
85.4
21.2
44.0
124.9
37.9
BOD
(mg/1)
1.6
0.37
0.40
0.31
0.13
COD
2.7
0.79
1.28
8.07
0.32
SS
(mg/1)
0.26
0.09
0.36
0.21
0.10
Oil &
Grease
(mg/1)
0.27
0.16
0.38
0.68
0.03
-------
TABLE 23
A STATISTICAL DESCRIPTION OF THE WASTEWATER CHARACTERISTICS FOR
EDIBLE OIL REFINERY STORAGE AND HANDLING OPERATIONS
-H
VARIABLE
Flow (MGD)
Prod, (ton/day)
BOD (mg/1)
SS (mg/1)
COD (mg/1)
*FOG (mg/1)
BOD (Ib/day)
COD (Ib/day)
SS (Ib/day)
FOG (Ib/day)
Lb/Ton-BOD
kg/kkg-BOD
Lb/Ton-COD
kg/kkg-COD
Lb/Ton-SS
kg/kkg-SS
Lb/Ton-FOG
kg/kkg-FOG
BOD/COD Ratio
BOD/FOG Ratio
Flow Ratio
N
21
21
19
20
18
19
19
18
20
19
19
19
18
18
20
20
19
19
17
17
21
MEAN
0.021924
314.736095
6173.664211
5426.300000
21606.111111
4223.473684
943.534452
. 2952.337077
544.060244
511.752952
2.717597
1.356799
7.661564
3.630792
1.743992
0.871996
1.390295
0.695147
0.506325
53.865362
62.437146
STANDARD
DEVIATION
0.041975
87.556213
22709.529317
13955.541929
76048.3*12459
12911.602032
3380.270853
11485.645406
1863.214956
1940.916493
8.654179
4.327090
29.190665
14.595433
5.072639
2.536419
4.932987
2.466493
0.114649
201.794U36
103.963429
VARIANCE
0.00
7666.09
515722721.78
194757150.54
5783350390.69
166709467.04
11426231.04
13192464U.73
.'3471569.97
' 3767156.83
74.89
18.72
652.11
213.03
25.73
6.43
24.33
6.08
0.01
40720.99
10808,39
MINIMUM
0.004800
144.000000 -
70.000000
40.000000
130.000000
10.000000
24.634440
37.772808
6.002855
1.251750
0,096730
0.048365
0.180065
0.090043
0.027463
0.013741
0.004622
0.002411
0.292500
0.307523
14.201183
MAXIMUM
0.173000
498.000000
99000.000000
55600.000000
326000.000000
56705.000000
14870.790000
.48966,460000
8351.676000
8517,656050
37.639160
18.919560
124.601679
62.300640
21.251084
10.625542
21.673430*
10,836715
0.675676
836.666667
, 397.701149
COEFFICIENT
COVARIENCE
191.458
27.819
277.837
257.089
348.746
305.710
358.256
389.042
342.465
379.268
316.450
318.450
381.003
361.003
290.875
290.875
354.816
354.816
22.683-'
374.488
J66.509
OF
(X)
* FOG = Fats, Oils, and greases.
N * Number of data points
Note: Computer calculations for this table show no regard for significant figures.
-------
DRAFT
Total Processing Effluent
On a daily basis the total waste load from the Subcategory A 5 plants may
be quite variable due to 1) differences in the raw materials processed;
2) the numbers of tank cars washed; and 3) the general cleanup pro-
cedures utilized to clean up accidental oil spills. Data compiled
for caustic refining, tank car cleaning, and storage and handling
indicate that flows and BOD concentrations vary greatly from day
to day as is indicated in the large standard deviations calculated
for these parameters in Table 17.
Model Plant
Based upon the data compiled for this study, a hypothetical model
plant for a caustic refinery operation was formulated. The following
assumptions were made for Subcategory A 5 plants:
1. The model plant is assumed to have a production of 454 kkg
(500 ton) per day.
2. The model plant has separate discharge of process waters
and non-contact cooling water.
3. The model plant has approximately five tank cars washed
per day. Each tank car has a capacity of 68 kkg (75 ton).
4. The model plant has a waste load generated from storage and
handling areas based upon a 454 kkg (500 ton) per day
production.
The following pollutant parameter waste loads were calculated for
Subcategory A 5 plants by assuming a linear relationship between
production and wasteload generation. For example, the average
waste loading for caustic refinery from the compiled data base was
as follows:
Suspended Oil and
Production Flow BOD COD Solids Grease
(kkg) (cum/day) (kg/kkg) (kg/kkg) (kg/kkg) (kg/kkg)
32071.9 1.01 1.81 0.51 0.61
The waste load for a caustic refining operation with a production of
454 kkg (500 ton) per day was then calculated by multiplying each
waste load by a factor of 1.42 (i.e., 454 kkg/320 kkg = 1.42). Thus,
the model plant was assumed to have the following waste load char-
acteristics for caustic refining:
Suspended Oil and
Production Flow BOD COD Solids Grease
(kkg) (cu m/day) (kg/kkg) (kg/kkg) (kg/kkg) (kg/kkg)
454102 1.43 2.57 0.72 0.86
315
-------
DRAFT
Historical and verification survey data compiled for storage and
handling was also converted to a production of 454 kkg (500 ton) per
day by multiplying by a factor of 1.59.
In addition, the model plant was assumed to wash five tank cars daily.
Therefore, tank car cleaning data was converted to a production of
320 kkg (375 ton) per day by multiplying by a factor of 2.05.
The total waste load characteristics for establishments engaged in
the caustic refining of edible oils was then calculated as indicated
in Table 24. Therefore, the wastewater characteristics for the
hypothetical model plant Subcategory A 5 are as follows:
Production: 454 kkg (500 ton) per day
Flow 314 cu m per day (0.083 MGD)
BOD 6,600 mg/1
COD 16,600 mg/1
SS 3,600 mg/1
0&6 3,500
pH 5.5 to 11.0
BOD Ratio 4.59 kg/kkg (9.18 Ib/ton)
COD Ratio 11.49 kg/kkg (22.98 Ib/ton)
SS Ratio 2.49 kg/kkg (4.98 Ib/ton)
O&G Ratio 2.39 kg/kkg (4.78 Ib/ton)
SUBCATEGORY A 6 - PROCESSING OF EDIBLE OILS BY THE USE OF CAUSTIC REFINING
AND ACIDULATION METHODS
The major process waste streams associated with Subcategory A 6 plants
are the same as those for Subcategory A 5 with the addition of acidulation.
Acidulation
The major waste loading unit process for the edible oil refinery industry
results from the acidulation process for the recovery of fatty acides
from the soapstock generated by caustic refining. Data collected
from four plants found average pollutant concentrations and waste loadings
for the acidulation process to be:
Production 486 kkg
Flow 223 cu m/day (0.059 MGD)
BOD 12,000 mg/1; 4.70 kg/kkg (9.39 Ib/ton)
COD 22,000 mg/1; 14.97 kg/kkg (29.94 Ib/ton)
SS 3,800 mg/1; 1.66 kg/kkg (3.32 Ib/ton)
O&G 2,500 mg/1; 1.20 kg/kkg (2.40 Ib/ton)
BOD/COD 0.57
pH range 0.6 to 3.0
Table 25 presents a statistical description of the data collected
from four refining operations. Table 26 provides a summary of average
wasteload values calculated for each plant investigated.
316
-------
TABLE 24
SAMPLE CALCULATIONS FOR DETERMINING TOTAL
WASTE LOADINGS FOR SUBCATEGORY A 5 PLANTS
Unit Process
Caustic Refining
Storage and Handling
Tankcar Cleaning
Flow
(cu m/day)
102.2
132.5
79.5
BOD
(kg/kkg)
1.43
2.16
1.00
COD
(kg/kkg)
2.57
2.83
6.09
SS
(kg/kkg)
0.72
0.39
1.38
0 & G
(kg/kkg)
0.87
0.41
1.11
Total subcategory
A5 Plant wasteload 314.2 4.59 11.49 2.49 2.39
-------
TABLE 25
A STATISTICAL DESCRIPTION OF THE WASTEWATER CHARACTERISTICS FOR THE
EDIBLE OIL REFINERY SOAPSTOCK ACIDULATION PROCESS
co
00
VARIABLE
Flow (MGD)
Prod, (ton/day)
BOD (mg/1)
SS (mg/1)
COD (mg/1)
*FOG (mg/1)
BOD (Ib/day)
COD (Ib/day)
SS (Ib/day)
FOG (Ib/day)
Lb/Ton-BOD
kg/kkg-BOD
Lb/Ton-COD
kg/kkg-COD
Lb/Ton-SS
kg/kkg-SS
Lb/Ton-FOG
kg/kkg-FOG
BOD/COD Ratio
BOO/FOG Ratio
Flow Ratio
N
43
43
30
40
35
42
30
35
40
42
30
30
35
35
40
00
42
42
25
29
43
MEAN
0.059456
536.183721
12075.600000
3855.150000
21741.657143
2508.904762
6625.366627
15rie. 169307
2518.823840
1667.210951
9.390904
4.695452
29.936496
14.966248
3.321062
1.660531
2.403090
1.201745
0.578246
77.669636
150.349585
STANDARD
DEVIATION
0.039224
506.130398
13817.169852
10049,563189
25933.071775
6437.980083
9074.811981
'21517.547446
6539.001880
4265.336623
10.154349
5,077174
46.676662
2'3. 439331
7.672406
3.836203
6.118084
3.059042
0.204302
153.556394
107.689029
VARIANCE
0.
256167.
190914182.
100993720.
672524211.
414«7587.
62352212.
463004648.
42758545.
16360109.
103.
25.
2197.
549.
58.
14.
37.
9.
0.
23579.
. . 11596.
MINIMUM
002
980
731
285
703
552
490
185
591
974
111
778
609
402
866
716
431
358
042
566
927
0
144
746
18
2730
5
224
800
6
1
0
0
2
1
0
0
0
0
0
0
36
,006600
.000000
.000000
.000000
.000000
.000000
.113320
.452400
.308620
.752050
.730011
.365005
.677707
.438854
.018077
.009038
.005021
.002511
.204041
,801289
.406420
MAXIMUM
0.
1666.
62100.
56986.
121000.
34570.
39365.
78256.
36142.
21924.
38.
19.
203.
101.
39.
19.
34.
17.
0.
673.
593.
226000
000000
000000
000000
000000
000000
062000
405800
929360
965400
595625
297813
792723
896362
977766
988883
318813
159006
948200
333333
750000
COEFFICIENT
COVARIENCE
65.971
94.395
114.422
260,679
119.278
256.605
132.957
139.560
259.605 .1
257.036
108.130
108.130
156.594
156.594
231.023
231.023
254.550
254.550
35.331
197.654
71.626 v
OF
(X)
* FOG = Fats, 011s, and greases.
N = Number of data points
Note: Computer calculations for this table show no regard for significant figures.
-------
CO
10
TABLE 26
POLLUTANT WASTE LOADINGS FOR THE EDIBLE OIL REFINERY ACIDULATION PROCESS
Edible Oil
Refinery by
Process Code
75A10
75A15
75A09
75A08
Production
(kkg/day)
279
337
345
1445
Volume of
Wastewater
Discharged
(cu m/day)
147.6
518.5
170.7
283.8
BOD
(mg/1)
1.29
18.43
7.05
6.06
COD
(mg/1)
3.09
59.4
11.07
10.72
SS
(mg/1)
0.25
6.34
0.45
2.91
Oil &
Grease
(mg/1)
0.23
5.31
0.04
1.86
-------
DRAFT
Model Plant
The hypothetical model plant described for Subcategory A 6 is assumed
to have a production of 454 kkg (500 ton) per day; to wash five tank
cars per day; and to have a separate discharge of process wastewater
and non-contact water. It is essentially the same plant described
in Subcategory A 5 with the addition of the unit process of acidulation.
By converting the acidulation data base to a 454 kkg (500 ton) per day
plant and adding this value to the model plant waste loads calculated for
Subcategory A 5 refineries, the following wastewater characteristics were
derived for Subcategory A 6 refineries.
Production 454 kkg (500 ton) per day
Flow 534 cu m/day (0.141 MGD)
BOD 7,600 mg/1
COD 21,600 mg/1
SS 3,400 mg/1
O&G 3,000 mg/1
pH range 0.6 to 3.0
BOD ratio 8.95 kg/kkg (17.90 Ib/ton)
COD ratio 25.41 kg/kkg (50.82 Ib/ton)
SS ratio 4.03 kg/kkg (8.06 Ib/ton)
O&G ratio 3.51 kg/kkg (7.02 Ib/ton)
SUBCATEGORY A 7 - PROCESSING OF EDIBLE OILS BY CAUSTIC REFINING.
ACIDULATION, OIL PROCESSING. AND DEQDORIZATION"
The individual unit processes and assumptions for the hypothetical
Subcategory A 7 refinery are identical to Subcategory A 6 plants with
the addition of the unit processes of deodorization and oil processing.
Deodorization
The contact cooling water blowdown generated from deodorization
barometric condenser units represents a major contribution to the total
waste load of an edible oil refinery. Average concentrations of pol-
lutants from six refining operations were a BOD of 4,000 mg/1, a COD
of 7,900 mg/1, a suspended solids of 730 mg/1, and oil and grease of
range was from 3.3 to 7.3. The average waste loadings were as follows:
BOD 2.21 kg/kkg (4.42 Ib/ton)
SS 0.32 kg/kkg (0.63 Ib/ton)
O&G 0.30 kg/kkg (0.60 Ib/ton)
BOD/COD 0.53
Table 27 presents a statistical description of the data compiled
for contact cooling tower blowdown from the six plants investigated.
Oil Processing
Oil processing includes the floor washing and general cleanup waste-
water discharges from the hydrogenation, winterization, bleaching,
320
-------
TABLE 27
A STATISTICAL DESCRIPTION OF THE WASTEWATER CHARACTERISTICS FOR
EDIBLE OIL REFINERY CONTACT COOLING TOWER SLOWDOWN FROM BAROMETRIC CONDENSERS
CO
ro
VARIABLE
Flow (MGD)
Prod, (ton/day)
BOD (mg/1)
SS (mg/1)
COD (mg/1)
*FOG (mg/1)
BOD Ob/day)
COD (lb/day)
SS (lb/day)
FOG (Ib/day)
Lb/Ton-BOD
kg/kkg-BOD
Lb/Ton-COD
ka/kka-COD
^y/ iN»Ny \rfUW
Lb/Ton-SS
ka/kka-SS
ivy/ Msy *j«j
Lh/Ton-FOG
UU/ 1 VII 1 1 \J*J
ka/kka-FOG
^y/ M>y r \j\j
BOD/CCD Ratio
\J\JU / \s\JD r\Q U 1 U
BOD/FOG Ratio
v\ju / i uu r\o \r lu
Flow Ratio
N
44
44
38
42
39
44
38
39
43
44
36
36
39
39
43
43
44 '
44
33
36
44
MEAN
0.047044
383.906*18
4061.8*6421
733.325581
7895.589744
758.522727
10S1. 815527
2288.022942
208.494655
255.700795
4.419554
2.209777
6.465459
4.242730
0.630163
0.315091
0.600192
0.300096
0.526206
15.911064
146.079543
STANDARD
DEVIATION
0.035721
291.499940
5341.475771
1267.460222
6771.247567
986.107676
1157.773150
206C. 853355
240.941 174
432.449137
7.015463
3.507732
11.469061
5.734540
0.734637
0.367319
0.633646
0.341923
0.164917
25.785414
100.262196
VARIANCE
0.0013 '
84972.2151 '
28531363.4147
1606455.4153
76934783.8799
972406.3483
1340438.6656
4247116.5517
. 58052.6496
187012.2563
49.2167
12.3042
131.5398
32.8650
0.5397
0.1349
0.4676
0.1169
0.0272
664.6676
10056,5169
MINIMUM
0.004320
. 96.500000
60.000000
15.000000
15S.OOOOCO
• 20.000000
15.P62176
33.687376
4.506300
15.321420
0.036878
0.019439
0.083057
0.041529
0,018902
0.009451
0.049039
. . 0.024520
'" 0,191964
0,776952
10.566235
MAXIMUM
0.173000
1144.000000
26700.000000
7960.000000
44520.000000
3683.000000
5793.099000
9659.504400
1129.245400
2494.771130
40.035238
20.017619
66.755386
33,377693
3,678324
1.839162
3.158095
i. 579047
0.660210
100.754717
607.017544
COEFFICIENT
COVARIENCE
75.932
75.930
131.503
172.837
111.090
130.004
110.074
90.071
115,562
169.123
158.737
158.737
135,162
135.162
116.575
116.575
113,938
113,936
31.341
162.059
67.722
OF
(X)
* FOG = Fats, 011s, and greases.
N ° Number of data points
Note: Computer calculations for this table show no regard for significant figures.
-------
DRAFT
and deodorization unit process operations. The wastewaters dis-
charged from these operations represent a relatively minor waste loading
in comparison to the other unit processes previously identified.
The average flow was 26.5 cu m/day (0.007 MGD). Average pollutant
concentrations were calculated as follows:
BOD 1800 mg/1
COD 5000 mg/1
SS 1100 mg/1
O&G 1300 mg/1
pH 7.3 to 13.0
Average waste loadings from oil processing were as follows:
BOD 0.09 kg/kkg (0.18 Ib/ton)
COD 0.22 kg/kkg (0.45 Ib/ton)
O&G 0.024 kg/kkg (0.48 Ib/ton)
The BOD/COD ratio was 0.49. Table 28 presents a statistical
description of the compiled data base from six plants.
Model Plant
Deodorization and oil processing data were converted to a 454 kkg
(500 ton) per day plant by the factors 1.32 and 1.16, respectively.
The waste loads from these unit processes were then added to the total
wasteload of the 454 kkg (500 ton) per day plant described for
Subcategory A 6 refineries. The following data represents the waste-
water characteristics of a Subcategory A 7 refining operation con-
sisting of the unit operations of caustic refining, acidulation,
deodorization, and oil processing:
Production 454 kkg
Flow 1147 cu m/day (0.303 MGD)
BOD 6.400 mg/1
COD 15,000 mg/1
SS 3,100 mg/1
O&G 1,500 mg/1
pH range 7.3 to 13.0
BOD ratio 16.09 kg/kkg (32.18 Ib/ton)
COD ratio 36.91 kg/kkg (73.82 Ib/ton)
SS ratio 7.84 kg/kkg (15.69 Ib/ton)
O&G ratio 3.93 kg/kkg (7.86 Ib/ton)
SUBCATEGORY A 8 - PROCESSING OF EDIBLE OILS UTILIZING CAUSTIC REFINING.
OIL PROCESSING, AND DEODORIZATION
Subcategory A 8 is essentially the same as Subcategory A 7 with the
deletion of the unit process of acidulation. As a result, the model
322
-------
TABLE 28
A STATISTICAL DESCRIPTION OF THE WASTEWATER CHARACTERISTICS FOR
EDIBLE OIL REFINERY OIL PROCESSING**
co
ro
CO
MEAN
0.006681
429.171029
1793.0R3333
1C89. 615385
5073.000000
1336.500000
67.421430
168.050SOU
34.652700
19. 170*568
0.185197
0.092599
0.446931
0.223465
0.093575
0.046787
0.047746
0.023874
0.485717
13.991234
25.060972
STANDARD
DEVIATION
0,005754
234.008594
3831.751806
1704.931550
6219.576185
1699.344161
175.611274
364.261552
61.722045
29.367554
0.455263
0.227632
0.945658
0.472829
0.160112
0.080056
0.077120
0.038560
0.226269
34.332616
24.822636
VARIANCE
0.0000
54760.0222
14682321.9015
2906791.5897
67561432.6667
2887770.5769
30839.6709
132686.4779
3809.6109
863.6263
0.2073
0.0518
0.89U3
0.2236
0.0256
0.0064
0.0059
0.0015
0.0512
1178.7266
616.1632,
MINIMUM
O.OOC330
238.400000
105.000000
35.000000
14.0.000000
2.000000
1.518790
13.791782
C. 841176
0.216970
0.002075
0.001038
0.018846
0.009423
0.001149
p. 000575
0.000910
0.000455
0,110123
0,351937
0.728711
MAXIMUM
0,013000
960.600000
13600.000000
5720.000000
.26400.000000
6000.000000
624.206000
1303.469000
161.386920
102.810400
1,625536
0.612768
3.394503
1.697251
0.472362
0,236181
0.267735
0.133868
0,890909
122,500000
54,530201
COEFFICIENT
COVARIENCE
86.122
54.526
213.696
156,471
162.026
126,959
260.469
216,757
177.094
153.292
245.626
245.826
211.589
211,589
171.106
171.106
161.513
161,513
46.584 j
245.387 '
99,049
OF
(X)
VARIABLE N
Flow (MGD) 14
Prod, (ton/day) 14
BOD (mg/1) 12
SS (mg/1) 13
COD (mg/1) 13
*FOG (mg/1) 14
BOD (Ib/day) 13
COD (Ib/day) 13
SS (Ib/day) 13
FOG (Ib/day) 1«
Lb/Ton-BOD 12
kg/kkg-BOD 12
Lb/Ton-COD 13
kg/kkg-COD 13
Lb/Ton-SS 13
kg/kkg-SS 13
Lb/Ton-FOG 1"
kg/kkg-FOG 1«
BOD/COD Ratio 12
BOD/FOG Ratio 12
Flow Ratio 1«
* FOG = Fats, Oils, and greases.
N » Number of data points
** Includes floorwashing and general cleanup for the following unit processes: hydrogenation, deodorizatlon,
bleaching, and winterization.
Note: Computer calculations for this table show no ragard for significant figures.
-------
DRAFT
plant for Subcategory A 8 will have a lower waste loading and flow then
Subcategory A 7 plants.
Model Plant
Assuming the same production rates and assumptions made for Sub-
category A 7 refineries, the model plant for Subcategory A 8 was
calculated to have the following concentrations and waste loading:
Production 454 kkg
Flow 927 cu m/day (0.245 MGD)
BOD 5,750 mg/1
COD 11,300 mg/1
SS 3,100 mg/1
O&G 1,400 mg/1
pH range 6 to 9
BOD ratio 11.73 kg/kkg (23.46 Ib/ton)
COD ratio 22.99 kg/kkg (45.98 Ib/ton)
SS ratio 6.30 kg/kkg (12.60 Ib/ton)
O&G ratio 2.81 kg/kkg (5.62 Ib/ton)
SUBCATEGORY A 9 - PROCESSING OF EDIBLE OILS BY THE USE OF CAUSTIC
REFINING. ACIDULATION. OIL PROCESSING. DEODORIZATION AND, THE
PRODUCTION OF SHORTENING AND TABLE OILS:
Subcategory A 9 is identical to Subcategory A 7 with the addition
of the plasticizing and packaging operations associated with a
shortening and table oils processing.
Shortening and Table Oil Production
Wastewater resulting from shortening and table oils plasticizing
and/or packaging operations are primarily generated from floor washing
and periodic equipment cleanup procedures. Wastewaters generated
from these operations represent a relatively insignificant waste
loading to the total refinery effluent. Average pollutant waste loads
for the production of shortening and table oils are discussed in detail
in Subcategory A 14.
Total Processing Effluent
Although the model plant for Subcategory A 9 has an additional unit
process waste stream its total waste load is observed to be less
concentrated than Subcategory A 7 due to the dilution effect attri-
butable to the relatively low waste load contributed by shortening
and table oil processing.
324
-------
DRAFT
Model Plant
The Subcategory A 9 model plant is assumed to be identical to the
Subcategory A 7 model plant with the addition of plasticizing
and packaging of shortening and table oils (i.e., Subcategory
A 14). The shortening and table oils packaging waste loads were
converted to a 454 kkg (500 ton) per day operation and were added
to the total waste load for Subcategory A 9. The wastewater char-
acteristics for Subcategory A 9 plants are as follows:
Production 454 kkg
Flow 1320 cu m/day (0.349 MGD)
BOD 5,900 mg/1
COD 13,500 mg/1
SS 3,000 mg/1
O&G 1,500 mg/1
pH range 3 to 9
BOD ratio 17.12 kg/kkg (34.24 Ib/ton)
COD ratio 39.15 kg/kkg (78.30 Ib/ton)
SS ratio 8.68 kg/kkg (17.36 Ib/ton)
O&G ratio 4.35 kg/kkg (8.70 Ib/ton)
SUBCATEGORY A 10 - PROCESSING OF EDIBLE OILS BY CAUSTIC REFINING.
OIL PROCESSING, DEODORIZATION, AND THE PLASTICIZING AND PACKAGING
OF SHORTENING AND TABLE "OILS"
The model plant developed for Subcategory A 10 is principally the
same as Subcategory A 9 with the deletion of the unit process of
acidulation.
Total Process Effluent
As a result of the deletion of acidulation, the total processing
effluent from Subcategory A 10 plants will be significantly reduced.
Model Plant
The model plant for Subcategory A 10 is identical to the Subcategory
A 9 model plant with the deletion of acidulation. The model plant
assumes a 454 kkg (500 ton) per day production for both the re-
fining operations and the filling and packaging of shortening
and table oils. The wastewater characteristics of Subcategory A 10
plants are as follows:
Production 454 kkg
Flow 1101 cu m/day (0.291 MGD)
BOD 5,250 mg/1
COD 10,400 mg/1
SS 3,000 mg/1
O&G 1,300 mg/1
325
-------
DRAFT
pH range 6 to 9
BOD ratio 12.76 kg/kkg (25.52 Ib/ton)
COD ratio 25.23 kg/kkg (50.46 Ib/ton)
SS ratio 7.14 kg/kkg (14.28 Ib/ton)
O&G ratio 3.23 kg/kkg (6.46 Ib/ton)
SUBCATE60RY A 11 - PROCESSING OF EDIBLE OILS BY CAUSTIC REFINING,
ACIDULATION. OIL PROCESSING. DEODORIZATION, AND THE PLASTICIZING AND
PACKAGING OF SHORTENING. TABLE OILS. AND MARGARINE
Subcategory A 11 is a combination of Subcategory A 7 (i.e., edible oil
caustic refining, acidulation, oil processing and deodorization) with
the addition of shortening, table oils, and margarine processing waste
load data presented in Subcategories A 13 and A 14. It is assumed that
the refining unit processes operate at a 454 kkg per day level. Sub-
category A 11 also assumes that the two additional unit processes (i.e.,
shortening, table oils packaging, and margarine packaging) operate each
at 227 kkg (250 ton) per day.
Total Processing Effluent
The total process effluent from Subcategory A 11 refineries represents
the highest pollutant wasteloading calculated for all the edible oil
refining model plants developed for this report.
Model Plant
It is assumed that the Subcategory A 11 plant has the same waste load
characteristics of Subcategory A 7, with the addition of: 1) a short-
ening, table oils plasticizing and/or packaging room and 2) a margarine
plasticizing and packaging room. Each packaging operation is assumed to
operate at a production rate of 227 kkg (250 ton) per day. The waste-
water characteristics of Subcategory A 11 plants are as follows:
Production 454 kkg
Flow 1574 cu m/day (0.416 MGD)
BOD 5,900 mg/1
COD 13,500 mg/1
SS 3,200 mg/1
O&G 2,800 mg/1
pH range 3 to 9
BOD ratio 20.57 kg/kkg (41.14 Ib/ton)
COD ratio 46.60 kg/kkg (93.2 Ib/ton)
SS ratio 10.98 kg/kkg (21.96 Ib/ton)
O&G ratio 9.95 kg/kkg (19.90 Ib/ton)
326
-------
DRAFT
SUBCATEGORY A 12 - PROCESSING OF EDIBLE OILS BY CAUSTIC REFINING,
OIL PROCESSING, DEQDORIZATION, AND THE PLASTICIZING AND PACKAGING"
OF SHORTENING. TABLE OILS. AND MARGARINE
Subcategory A 12 is identical to Subcategory A 11 with the deletion of
the unit process of acidulation. As a result, the final discharge from
the Subcategory A 12 plant will have a significantly higher pH and lower
pollutant waste load than Subcategory All.
Model Plant
The hypothetical Subcategory A 12 model plant is assumed to have the same
daily production rates, assumptions, and waste loadings per unit process
as the Subcategory A 11 model plant with the deletion of the unit process
for acidulation. The wastewater characteristics of Subcategory A 12 edible
oil refineries are as follows:
BOD 5,400 mg/1
COD 10,900 mg/1
SS 3,200 mg/1
O&G 3,200 mg/1
pH range 6 to 9
BOD ratio 16.20 kg/kkg (32.40 Ib/ton)
COD ratio 32.68 kg/kkg (65.36 Ib/ton)
SS ratio 9.44 kg/kkg (18.88 Ib/ton)
O&G ratio 8.83 kg/kkg (17.66 Ib/ton)
SUBCATEGORY A 13 - PLASTICIZING AND PACKAGING OF MARGARINE
Historical data submitted by the National Association of Margarine Manu-
facturers (NAMM) for four plants with supporting verification sampling
represents the data base compiled for Subcategory A 13 margarine pro-
cessing plants.
There are principally three sources of wastewater generated from mar-
garine plasticizing and packaging operations: 1) wastewater discharged
from margarine reclamation rooms; 2) wastewater discharged from general
floor washing operations containing detergents and chlorine; and 3) the
daily cleanup of CIP (clean-in-place) equipment utilizing the following
cleaning cycles: hot rinse, caustic wash, chlorine rinse, final rinse,
sanitation, and air drying. The amounts of wastewater generated from
these operations is primarily dependent upon the cleanliness and effi-
ciency of the above three operations. Margarine production requires
considerably more sanitation procedures than other edible oil finished
product packaging operations due to its ability to provide a growth
medium for pathogenic bacteria. As a result, cleanup operations of CIP
equipment and floor washing procedures require relatively larger volumes
of water. Average pollutant concentrations, flow, and production for
the four plants investigated were as follows:
327
-------
DRAFT
Production
Flow
BOD
COD
SS
O&G
pH
BOD ratio
COD ratio
SS ratio
O&G ratio
BOD/COD ratio
112 kkg
170 cu m/day (0.045 MGD)
1440 mg/1
4470 mg/1
900 mg/1
1760 mg/1
6 to 8
1.93 kg/kkg (3.86 Ib/ton)
4.22 kg/kkg (8.45 Ib/ton)
1.34 kg/kkg 2.69 Ib/ton)
2.86 kg/kkg (5.72 Ib/ton)
0.53
Table 29 presents a statistical description of the data base collected
indicating mean, standard deviations, and minimum and maximum values.
Table 30 presents the calculated averaged data for each of the three
plants investigated.
Total Processing Effluent
The total waste load resulting from a margarine processing operation in
combination with an edible oils refinery represents a significant waste
load to the total processing effluent. Based upon the data provided by
the NAMM, it is evident that the wastewater characteristics for margarine
processing is highly variable from plant to plant with higher waste loads
being correlated with larger production rates.
Model Plant
The hypothetical margarine processing plant for Subcategory A 13 was
assumed to operate at a production rate of 227 kkg/day (250 ton/day).
The wastewater characteristics for Subcategory A 13 plants are as
follows:
Production 227 kkg
Flow 340 cu m/day (0.09 MGD)
BOD 2600 mg/1
COD 5700 mg/1
SS 1800 mg/1
O&G 3900 mg/1
pH range 6 to 8
BOD ratio 3.92 kg/kkg (7.84 Ib/ton)
COD ratio 8.57 kg/kkg (17.14 Ib/ton)
SS 2.72 kg/kkg (5.44 Ib/ton)
O&G 5.81 kg/kkg (11.62 Ib/ton)
SUBCATEGORY A 14 - PLASTICIZING AND PACKAGING OF SHORTENING AND TABLE OILS
The plasticizing and packaging of shortening and table oils represents a
relative insignificant waste load in comparison to Subcategory A 13,
margarine processing. In general, shortening and table oils processing
328
-------
TABLE 29
A STATISTICAL DESCRIPTION OF THE WASTEWATER CHARACTERISTICS
FOR MARGARINE PROCESSING
co
ro
10
VARIABLE
Flow (MGD)
Prod, (ton/day)
BOD (mg/1)
SS (mg/1)
COO (mg/1)
*FOG (mg/1)
BOD (Ib/day) •
COD (Ib/day)
SS (Ib/day)
FOG (Ib/day)
Lb/Ton-BOD-
kg/kkg-BOD
Lb/Ton-COD
kg/kkg-COD
Lb/Ton-SS
kg/kkg-SS
Lb/Ton-FOG
kg/kkg-FOG
BOD/COD Ratio
BOD/FOG Ratio
Flow Ratio
N
32
32
25
31
19
30
25
19
31
30
25
25
19
19
31
31
30
30
17
25
32
MEAN
0. 04456!
123.100000
1437.940000
904.129032
4467.000000
1760.600000
653.976456
1352.209216
482.167910
972..3S2543
3.860014
1.930007
8.456009
4.226004
2.689364
1.344682
5.724866
2.862433
0.526437
4.143415
329.496866
STANDARD
DEVIATION
0.036737
66.897688
2175.084452
1273.862963
7270.608908
266«, 526167
2091.200180
2815. 409613
968.699670
2284.817523
8.123545
4,061772
11.268122
5.634061
4.818516
2.409258
11.221208
5.610604
0.2250C4
4.289584
139.962332
VARIANCE
0.0013
0475.3006
4730992.3733
1622726.8495
52861753.8669
7099699.6966
4373116.1909
7926503.1338
977527.0368
522039U1341
65.9920
16.4980
126.9706
31.7426
23.2181
5.6045
125.9)55
31.4789
0.0506
18.4005
19589.4544
MINIMUM
0.008000
30.000000
135.000000
24.0000CO
030,000000
22.0000CO
22.761850
119.592195
3.972220
4.836762
0.399981
0.199951
1.161089
0.580545
0,054639
0.027320
0,026416
0.013208
0.288925
0.367647
Ji8.5iesi9
MAXIMUM
0.111400
251.500000
11433.000000
4476.000000
32554,000000
11907.000000
10628.494089
11033. 8H077
3702.926850
10592.193339
42.260414
21,130207
63.672024
21.936012
19.235984
9.617992
51.795566
25.897783
0.976744
20.434783
615.384615
COEFFICIENT
COVARIENCE
82.443 '
54.344
151,262
140.894
162.763
151.342
3190766
208.206
205,044
234.971
210.454
210,454
133.256
133.256
179.169
179.169
196.006
1960e08
42,741
103.528 :
. 12.478
OF
(SO
* FOG • Fats, Oils, and greases.
N « Number of data points
Note: Computer calculations for this table show no regard for significant figures.
-------
TABLE 30
POLLUTANT WASTE LOADINGS FOR THE PROCESSING OF MARGARINE
CO
co
0
Edible Oil
Refinery by
Process Code
79M03
79M06
79M05
Production
(kkg/day)
134.0
119.4
63.9
Volume of
Wastewater
Discharged
(cu m/day)
219.5
176.7
59.4
BOD
(mg/1)
4.05
0.95
1.36
COD
(mg/1)
7.01
0.74
2.38
SS
(mg/1)
2.65
0.33
0.51
Oil &
Grease
(mg/1)
6.58
0.19
0.42
-------
DRAFT
employs strictly mechanical treatment of oils for the conversion of bulk
quantities of hardened oil into consumer sized packaging. The waste-
waters generated from these operations are principally from general
sanitation of filling and packaging equipment and general floor washing
procedures. The volume of water generated from the process is signifi-
cantly less than that for margarine processing due to the fact that
the finished products do not support bacterial growth and therefore re-
quire less rigorous sanitation procedures. The average pollutant con-
centrations furnished by the Institute of Shortening and Edible Oils
(ISEO) from five plants were:
Production 195 kkg (215 ton/day)
Flow 74.9 cu m/day (0.0198 MGD)
BOD 1600 mg/1
COD 4000 mg/1
SS 750 mg/1
O&G 770 mg/1
pH 6 to 8
BOD 0.48 kg/kkg (0.96 Ib/ton)
COD 0.19 kg/kkg (0.37 Ib/ton)
SS 0.18 kg/kkg (0.36 Ib/ton)
O&G 0.19 kg/kkg (0.36 Ib/ton)
BOD/COD ratio 0.52
Table 31 presents a statistical description of the means, standard
deviations, and minimum and maximum values calculated from the five
plants investigated and sampled. Table 32 presents a description of
the shortening data collected at each plant.
Model Plant
The hypothetical shortening and table oil processing model plant was
assumed to operate at a production level of 227 kkg (250 ton) per day.
The data base collected was converted to a daily production rate of
227 kkg by multiplying by a factor of 1.16 (i.e., 227 kkg/195 kkg =
1.16). The wastewater characteristics of Subcategory A 14 plants are
as follows:
Production 227 kkg
Flow 87 cu m/day (0.023 MGD)
BOD 1500 mg/1
COD 3000 mg/1
SS 1100 mg/1
O&G 550 mg/1
BOD ratio 0.56 (0.11 Ib/ton)
COD ratio 1.12 (2.24 Ib/ton)
SS ratio 0.42 (0.84 Ib/ton)
O&G ratio 0.21 (0.42 Ib/ton)
BOD/COD ratio 0.52
331
-------
TABLE 31
A STATISTICAL DESCRIPTION OF THE WASTEWATER CHARACTERISTICS FOR
SHORTENING AND TABLE OIL PACKAGING OPERATIONS
co
co
ro
VARIABLE
Flow (MGD)
Prod. (ton/day)
BOD (mg/1)
SS (mg/1)
COD (mg/1)
*FOG (mg/1)
BOD Mb/day)
COD (Ib/day)
SS (Ib/day)
FOG (Ib/day)
Lb/Ton-BOD
kg/kkg-BOD
Lb/Ton-eOD
kg/kkg-COD
Lb/Ton-SS
kg/kkg-SS
Lb/Ton-FOG
kg/kkg-FOG
BOD/COD Ratio
BOD/ FOG Ratio
Flow Ratio
N
24
24
22
23
16
24
22
16
23
2(1
22
22
16
16
23
23
20
2
-------
TABLE 32
POLLUTANT WASTE LOADINGS FOR SHORTENING AND TABLE OIL PROCESSING
co
CO
CO
Edible Oil
Refinery by
Process Code
79S05
79S08
79S09
79S06
79S17
Production
(kkg/day)
142
250
268
154
321
Volume of
Wastewater
Discharged
(cu m/day)
12.9
7.19
17.9
235.4
181.7
BOD
(mg/1)
0.13
0.046
0.11
1.51
__._
COD
(mg/1)
0.22
0.12
0.24
._._
ss
(mg/1)
0.052
0.041
0.12
1.03
____
Oil &
Grease
(mg/1)
0.056
0.052
0.067
0.28
1.87
-------
DRAFT
SUBCATEGORY A 15 - OLIVE OIL REFINING
The refining of olive oil is similar to the refining of other edible oils
except that it is done on a much smaller scale. The only wastewater gen-
erated is from caustic refining wash water with the following character-
istics:
Flow 1.13 cu m/day (0.003 MGD)
BOD 5700 mg/1
SS 296 mg/1
FOG 195 mg/1
Model Plant
Plant 79102 is the only olive oil refiner in the country using caustic
refining. Thus, the model plant is Plant 79102 and is illustrated in
Figure 111. The plant will have wastewater characteristics as listed
above.
BEVERAGES
SUBCATEGORY A 16 - NEW LARGE MALT BEVERAGE BREWERIES
In order to determine the wastewater characteristics of the malt beverage
industry, information was collected from several sources. The United
States Brewers Association (USBA) circulated one of two types of surveys
to all known breweries. They then produced a report entitled "1974 Brewery
Effluent Wastewater Characteristics" (56). Eleven breweries were visited
during the study and four breweries were sampled. An extensive literature
search was made to locate any existing historical data.
Process Waste Streams
As was noted in Section III, the sources of brewery waste can be identified
but the methods of disposal vary for each individual brewery. Further,
individual breweries may vary their methods of disposal based upon economic
or environmental factors. For these reasons there can be no fixed ranking
of the strengths of waste streams for the entire industry, however several
generalizations can be made.
Spent Grain Liquor - This is one of the most significant sources of waste
in the brewing process. It is essentially carbohydrate material, high in
BOD and suspended solids and low in pH. According to LeSeelleur (57)
average concentrations of BOD and suspended solids in spent grain liquor
are 15,000 mg/1 and 20,000 mg/1, respectively. At these concentrations
spent grain liquor, if discharged, can be expected to comprise 30 to 60
percent of the total plant load. As reported by Stein (58), spent grain
liquor from Plant 82B08IP9 represented 43.5 percent of the total pounds
of BOD and 60.3 percent of the total pounds of suspended solids. Most of
the breweries in subcategory A 16 do not discharge spent grain liquor.
Feed Recovery - These systems are operated by breweries not selling wet
spent grains or not disposing of spent grain liquor to sewers. As ex-
plained in Section III, the major wastewater discharge from feed recovery
334
-------
POOR QUALITY OLIVE OIL
COOLING
POND
DEGUM
>'
CAUSTIC
WASH
ACIDULATION
BLEACHING
PRESSURE
CLAY
FILTER
WINTERIZATION
OEODORIZATION
I
I
WASTEWATER TO
HOLDING TANK
CLAY SLUDGE
TO SOLID WASTE
REFINED OIL
FIGURE III
SUBCATEGORY A 15
OLIVE OIL CAUSTIC REFINING PROCESS
MODEL PLANT
335
-------
DRAFT
is the evaporator condensate. This is a high volume effluent with little
or no suspended solids and up to 300 mg/1 BOD. The concentration varies
from plant to plant depending on whether yeasts lost beer, or other wastes
are evaporated along with spent grain liquor. Wet scrubbers9 if utilized,
comprise a minor part of total feed recovery wasteload. Some plants
utilize cyclones, thus eliminating wet scrubber discharge. More than half
of the breweries in Subcategory A 16 operate feed recovery systems.
Lost Beer - This may represent from four to eight percent of beer produced.
Lost beer is primarily derived from packaging, fermentation, and finishing.
Since beer has a BOD concentration of approximately 125,000 mg/1, this can
account for a considerable part of the total plant load. Plant 82A02IP9,
for example, estimated beer loss at 40 percent of the total pounds of BOD
discharged per day. Assuming no recovery, a four percent beer loss would
amount to a BOD load of 5.02 kg/cu m (1.3 Ib/barrel). Four of the breweries
in Subcategory A 16 practice some form of beer recovery.
Spent Hops, Trub, and Yeast - These are grouped together simply because
their method of disposal may be similar. They are all suitable for
addition to spent grains since they contain only carbohydrates, protein
materials, yeast, and beer residues. None of the plants in Subcategory
A 16 discharge hops, trub, or yeast to sewers.
Filter Aid - This must either be hauled away to land disposal or sewered.
Considerable suspended solids would result were this waste to be dis-
charged, hence all but one of the plants in Subcategory A 16 recover
filter aid by decant tanks, vacuum, or pressure filters.
Alkaline Wastes - These are generated from vessel cleanup and bottle
washers.Residue from vessel walls is combined with caustic during
vessel cleanup. Paper labels, sodium aluminate from aluminum labels,
and glue are combined with caustic discharges from bottle washers.
Although alkaline wastes may be readjusted and reused, they are eventu-
ally sewered. Several plants in Subcategory A 16 meter caustic into
sewers from holding tanks.
Combined Process Flow
Data from 77 breweries were catalogued in the USBA wastewater character-
istics report. These brewers represented 87 percent of total sales for
the industry in 1973. Each brewery reported the ratio of flow (barrels),
BOD (lb), and suspended solids (lb)8 to production (barrels) for a full-
capacity day. A full-capacity day was defined as the maximum output which
could be sustained for a number of consecutive days. Each brewery was
assigned a reliability number based on the amount of accumulated data
and on sampling technique. Reliability numbers ranged from 0 to 10, with
the higher numbers corresponding to those breweries with more accurate
data. Breweries with reliability ratings of 8 to 10 were characterized
by continuous metering with short interval flow proportional sampling on
a daily basis for six or more months. Breweries with reliability ratings
336
-------
DRAFT
of zero had no data or data which would have an extremely high probability
of yielding misleading results. The year of initial construction and last
major expansion was presented for as many brewers as possible.
Based on the survey data, the arithmetic mean for all brewers was as
follows:
Flow Ratio 7420 1/cu m (7.42 Ib/bbl)
BOD Ratio 9.43 kg/cu m (2.44 Ib/bbl)
SS Ratio 3.83 kg/cu m (0.99 Ib/bbl)
Data for breweries in Subcategory A 16 are itemized and summarized in
Table 33. Scatter diagrams of flow, BOD, and suspended solids ratios versus
production for Subcategory A 16 are illustrated in Figures 112, 113 and 114.
Log normal probability plots of flow, BOD, and suspended solids ratios
are illustrated in Figures 115, 116, and 117.
Other significant parameters for combined process flow are pH, nitrogen,
and phosphorus. Several studies have documented the fact that pH may
vary widely over a 24 hour period. In fact, fluctuations of pH from 2
to 12 can be expected due to the batch nature of the brewing process.
In general, the pH of breweries in Subcategory A 16 can be expected to
remain between 5 and 11 due to the large number of compensating opera-
tions taking place simultaneously. Metering of caustic from holding
tanks can be expected to further buffer variations. Brewery waste is
known to be deficient in nitrogen. O'Rourke and Tomlinson (59) defined
an average BOD/N ratio of 43.2. Tests at Plant 82K32MP9 (60) established
a BOD/N ratio of 50.7. These appear to be representative of the industry
as a whole. Based on treatment systems in operation the waste appears
to contain adequate phosphorus.
In order to demonstrate the daily variability of brewery waste, the flow,
BOD, and suspended solids ratios for Plant 82A43 have been plotted for a six
month period as shown in Figures 118, 119, and 120. The means and standard
deviations are also given, It is obvious from these figures that treat-
ment system design and effluent limitations must take into account the
highly variable nature of brewery waste.
Model Plant
The raw waste loads for the model plant for Subcategory A 16 are based on
the mean ratios presented in Table 33. It should be noted that these
means were calculated by excluding the data from Plant 82A16. This plant
has demonstrated superior in-house waste reduction procedures. It was
felt, however, that the raw waste loads for this plant were not necessarily
economically achievable for the other brewers in this subcategory in their
present configurations. For treatment system design purposes, an average
production for this subcategory was calculated to be 1500 cu m (12,800
337
-------
DRAFT
TABLE 33
WASTEWATER CHARACTERISTICS
SUBCATEGORY A16
(NEW LARGE BREWERIES)
Plant
82A01
82A02
82A05
82B07
82B08
82A09
82A16
82B35
82A43
82B56
82A58
82A61
82A62
82A63
Flow Ratio
(I/cum)
4640
5200
7020
6850
9860
4970
1620
3550
4520
4600
5870
4860
4660
3730
BOD Ratio
(kg cu m)
9.62
7.88
10.40
12.90
17.40
7.38
1.74
9.00
8.81
11.00
15.00
8.58
11.20
7.57
SS Ratio
(kg/cu m)
3.48
3.40
2.40
4.41
8.35
5.06
1.08
2.24
2.98
3.94
4.68
2.78
3.59
2.94
Reliability
Number
10
8
8
7
5
8
10
5
10
7
8
10
9
9
Mean* 5410 10.50 3.86
(5.41 bbl/bbl) (2.72 Ibs/bbl) (1.00 Ibs/bbl)
Calculated without data from plant 82A16
338
-------
CO
co
vo
20000
13000
ICiOOO
•
14000
"=• 12000
3
O
^10000
+J
10
c:
£ 8000
6000
*000
2000
1C
3 T
It
• •
100
1000
10000
Production (cu m/day)
FIGURE 112
SUBCATEGORY A 16
RLQW.VS GARACJTY
-------
40
30
E
a 20
u
S
8
CO
10
s-
o .
10
100
1000
Production (cu ID/day)
FIGURE 113
SUBCATEGORY A 16
BOD VS CAPACITY
10000
-------
12
10
- 8
00
2.
10
100 1000
Production (cu in/day)
FIGURE 114
SUBCATEGORY A 16
SUSPENDED SOLIDS VS CAPACITY
10000
-------
100000
60000
40000
^ 20000
= 10000
* 8000
6000
4000
2000
5 10 20 40 60 80 90 95
Percent <_ Value Indicated
FIGURE 115
SUBCATEGORY A 16
FLOW PROBABILITY DIAGRAM
99
-------
30
u
o>
JO
-P»
CO
S 3
**
s
8
10 20 40 60 80 90 95
Percent <_ Value Indicated
FIGURE 116
SUBCATEGORY A 16
BOD PROBABILITY DIAGRAM
99
-------
30
OJ
-Pi
u
en
10 20 40 60 80 90 95 99
Percent <_ Value Indicated
FIGURE 117
SUBCATEGORY A 16
SUSPENDED SOLIDS PROBABILITY DIAGRAM
-------
25,000
20,000
£ 15,000
10,000
5,000
. ••".'" • ••'.-.'
J_
2 3
MONTH
FIGURE 118
DAILY FLOW VARIABILITY
PLANT 82A43
345
-------
38.7
34.8
30.9
27.1
23.2
o
C 19.3
O
s
15.5
11.6
7.7
3.9
3
MONTH
FIGURE 119
DAILY BOD VARIABILITY
PLANT 82A43
346
-------
34. e-
30. *•
27.1
at.z
S .9.3
It.5
11.6
7.7
3.9
_L
1234
MONTH
FIGURE 120
DAILY SUSPENDED SOLIDS VARIABILITY
347
-------
DRAFT
barrels) per day. Process waste and non-contact water are assumed to be
separated. Based on these assumptions, the model plant is defined as
follows:
Flow (MGD) 2.2
BOD (mg/1) 1900
SS (mg/1) 700
Total KN 40
pH 2 to 12
SUBCATEGORY A 17 - OLD LARGE MALT BEVERAGE BREWERIES
The methodology for determining wastewater characteristics for this sub-
category was the same as for Subcategory 16.
Process Waste Streams
Management questionnaires were available for three of the four brewers
in this subcategory. From the questionnaire responses and from plant
visits the following generalizations can be made. Due to the original
design of breweries in this subcategory there is a tendency for spent
wet grains to be sold and for spent grain liquor to be sewered instead
of evaporated. Spent hops, trub, and yeast are generally added to the
wet spent grains, while lost beer, filter aid, and caustic are usually
sewered.
Combined Process Flow
Data for breweries in this subcategory are itemized and summarized in Table
35. Scatter diagrams of flow, BOD, and suspended solids ratios versus pro-
duction are plotted in Figures 121, 122, and 123. Log normal probability
plots of flow, BOD, and suspended solids ratios are illustrated in Figures
124, 125, and 126. Age of facilities and efficiency of operation
result in higher raw wasteloads for this subcategory. Smaller tankage
is common, thus causing more water to be used in cleaning operations.
Collection and disposition of wastes is made more difficult by old and
intricate piping.
Model Plant
The raw waste load for this subcategory are based on the mean values
presented in Table 35. The average production was calculated to be
2600 cu m (2,200 barrels) per day. Process waste and non-contact water
are assumed to be separated. Based on these assumptions the model plant
is defined as follows:
Flow (MGD) 7.5
BOD (mg/1) 1700
SS (mg/1) 670
Total KN 35
pH 2 to 12
348
-------
JKMr
TABLE 35
WASTEWATER CHARACTERISTICS
SUBCATEGORY A17
(OLD LARGE BREWERIES)
Plant
82F04
82H36
82G46
82G64
Flow Ratio
(1/cu m)
14,700
9380
9870
10,200
BOD Ratio
(kg/cu m)
18.9
-
20.9
16.7
SS Ratio
(kg/cu m)
9.62
-
7.85
4.64
Reliability
Number
5
0
2
2
Mean 11,000
(11.0 bbl/bbl)
18.8
(4.87 Ibs/bbl)
7.34
(1.90 Ibs/bbl)
349
-------
co
in
o
20000
10000
IfiOOO
14000
1? 12000
3
O
^c 10000
*J
cl
£ sooo
cooo
2001
10
o
»—I
£
»—•
>
III
100
1000
Production (cu Hi/day)
FIGURE 121
SUBCATEGORY A 17
FLOW VS CAPACITY
10000
-------
CO
en
40
30
E
3
U
-x
o
*^r
O
•^
•«->
s.
20
10
10
100
1000
10000
Production (cu m/day)
FIGURE 122
SUBCATEGORY A 17
BOD VS CAPACITY
-------
CO
in
ro
12
10
•E 8
3
U
10
100 1000
Production (cu in/day)
FIGURE 123
SUBCATEGORY A 17
SUSPENDED SOLIDS VS CAPACITY
10000
-------
CO
tn
CO
100000
60000
40000
20000
10000
3000
6000
4000
2000
5 10 20 40 60 80 90 95
Percent £ Value Indicated
FIGURE 124
SUBCATEGORY A 17
FLOW PROBABILITY DIAGRAM
99
-------
30
3
U
co
en
-
£
I
10 20 40 60 80 90 95
Percent <. Value Indicated
FIGURE 125
SUBCATEGORY A 17
BOD PROBABILITY DIAGRAM
99
-------
30
E
o
CO
cn
01
.2 3
10
10 20 40 60 80 90 95
Percent <_ Value Indicated
FIGURE 126
SUBCATEGORY A 17
SUSPEWED SOLIDS PROBABILITY DIAGRAM
99
-------
DRAFT
SUBCATEGORY A 18 - ALL OTHER MALT BEVERAGE BREWERIES
The methodology for determining wastewater characteristics for this
subcategory was the same as for Subcategory A 16 and A 17.
Process Waste Streams
Management questionnaires were not completed by all breweries in this sub-
category hence no comprehensive analysis of methods of disposal is possi-
ble. The constituency of the process streams remains identical to that
described in Subcategory A 17.
Combined Process Flow
Eighty-five breweries are included in this subcategory. The 27 plants
not responding to the USBA survey form part of this group. Twenty-five
plants responding, but reporting no data, are also included in this
group. Data for those breweries responding is itemized and summarized
in Table 36. Only six of these breweries have a reliability rating of
four or higher. The standard deviation for the group is quite high,
indicating the lack of a definitive data base. Scatter diagrams of flow,
BOD, and suspended solids ratios verus production are plotted in Figures
127, 128, and 129. Log normal probability plots of flow, BOD and sus-
pended solids ratios are illustrated in Figures 130, 131, and 132.
Model Plant
The raw waste loads for the model plant are based on the 80 percent
values for this subcategory as presented in Table 36. This assumption
took into account the statistical variance of the group in addition to
the fact that those six plants with reliable data tended to exceed the
mean in several cases. The average production for this subcategory was
calculated to be 470 cu m (4000 barrels) per day. Process waste and
non-contact water are assumed to be separated. Based on these assumptions
the model plant is defined as follows:
Flow (M6D) 1.2
BOD (mg/1) 1400
SS (mg/1) 640
Total KN 30
pH 2 to 12
SUBCATEGORY A 19 - MALT
In order to determine the wastewater characteristics of this industry
a survey was conducted of all known maltsters. Three plants were visited,
two plants were sampled, and a search was made for any existing historical
data.
t
Process Waste Streams
As far back as 1935 Ruf (61) identified steeping and germinating as the
primary and secondary waste sources, respectively, from a malt house.
356
-------
DRAFT
TABLE 36
WASTEWATER CHARACTERISTICS
SUBCATEGORY A 18
PLANT
82059
82K32
82K44
82K50
82K55
82L03
82L10
82L14
82L17
82L20
82L21
82L23
82L24
82L25
82L26
82L27
82L28
82L29
82L33
82L40
82L42
82L45
82L47
82L48
82L57
82L60
82L65
82L68
82L74
82M12
82M13
82M15
82M18
82M19
82M22
82M30
82M31
82M34
82M37
82M38
82M39
FLOW RATIO
(1/cu m)
4300
920
6810
3710
4100
9700
4630
2820
12900
8990
5380
3110
3720
7190
760
18060
66820
1030
4160
1660
4690
6450
4810
6500
11730
21510
7650
2470
10860
3910
10750
5960
3010
1130
7460
1090
810
BOD RATIO
(kg/cu m)
9.05
2.20
9.82
7.50
7.42
15.08
1.66
6.88
19.34
1.66
28.35
10.71
8.62
5.99
5.76
8.97
2.55
5.41
1.70
1.66
14.93
14.31
4.10
0.66
5.88
15.47
5.45
3.40
10.32
____
0.04
2.32
SS RATIO
(kg/cu m)
1.62
0.27
6.26
1.93
2.55
5.03
1.04
2.94
9.40
0.19
0.89
4.95
2.98
1.47
1.24
5.88
1.66
1.55
1.01
0.54
4.76
6.96
-.81
0.39
15.82
6.46
2.63
0.66
4.14
_-__
0.12
3.48
RELIABILITY
NUMBER'
8
6
7
7
6
1
1
1
1
1
1
4
2
1
2
1
1
1
1
1
2
1
2
1
1
1
3
1
3
0
0
0
0
0
0
0
0
0
0
0
0
357
-------
DRAFT
TABLE 36 (CONT'D)
PLANT
FLOW RATIO BOD RATIO SS RATIO
(1/cu m)(kg/cu m) (kg/cu to)
RELIABILITY
'NUMBER
82M41
82M49
82M51
82M52
82M53
82M54
82M65
82M67
82M69
82M70
82M71
82M72
82M73
82M75
82M76
82M77
8310
9160
- TOO
9220
5.41
0.50
6.61
2.44
0.04
3.33
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
MEAN
80 Percent
Value
(7710 1/bbl) (8.47 1/bbl) (3.64 1/bbl)
10000 13.53 6.19
(lO.Obbl/bbl) (3.5bbl/bbl) (1.60bbl/bbl)
358
-------
en
20000
10000
16000
1400C
c 12000
3
U
10000
BOOO.
eooo
",000
200^
*c
in
100
1000
Production (cu n/day)
FIGURE 127
SUBCATEGORY A 18
FLOW VS CAPACITY
10000
-------
oo
o»
O
40
30
20
§
<° 10
10
LO
O
£ •
O
O
100
1000
10000
Production (cu n/day)
FIGURE 128
SUBCATEGORY A 18
BOD VS CAPACITY
-------
CO
12
10
- 8
o
5 6
10
CO
o
100 1000
Production (cu m/day)
FIGURE 129
SUBCATEGORY A 18
SUSPENDED SOLIDS VS CAPACITY
10000
-------
100000
60000
40000
20000
= loooo
3000
6000
4000
2000
5 10 20 40 60 80 90 95
Percent <_ Value Indicated
FIGURE 130
SUBCATEGORY A 18
FLOW PROBABILITY DIAGRAM
99
-------
30
CO
cr»
CO
5 3
8
••'
2 5 10 20 40 60 80 90 95
Percent <_ Value Indicated
FIGURE 131
SUBCATEGORY A 18
BOD PROBABILITY DIAGRAM
99
-------
30
c*>
.2 3
4->
-------
DRAFT
According to Isaac (62) the steep liquor is a strong, deeply colored,
putrescible liquid which may contain high levels of suspended solids.
The quantity and quality of steep liquor varies according to the number
of steep water changes and according to the contact time for each change.
In general, the strength of the waste (as measured by BOD) decreases ap-
proximately 75 percent from the first to the last steep. This is illus-
trated by data from Isaac (62) and Simpson (63) presented in Table 37.
Wastes from germination are known to'be smaller in volume and concentra-
tion than those from steeping, although insufficient data is available
to establish a specific proportion between the two.
Combined Process Flow
The significant parameters for this industry are flows BOD, and suspended
solids. The ratios of these parameters to the number of barley bushels
processed were calculated for each of the 18 plants which responded to
the industry survey. These responses are itemized and summarized in
Table 38. In addition, a reliability number was assigned to each plant
based on the method and duration of sampling as follows:
Reliability 1 - 24 hour flow proportional sampling for 5 con-
secutive days or more.
Reliability 2-24 hour flow proportional sampling for less
than 5 consecutive days.
Reliability 3 - Flow metered, grab samples.
Reliability 4 - Flow estimated, grab samples.
A separate arithmetic mean was calculated for those plants with reli-
ability numbers 1 and 2. A log mean was calculated to check the dis-
tribution of the. data.
In order to demonstrate the variability of malt waste, one plant was
selected which had conducted several periods of five-day, 24-hour, flow
proportional sampling. Table 39 gives the results of those tests with
the standard deviation for each measured parameter.
Malting effluents can be characterized as consisting of highly soluble
organic materials. Based on the even distribution of high reliability
plants throughout the spectrum of production in the industry, it is felt
that'the following levels are typical.
BOD Ratio 4.55 kg/kkg (0.218 Ib/bu)
SS Ratio 0.770 kg/kkg (0.0369 Ib/bu)
Flow Ratio 7410 1/kkg (42.6 gal bu)
The pH of the waste varies between 6.0 and 8.0 as reported by Isaac (62).
The waste is deficient in nitrogen, a fact which was confirmed by wet
sampling at plant 83A13.
365
-------
DRAFT
TABLE 37
ANALYSES OF MALTING STEEP WATER WASTES
BOD CONCENTRATION (mg/1)
Plant
Designation _A
1st Steep
2nd Steep
3rd Steep
4th Steep
960
920
185
254
1100
900
700
140
750
890
400
50
2800
2250
1900
490
1900
1630
1890
450
2750
1300
1800
870
366
-------
DRAFT
TABLE 38
RESULTS OF MALT INDUSTRY WASTEWATER SURVEY
PLANT
83A02IS9
83A07IP9
83A08IP9
83A09IS9
83A12IP9
83A13IS9
83A15IP9
83A19IS9
83A22IP9
83A25IP9
83A27IS9
83A28IS9
83A29IP9
83A30IP9
83A31IP9
83A32IP9
83A33IP9
83A34IP9
MEAN
(ALL)
MEAN
(1.2)
LOG-MEAN
(ALL)
FLOWR
BODR
SSR
11,800
8,780
682
7,080
6,240
6,960
6,240
4,430
6,180
9,700
10,800
31,100
11,300
5,690
4,580
4,190
5,570
5,210
8,140 1/kkg
46.8 gal/bu
7.29
4.41
0.459
6.05
3.74
5.29
2.72
3.43
3.52
4.03
14.59
3.66
5.44
3.16
2.93
2.92
3.37
4.84
4.60 kg/kkg
0.221 Ib/bu
0.446
1.45
0.0914
0.892
0.543
0.586
0.713
0.625
0.506
0.928
5.52
1.80
1.14
0.171
0.458
0.477
0.885
0.836
1.00 kg/kkg
0.048 Ib/bu
RELIABILITY
2
4
4
3
4
2
3
2
4
1
4
2
2
2
2
4
4
2
*7410 1/kkg 4.55 kg/kkg 0.770 kg/kkg
42.6 gal/bu 0.218 Ib/bu 0.036 Ib/bu
6460 1/kkg 3.70 kg/kkg 0.682 kg/kkg
37.1 gal/bu 0.177 Ib/bu 0.033 Ib/bu
Calculated without Plant 83A28 which had combined process and
cooling water.
367
-------
DRAFT
TABLE 39
DAILY VARIABILITY OF MALT WASTE
DAY FLOW FLOWR BOD BODR SS SSR
Wkkg) TJiG/1) TkgTkkg) WG/D TJg/kkg)
1
2
3
4
5
MEAN
0.365
0.373
0.365
0.378
0.444
0.385
9,210
9,420
9,210
9,560
11,200
9,700
485
475
300
370
451
416
4.43
4.44
2.74
3.51
5.01
4.03
92
59
125
90
113
95.8
0.843
0.552
1.14
0.850
1.26
0.928
STD.
DEVIATION 0.0334 79.1 25.3
368
-------
DRAFT
Model Malt Plant
For the purpose of developing control and treatment technology and for
conducting cost analyses a model plant has been designed. The model
plant for Subcategory A 19 operates 24 hours per day, 365 days per year.
It processes 350 kkg (16,000 bu) of barley per day based on the mean
production of those plants surveyed. Suspended solids in the waste,
consisting mostly of grain and sprouts, are assumed to be removed by
screening prior to discharge. Non-contact and process water are assumed
to be separated. Based on the above ratios the model plant has the fol-
lowing wastewater characteristics:
Flow (MGD) 0.685
BOD (mg/1) 615
SS (mg/1) 104
Total KN (mg/1) 17
Total P (mg/1) 7
pH 6 to 9
SUBCATEGORY A 20 - WINERIES WITHOUT STILLS
In order to determine the wastewater characteristics for the wine
industry (Subcategories A 20 and A 21) 11 wineries were visited, 5
wineries were sampled, and an extensive literature search was con-
ducted.
A short discussion of the methodology to be used in this section is re-
quired. Basically, a building block approach will be used. First, wineries
without stills will be described. Since many wineries in New York dis-
charge to navigable waters and since wineries in California do not,
the raw waste and effluent monitoring in New York were understandably
more extensive. For this reason wastewater characteristics for wineries
without stills rely heavily on New York data. Second, wineries with
stills will be described. These wineries are all located in California.
They produce the same wastewater as wineries without stills plus waste-
water associated with stillage. Since the characteristics of stillage
are fairly well defined, the total effluent for wineries with stills
during crushing will be the sum of the wastewater produced by distilling
added to the wastewater produced by wineries without stills. During the
processing season all wineries will be assumed to operate with the same
wastewater characteristics except as noted.
Process Haste Streams
The percentage of wastewater that each unit process contributes to the
total winery effluent has not been well documented. As identified in
Section III the sources of wastewater during processing are as follows:
lees, or washdown of filter presses or centrifuges with lees and filter
aid residue; fermenter washdown; finishing tank washdown; aging tank
369
-------
DRAFT
washdown; transfer hose, pipeline, and pump washdown; boiler and cooling
tower blowdown; water conditioning and regeneration rinses; and general
winery sanitation. During crushing, wastewater may be generated from all
of the above plus crusher/stemmer and pomace press washdown.
Combined Process Flow
It is recognized that wastewater characteristics differ during the crush-
ing and processing season. For that reason waste loading has been sepa-
rately correlated to kkg (tons) of grapes crushed and to cu m (gallons)
of wine produced. The ratio of flow, BOD, and suspended solids to grapes
crushed for four New York wineries is presented in Table 40. Three of
these four wineries have 24 hour flow proportional sampling with daily COD
and weekly BOD analyses. Based on the weighted mean for these wineries,
it is felt that the following ratios are typical for a winery without
stills during crushing:
Flow Ratio BOD Ratio SS Ratio
(1/kkg) (kg/kkg) (kg/kkg)
1528 3.57 1.16
(365 gal/ton) (7.14 Ib/ton) (2.32 Ib/ton)
It is noted that although these values are derived from New York wineries
they apply equally well to California wineries which are estimated to pro-
duce wastewater during crushing at 2.1 kg/kkg (4.2 Ib/ton) (64).
Wastewater generated during processing has been correlated to finished
wine produced. The flow, BOD, and suspended solids to wine produced ratios
are presented in Table 41. Based on the weighted mean for these wineries
it is felt that the following ratios are typical for a winery without
stills during processing.
Flow Ratio BOD Ratio SS Ratio
Q/cu m) (kg/cu m) (kg/cu m)
5510 6.63 2.33
(5510 gal/1000 gal) (55.3 lb/1000 gal) (19.4 lb/1000 gal)
Here again the values correlate to estimates from California wineries of
2.96 kg/cu m BOD and 0.6 kg/cu m suspended solids, since the production for
the New York wineries has been increased by amelioration and blending.
Other parameters which are significant for treatment system design are pH,
nitrogen, and phosphorus. In general the pH varies annually from 4.0 to
10.0 with a daily average of 7.9. Based on over 100 samples from plants
84*02 and 84*03 the waste can be characterized as deficient in both nitro-
gen and phosphorus. BOD/N ratios vary from 78:1 to 690:1 with those during
crushing being somewhat higher than those during processing. BOD/P ratios
remain fairly consistent between 162:1 and 208:1.
370
-------
DRAFT
TABLE 40
RAW WASTE CHARACTERISTICS DURING CRUSHING
WINERIES WITHOUT STILLS
Plant
84E01
84E02
84E03
84E04
Mean
Log Mean
Weighted**
Mean
Flow Ratio
(1/kkg)
1,970
7,290
1,087
1,090
1,380*
2,090
1,528
BOD Ratio
(kg/kkg)
3.42
4.96
2.88
3.03
3.57
3.64
3.57
SS Ratio
(kg/kkg)
1.47
1.57
0.44
0.32
0.95
0.76
1.16
Number
of
Samples
12
16
16
5
(365 gal/ton) >(7.14 Ib/ton) (2.32 Ib/ton)
* Calculated without plant 84E02 which has combined process
and cooling water.
** Excludes FLOWR and SSR for plant 84E04 due to method of
sampling.
371
-------
DRAFT
TABLE 41
RAVI WASTE CHARACTERISTICS DURING PROCESSING '
WINERIES WITHOUT STILLS
Plant
84A01
84A02
84A03
84A04
Mean
Log Mean
Weighted**
Mean
Flow Ratio
(1/cu m)
7,280
12,400
2,940
1,290
3,840*
4,300
5,510
.5,510 gal.
BOD Ratio
(kg/cu m)
14.1
6.35
6.91
30.4
14.4
11.7
6.63
,_55.3 Ib .
SS Ratio
(kg/cu m)
4.70
1.52
0.79
4.05
2.76.
2.19
2.33
, 19.4 Ib v
Number
of
Samples
36
47
65
5
* Calculated without plant 84E02 which has combined process
and cooling water.
** Calculated without plant 84A04 due to size and method of
sampling. Labor calculated without plant 84A01 due to in-
plant reduction required (See Section VII).
372
-------
DRAFT
Model Plant
For the purposes of control and treatment technology and cost analysis a
model plant has been designed. The production for wineries without stills
during crushing is 180 kkg (200 tons) per day based on average operating
levels for New York and California wineries. The production during pro-
cessing is 41 cu m (10,800 gal) of finished wine based on the average for
New York wineries. It is recognized that this figure may be a little
higher than California wineries without stills due to the practice of
New York wineries to blend in up to 25 percent of California wines; i.e.,
a typical California production during a 70 day season would be 25 cu m
(6,730 gal). Based on this production level the raw waste loads for the
model plant are as follows:
Crushing Processing
Flow (MGD) 0.0730 0.060
BOD (mg/1) 2300 1200
SS (mg/1) 760 420
Total KN mg/1) 7 4
Total P (mg/1) 13 -7
pH 4 to 10 4 to 10
The following process operations are assumed:
1) Stems are considered a solid waste to be spread on vineyard
property.
2) Pressed pomace may be used for distilling material, may be
spread on vineyard property, or recovered as a by-product.
3) Diatomaceous earth (filter aid) is considered a solid waste to
be spread on vineyard property.
4) No distilling takes place on premises.
5) Final effluent is screened to remove solids.
SUBCATEGORY A 21 - WINERIES WITH STILLS
As previously described, the wastewater for wineries in this subcategory
will be the same as that for wineries without stills, plus the wastewater
associated with stillage.
Process Waste Streams
As explained in Section III the raw material for distillation may be lees,
pomace, or wine. Although stillage characteristics will vary with the
373
-------
DRAFT
source of distilling material, it can generally be classified as a high-
strength organic waste with low pH. Typical values for different types
of stillage are reported in Table 42 ( 65).
In order to determine the wastewater effluent due to stillage an average
volume and concentration must be defined which will apply equitably to
the wineries with stills. In order to calculate the average flow, data
from 19 California wineries with stills was obtained to determine the
average amount of distilling material produced per ton of grapes crushed.
This data is itemized in Table 43 (66). Based on this average of 746
1/kkg (179 gal/ton) the total quantity of still age produced would be
the amount of distilling material increased by 15 percent due to steam
introduced in the still. The average volume of stillage per unit of
grapes crushed, therefore, is 853 1/kkg (206 gal/ton). As acknowledged,
the concentration of stillage varies depending on the type of distilling
material used. Table 44 presents data from Skofis (67) and wet sampling
at plant 84C80 which has been used to verify the ranges of values ex-
pected. In both cases 24 hour flow proportional samples were taken for
five or more days. Based on these data and that presented in the litera-
ture (68, 69) it is felt that typical values for stillage are as follows:
BOD (mg/1) 12,000
SS (mg/1) 14,000
By combining these values with the flow volume of 858 1/kkg (206 gal/ton)
the ratios (pounds of pollutant to tons of grapes crushed) contributed by
stillage are:
BOD 10.3 kg/kkg (20.6 Ib/ton)
SS 12.0 kg/kkg (24.0 Ib/ton)
Combined Process Flow
The total effluent during crushing for a winery with stills, then, is a
combination of stillage and crushing wastes as shown below:
Due to Crushing Due to Stillage Total
Flow 1528 1/kkg 859 1/kkg 2390 1/kkg
(365 gal/ton) (206 gal/ton) (571 gal/ton)
BOD 3.57 kg/kkg 10.3 kg/kkg 13.9 kg/kkg
(7.14 Ib/ton) (20.6 Ib/ton) (27.7 Ib/ton)
SS 1.16 kg/kkg 12.0 kg/kkg 13.6 kg/kkg
(2.32 Ib/ton) (24.0 Ib/ton) (27.3 Ib/ton)
As evidenced by these calculations, stillage contributes 36 percent of the
flow and 74 percent of the BOD ,and suspended solids in winery waste during
374
-------
DRAFT
TABLE 42
STILLAGE CHARACTERISTICS
Total Solids
Volatile Solids
Suspended Solids
BOD
Total Acidity (CaC03)
pH
Total N
Total P
Conventional
Still age
(mg/1)
20,100
87.4
3,120
11,000
3,170
4.7
271
11,150
Lees
Still age
(mg/1 )
68,000
86.5
59,000
20,000
9,870
3.8
1,532
4,284
Pomace
Still age
(mg/1 )
13,180
77.0
18,700
2,400
1,220
3.7-6.8
330
1,310
375
-------
TABLE 43
DISTILLING MATERIAL PRODUCED PER TON OF GRAPES CRUSHED
Tons of Grapes Gallons Distilling
Gallons Distilling
Materi "" Per
Plant
(A)
(B)
(C)
(D)
(E)
(F)
(G)
(H)
(I)
(J)
(K)
(L)
(M)
(N)
(0)
(P)
(Q)
(R)
(S)
TOTAL
Received
79,633
58,448
53,514
34,187
50,488
39,769
24,480
208,603
45,909
17,846
131,381
27,822
113,050
34,520
28,869
26,800
25,920
6,296
24,762
1,032,297
Material Produced
21,659,432
22,532,405
7,898,299
6,768,676
14,292,949
10,334,742
7,460,034
12,271,927
10,275,021
2,620,889
33,995,334
5,061,980
7,323,023
8,148,219
4,207,165
2,080,224
2,931,093
1,701,137
3,164,583
184,727,132
Ton of arapes
270.968
385.500
147.592
197.99
283.00
259.868
304.74
58.828
223.812
146.86
258.754
181.93
64.776
236.043
145.732
77.620
113.082
270.193
127.799
184 7?7
1 032 297 = 178-947 wine Gallons Distilling Material
5 ' per Ton of Grapes Received
376
-------
DRAFT
TABLE 44
STILLAGE CHARACTERISTICS
PLANT 84C80
Day
1
2
3
4
5
6
7
8
Average
BOD
(mg/i)
6,650
14,400
9,620
11,100
10,300
12,000
18,300
7,650
11,300
Day
1
2
3
4
5
Average
SS
(mg/i )
23,100
11,400
10,200
10,700
4,060
13,400
33,200
10,300
14,500
DATA FROM
BOD
(mg/1)
12,008
14,211
9,925
13,864
13,650
12,732
N
(mg/i)
369
380
184
185
182
268
203
231
250
SKOFIS
SS
(mg/i )
5,289
3,784
6,084
2,096
2,916
4,033
P
(mg/i)
321
321
204
273
242
308
425
209
288
£H
3.98
3.92
3.89
3.82
3.95
3.91
2H
3.8
3.8
3.8
3.9
3.8
3.8
3.9
3.9
3.8
377
-------
DRAFT
crushing. A winery with stills is assumed to have the same wastewater
loads during processing as a winery without stills.
Model Plant
For the purposes of control and treatment technology and cost analysis a
model plant has been designed. The production for wineries with stills
during crushing is 700 kkg (776 tons) per day based on the average of
the 19 wineries itemized in Table 43 over 70 days in 1974. The pro-
duction per day during a 70 day processing season is estimated to be
91 cu m (23,900 gal).
Based on these production levels the raw waste loads for the model plant
are as follows:
Crushing Processing
Flow (MGD) 0.422 0.132
BOD (mg/1) 5830 1210
SS (mg/1) 5750 424
Total KN (mg/1) 103 4
Total P (mg/1) 494 7
pH 3.5 to 6 4 to 10
SUBCATEGORY A 22 - GRAIN DISTILLERS OPERATING STILLAGE RECOVERY SYSTEMS
In order to determine the wastewater characteristics for the distilled
spirits industry (Subcategories A 22, A 23, A 24 and A 25) 59 plants were
contacted to obtain existing historical data, 13 plants were visited,
3 plants were sampled, and a complete literature search was conducted.
Process Waste Streams
Extensive unit process research has been conducted in distilled spirits
plants since the 1930's. As a result of this research, process waste
streams can be defined quite accurately. Table 45 illustrates
the percent of total plant wasteload attributable to each unit process
as reported by plants 85A01 and 85A13 (70 and 71 ). These figures
are reported merely to establish a general hierarchy of waste loading
that is felt to be representative of the industry. Additional waste
reduction measures employed by these plants since testing are reported
in Section VII.
Feed Recovery - The major source of wastewater within the feed recovery
plant is evaporator condensate. Evaporator condensate flows will vary
based on mash concentration in the fermenters, percent of "backset,"
and beer still dilution. By reducing beer gallonage toward 110 1 (28 gal)
per bushel, the liquid load to the evaporators is reduced. By increasing
the percent of "backset," less water must be added to obtain a given beer
378
-------
TABLE 45
PROCESS WASTE STREAMS
GRAIN DISTILLERS WITH STILLAGE RECOVERY
Subcategory A 22
Percent of Total Waste Load
Feed Recovery
Cooking-Mashing
Recti fyi ng-Bottl i ng
Distilling
Fermenting
Power House
Domestic
TOTAL
Plant
85A01
79
12
4
1
1
2
1
100
Plant
85A13
75
11
8
2
1
2
1
100
379
-------
gallonage. By using a reboiler, either internal or external to the still,
the amount of liquid added to spent still age is reduced in comparison
with liquid added by sparging live steam. The flow for evaporator
condensate might vary between 97 1 (15 gal) and 79 1 (21 gal) per
bushel. The concentration of the condensate (as measured by BOD in
mg/1) varies mainly according to the design and operation of the
evaporator. Data presented by Rullman in Table 46 (72)illustrates the
range of values that might be expected. As reported by Hurst (73)
Plant 85A04 has achieved BOD concentrations of 300 mg/1 using a
mechanical recompression evaporator. Results of other tests (74 and 75)
indicate that BOD concentrations of from 600 to 800 mg/1 are generally
representative of the industry. An analysis of evaporator condensate
from Plant 85A01 is shown in Table 47'. The main constituents of
the condensate are free organic acids, volatile with steam at reduced
pressure and, hence, not included in total solids figures.
Both barometric and surface condensers are being used on evaporators in
the industry. Flows for these discharges might vary between 380 and
570 1 (100 and 150 gal) per bushel. BOD concentrations for barometric
discharges are generally quite low depending on the quality of the
intake water. The temperature of these discharges as they leave the
plant might range from 83° to 99°C (180° to 210°F).
Barometric discharges are currently separated from process wastestreams
and routed to surface waters. Wet scrubber discharge containing parti-
culate from drum and/or grain dryers may constitute the secondary waste-
load from feed recovery operations. Before their elimination, Plant 85A13
estimated these discharges at 37 percent (70) of the population
equivalent for feed recovery. Several plants have installed cyclones
to recover particulate for addition to feeds, thereby eliminating the
wasteload.
Mashing - Cooking - Mash pressure cooking in batch cookers with
vacuum cooling to malting temperature will produce approximately
7.6 1 (2 gal,) of condensate per bushel of grain mashed. Analyses at
Plant 85A23 (72) indicate this condensate may average 900 mg/1
BOD. The flow from continuous cookers would be some what higner.
Here again, both barometric and surface condensers are employed
in the industry. As indicated in Section III, cooling may be by shell
and tube heat exchanger, thus reducing the wasteload.
Cooking vessel cleanup must also be taken into account. In most plants
the mash will simply be washed to the following cook, then cleaned
with caustic during the weekend. Unpumpable mash which is low in
volume but high in BOD and suspended solids concentration will in-
evitably be sewered.
Rectifying - Bottling - As described in Section III, the potential
wastes generated from rectifying are fusel oil column tails and rec-
tifying column tails. A balance sheet for Plant 85A22 operating at
200 kkg (7200 bushels) per day with a grain neutral spirits unit
380
-------
TABLE 46
VARIABILITY IN BOD CONCENTRATION OF
GRAIN DISTILLERY EVAPORATOR CONDENSATE
SUBCATEGORY A 22
Type of Evaporator
A. Standard short tube vertical
type, triple effect and fin-
ishing pan, natrual cir-
culation, basket type
separators.
B. Same as above.
C. Same as above.
D. Same as above.
E. Standard vertical, quadruple
effect and finishing pan.
Forced circulation in 4th
effect and finishing pan.
Centrifugal separators.
F. Standard vertical, short tube,
triple effect and finishing
pan, natural circulation.
Baffle type separators.
G. Vertical long tube, outside
colandria, triple effect
and finishing pan. Natural
circulation.
H. Same as above, after larger
vapor bodies and basket type
separators and automatic level
controls installed.
Remarks
Operated at maximum
capacity. Automatic
level controls.
BOD
(mg/V
675
Operated at 3/4 capacity 570
Operated at 1/2 capacity 510
Operated at 1/2 capacity, 800
manual level controls.
1/2 capacity. Automatic 650
level controls.
Full capacity. Manual 1580
controls.
3/4 capacity. Semi- 3210
Auto level controls.
3/4 capacity. Automatic 520
level controls.
381
-------
TABLE 47
ANALYSIS OF GRAIN DISTILLERY EVAPORATOR CONDENSATE
SUBCATEGORY A 22
Biochemical Oxygen Demand (BOD) 1,100
Total Solids (8.0 gms/100 1) 80
Ethyl Alcohol (0.04 Proof) - 135
Aldehydes Trace
Esters Trace
Organic Acids, calc. as acetic 550
Fusel Oil, AOAC - - Less than 10
Nessler Nitrogen 12
pH - - - 3.6
382
-------
DRAFT
consisting of aldehyde, rectifying, fusel oil, and vacuum columns is
presented in Table 48 (72). As jnpted, discharges at the plant
from the fusel oil column amount to approximately 7.6 1 (2 gal) per
bushel at 35 to 40 mg/1 BOD, and discharges from the rectifying column
approximately 38 1 (10 gal) per bushel at 300 mg/1 BOD. It should be
noted that at these rates the waste load from rectifying would comprise
more than four percent of the total plant load. Sobolov (76)
indicates that gin process residues would be higher due to the
presence of spent botanicals.
Bottling wastes, consisting of glue, paper, and alcohol, appear to be
negligible.
Distilling - As noted in Section III, possible sources of waste from
distilling are doubler discharge and beer still cleanup. If the doubler
discharge is sewered then the approximate flow for a doubler raising
proof from 115° to 130° would be calculated as follows (72):
One Bushel Mashed = Five Proof Gallon (Yield)
Five Proof Gallon =4.35 Wine Gallons at 115 Proof
= 3.85 Wine Gallons at 130 Proof
One Bushel Mashed =0.5 Gallons to Sewer
At these rates BOD concentratforTmay be from 5000 to 6000 mg/1.
Beer still cleanup discharges will of course vary throughout
the industry. Plant 85A01 has estimated 10,300 1 (8000 gal) at
1500 mg/1 BOD for a weekly water and caustic wash. Plant 85A30
(77) reports 38,000 1 (10,000 gal) at 2500 mg/1 BOD once per
week.
Fermenting - Steam and water are normally employed to wash the
mash into the still during processing. Weekend cleanups here again,
will comprise the major discharge. Plant 85A01 (74) has
estimated fermenter wash at 3800 1 (1000 gal) and 1000 to 2000 mg/1
BOD.
Cleanup - Wastes from weekend cleanups are normally estimated and averaged
on a daily basis to determine total plant discharges. In order to es-
tablish the general nature and magnitude of weekend cleanups, data is
presented in Table 49 (74). These loads would vary from plant
to plant according to operating procedure, numbers and types of equip-
ment, and plant design.
Combined Process Flow
The significant parameters for this industry are flow, BOD, and suspended
solids. The ratios of these parameters to bushels mashed were calcu-
lated for 16 plants. In addition, a reliability number was assigned
to each plant based on plant visits and the method and duration of
raw waste sampling as follows:
383
-------
DRAFT
TABLE 48
BALANCE SHEET FOR GRAIN NEUTRAL SPIRITS UNIT
GRAIN DISTILLERY
SUBCATEGORY A 22
Wine Gallons
Jji (Per/Hour)
High Wine Feed (115° Proof) 1285
Aldehyde Column Steam (5,000 #) 600
Rectifying Column Steam (13,000 #) 1560
Fusel Oil Column Steam (3,000 #) 360
Fusel Oil Column Dilution Water 125
3930
Out
Concentrating Column Heads 70
Rectifying Column Product (190° Proof) 700
Fusel Oil Column Tails to Sewer 527
(35 to 40 ppm BOD)
Rectifying Column Tails to Sewer 2633
(300 ppm BOD)
3930
384
-------
DRAFT
TABLE 49
POLLUTION LOAD FROM WEEK-END CLEANUPS
GRAIN DISTILLERY
SUBCATEGORY A 22
Source
Unpumpable mash
Venturi fermenter wash
Beer still caustic
Gin still drop
Mash line caustic
Evaporator water wash
Conveyor water wash
Centrifuge water wash
Flow
(gal)
80
1,000
8,000
3,500
9,000
29,000
300
8,000
BOD
(mg/1)
20,000
1,100
1,500
5,300
3,200
1,600
1,600
900
SS
(mg/1)
10,000
450
1,600
1,500
600
27,000
700
385
-------
DRAFT
Reliability 1 - 24 hour flow proportional
sampling for three or more months.
Reliability 2-24 hour flow proportional sampling
for at least one week.
Reliability 3 - Flow metered, grab samples.
Reliability 4 - Flow estimates, grab samples.
Reliability 5 - Plant estimates.
This data is itemized and summarized in Table (50). A separate
arithmetic mean was calculated for those plants with reliability
numbers 1 and 2. As reported in Table (50) the means are as
follows:
Flow Ratio BOD Ratio SS Ratio
(1/kkg) (kg/kkg) (kg/kkg)
All Plants 5572 3.95 2.57
(37.5gal/bu) (0.221 ]b/bu) (0.1441b/bu)
Plants (1,2) 6582 6.01 4.23
(44.3gal/bu) (0.3761b/bu) (0.237 Tb/bu)
Table 51 demonstrates the daily variability in distillery waste
as presented by Glower (77), 24 hour flow proportional composites
were taken for ten consecutive days in a large size plant with combined
process and cooling water.
Other parameters significant for treatment system design are pH,
nitrogen, and phosphorous. pH can be expected to fluctuate between
5 and 11 over a 24 hour period. The waste is deficient in nitrogen
and phoshporous. Based on tests conducted at Plant 85A01, (78)
probable levels of nitrogen and phosphorous are 1.59 kg (3.5 Ib) and
0.136 kg (0.3 Ib) per 45.5 kg (100 Ib) of BOD.
Model Plant
For the purpose of developing control and treatment technology and for
conducting cost analysis a model plant must be designed. Based on the
even distribution of plants with reliability numbers 1 and 2 through-
out the spectrum of the industry, it is felt that the following raw
waste ratios are typical:
Flowr 6500 1/kkg (43.7 gal/bu)
BODR 6.00 kg/kkg (0.33 lb/bu)
SSR 4.23 kg/kkg (0.237 lb/bu)
Since the range of production in the industry is large, two model
plants have been designed on the above ratios. Production for the two
plants was set at 380kkg (15,000 bushels) and 90 kkg (3500 bushels)
386
-------
DRAFT
TABLE 50
WASTEWATER CHARACTERISTICS
GRAIN DISTILLERY
SUBCATEGORY A 22
Plant
85A01
" 02
11 04
11 05
11 07
11 08
11 13
11 15
11 17
11 18
„ 22
11 23
11 26
11 27
11 29
11 30
MEAN
MEAN
(1,2)
Flowr
(1/kkg)
8120
3680
6770
3480
3450
3550
3690
5730
7230
7360
4050
70,500
7560
6120
7420
77,100
5572*
(37.5gal/bu) (0.
6582*
(44.3 gal/bu)(0.
BODR
(kg/kkg)
7.88
1.79
4.07
1.08
1.17
6.25
6.97
1.72
2.64
2.22
3.97
2.16
6.14
2.98
5.63
6.49
3.95
221 Ib/bu)
6.01
336 Ib/bu)
SSR
(kg/kkg)
3.52
.545
1.44
.463
.711
.592
6.18
3.56
3.53
1.43
5.77
.536
5.16
2.57
(0.144 Ib/bu)
4.23
(0.237 Ib/bu)
Reliability
1
3
4
4
3
5
2
4
4
4
3
3
2
2
2
2
* Averaged without 85A23 and 85A30 which have combined process & cooling water
387
-------
DRAFT
TABLE 51
DAILY VARIATIONS IN RAW WASTE
GRAIN DISTILLERY
SUBCATEGORY A 22
Flow
(MGD)
7.90
8.10
7.84
7.53
7.63
7.79
7.77
8.58
7.64
7.09
MEAN 7.92
STANDARD
DEVIATION
SS
(mg/1)
120
100
84
31
21
138
54
35
41
45
67
SS
(Ib/bu)
Ow504
0.480
0.349
0.138
0.082
0.573
0.248
0.159
0.184
0.168
0.308
5.5 kg/kkg
0.179
BOD
(mg/1)
40
109
53
121
124
100
58
48
91
100
84
BOD
(Ib/bu)
0.168
0.521
0.221
0.539
0.502
0.415
0.267
0.218
0.369
0.373
0.359
6.4 kg/kkg
0.136
388
-------
DRAFT
per day. Screening is assumed to remove grain solids prior to discharge.
Based on these assumptions the raw waste loads for the model plants are
as follows:
380 kkg 90 kkg
Flow (MGD) 0.650 0.150
BOD (mg/1) 930 950
SS (mg/1) 650 670
Total KM(mg/1) 33 33
Total P (mg/1) 3 3
pH 5 to 11 5 to 11
SUBCATEGQRY A 23 - GRAIN DISTILLERS NOT OPERATING STILLAGE RECOVERY
SYSTEMS
The methodology for determining the wastewater characteristics for this
subcategory was the same as for Subcategory A 22.
Process Waste Streams
Process streams are assumed to have the same characteristics as those
in Subcategory A 22 with the following exceptions:
Feed Recovery - Distilleries in this subcategory may operate in one of
two modes:T) wet spent stillage may be collected in holding tanks and
sold as cattle feed; 2) wet spent stillage may be screened, with solids
recovered by drying, and thin stillage collected in holding tanks for
sale as cattle feed. Since the load from evaporator condensate is non-
existent, the wastewater discharge is greatly reduced compared to dis-
tilleries in Subcategory A 24.
Rectifying-Bottling - Many distillers in this subcategory may produce
only straight whiskey. Wastes associated with multi-column operation
would therefore be eliminated, but doubler discharge would remain the
same. Also, whiskey may be shipped in bulk after maturation, thus
eliminating bottling discharges.
Combined Process Flow
Less data exists for these distilleries due to the fact that many either
sell to farmers or discharge their wastes to sewers. Data obtained from
three plants is presented below:
Flow Ratio BOD Ratio SS Ratio
Plant I/ kkg) (kg/kkg) (kg/kkg)
85B04 1530 0.628 0.736
85B28 1830 0.593 0.533
85B29 1975 1.62 0.101
Mean 1780 0.947 0.634
389
-------
DRAFT
The mean for the suspended solids ratio was calculated without data from
plant 85B29 due to sampling error.
As expected, these ratios are approximately 70 percent less than the
ratios for distilleries with complete feed recovery systems.
Model Plant
For the purpose of developing control and treatment technology and for
conducting cost analysis, a model plant has been designed. The daily
production for the subcategory was set at 50 kkg (2000 bushels). Based
on these assumptions the raw waste loads for the model plant are as
follows:
Flow 90.8 cu m/day (0.024 MGD)
BOD (mg/1) 210
SS (mg/1) 160
Total N 7
Total P 1
pH 5 to 11
SUBCATEGORY A 24 - MOLASSES DISTILLERS
In order to determine the wastewater characteristics for this subcate-
gory all known rum distillers were contacted. Two plants were visited
and a complete search of the literature was conducted.
Process Streams
Areas of wastewater generation in the rum distilling process are 1) spent
molasses stillage, 2) boiler and cooling waters and fermenter washdowns,
3) barrel washings and analytical laboratory wastewaters, and 4) bottling
wastes. Table 52 outlines the wastewater generated per proof gallon
produced as well as the percent contribution by type of waste stream
(79).
Spent Stillage
This stream accounts for approximately 66 percent of the waste flow, over
98 percent of the BOD and COD, and over 90 percent of the solids generated.
The chemical constituents can fluctuate depending on the variability of
ash and sugar contents of the molasses feed and the degree of acidifi-
cation prior to fermentation. Table 53 demonstrates the variability
of spent stillage based on the type of molasses used. Such variations
appear to have only minor effects on waste treatability. In addition,
both cane and citrus molasses are used by distillers in the United States.
According to plants 85C43 and 85C44 (78, 79) these raw materials also
produce no noticeable difference in wastewater effluent. Typical chemical
analyses and ionic concentrations of rum slops are presented in Tables 54
and 55 respectively (79). It should also be noted that the temperature
of this waste stream ranges from 80 to 90° C (165 to 220° F) with a dark
brown color of approximately 100,000 units.
39D
-------
TABLE 52
MOLASSES DISTILLERY WASTE STREAMS
co
vo
Waste
Parameter
or
Constituent
Total
Facility Waste
Generation
per
Proof Gallon
Volume
COD
BOD
Total Solids
Total Dissolved
Solids
Total Suspended
Solids
Total Kjehldahl
Nitrogen
Total Phosphate
55.6 1 (14.7 gal)
3.0 kg (6.6 Ib)
1.0 kg (2.3 Ib)
4.2 kg (9.2 Ib)
3.9 kg (8.6 Ib)
0.25 kg (0.56 Ib)
0.06 kg (0.14 Ib)
0.003 kg (0.007 Ib)
% Contribution by Type of Waste Stream
Slops Boiler/Cooling Barrel Water Treatment
Stream Water & Per- Washings & Analytical Lab,
menter Washdown Wastewaters
66%
98%
99%
91%
91%
97%
100%
100%
26% 5% 3%
1% 1%
1%
9%
9%
3%
_._ — —
___ ___ __ _
-------
DRAFT
TABLE 53
VARIABILITY OF MOLASSES STILLA6E
Type of Molasses
pH
Total Solids %
Insoluble solids %
Ash %
Total nitrogen %
Reducing substances (as
invert sugar) %
Ca %
Sulphate (as S04) %
5-day BOD p.p. 100,000
Cuban
High Test
3.5
2.81
0.25
0.42
0.06
0.13
870
Cuban
Low Test
4.2
7.12
0.68
2.3
0.13
1.0
0.26
0.52
1,950
392
-------
DRAFT
TABLE 54
CHEMICAL CHARACTERISTICS OF MOLASSES STILLAGE
Parameter Mean Range
Soluble COD (mg/1) 72,000 67,100 - 75,700
92,000 81,100 - 106,300
Total COD (mg/1) 74,800 71,500- 78,900
99,800 83,800 - 115,500
Soluble BOD (mg/1) 26,500 17,600 - 32,300
47,400 40,600 - 57,500
Total BOD flng/1) 32,900 19,800- 41,900
54,900 45,800 - 67,000
Alkalinity (mg/1 as CAC03) 912 806- 1,320
Volatile Acids (mg/1 as HAc) 4,920 3,610- 5,920
pH 4.36 4.28 - 4.45
Solids (mg/1)
Total 83,500 70,200 - 95,800
. total fixed 20,500 19,400 - 22,200
. total volatile 63,000 50,700 - 73,600
Total dissolved 77,700 77,400 - 85,600
. fixed dissolved 19,800 17,900 - 21,500
. volatile dissolved 57,900 45,600 - 64,000
Total suspended 6,220 2,540 - 10,280
. fixed suspended 800 40 - 1,720
. volatile suspended 5,400 2,500 - 9,620
Nitrogen (mg/1 as N)
.total KjehldahT U40 790- 1,450
.organic 1,060 770 - 1,280
Total Orthophosphate (mg/1
as P04 93 59 - 98
393
-------
DRAFT
TABLE 55
IONIC COMPOSITION OF MOLASSES STILLAGE
(Units of mg/1)
Constituent Mean Range Observations
Zn
Cd
Pb
Fe
Na
Cu
Co
Mn
Ca
Mg
Cr
K
AT
Cl
S04
9.89
0.18
1.10
81.0
372
32.8
0.60
10.6
2088
824
0.30
4259
0.38
2110
4120
2.38
0.09
0.77
42.0
209 -
2.0
0.19
2.38
1850 -
391
0.25
4011 -
0.10
1330 -
3500 -
- 19.93
- 0.32
- 1.60
- 150.0
523
- 124
- 0.76
- 15.6
2476
1728
- 0.33
4845
- 0.58
4400
4800
4
4
4
5
5
5
4
4
4
5
4
5
4
4
3
394
-------
DRAFT
The amount of mixing water added to the raw molasses can affect the
strength of the still age. When the water to molasses ratio is decreased,
a smaller volume of stillage results. Therefore, in order to minimize
cost, most rum distillers maintain a low water to molasses ratio.
The addition of NH4 and PCty nutrients appears to have little effect
on the wastewater characteristics. It is assumed that nearly 100
percent of the nutrients added are utilized by the yeast cells during
fermentation.
The use of indirect heat rather than live steam in the still also re-
sults in a lower volume of stillage. The total pollutant load remains
the same. The use of direct heat would result in a 15 to 30 percent
reduction in water usage. Only one rum distiller is currently in
the process of converting from live steam to indirect heating.
The unique solubility properties of calcium sulfate (gypsum), one of
the major components of rum stillage, has an impact on the treatability
of the slops stream. Unlike most compounds, gypsum becomes less soluble
with increased temperatures. Therefore, the formation of scale is an
important consideration, especially for evaporation.
Boiling/Cooling Water and Miscellaneous Washes - Boiling/cooling waters
can represent 20 to 25 percent of the total flow from a rum distillery.
Most of the wasteload is in the form of suspended and dissolved solids
(less than 10 percent of the RWL) resulting from solubility changes
due to the temperature fluctuations. Cooling water is used on a
non-contact basis to decrease the temperature of the molasses prior to
fermentation. Boiling water is used in pasteurization of the molasses
prior to cooling and fermentation. Such water is non-contact and
usually recycled, thus explaining the minor role in pollutant loadings.
Further uses of boiling/cooling waters are similar to those of the grain
distillation processes.
Washdown of fermenters usually is sent to the still with the "wort."
Some plants may follow with a caustic wash cycle which is then either
discharged or regenerated for future washings. Other plants use a
detergent wash cycle which is directly sewered. The initial holding
tanks for molasses seldom require washing since they are rarely empty.
A rinse once a year would be an exceptional case.
Barreling Operations - These operations involve a minimum of water usage
(approximately 1.3 gallons of water/gallon of rum). Since alcohol laws
for rum production permit the use of used oak barrels for aging, the
barrels are washed after usage. The resultant wasteloads are small
amounts of dissolved materials which have migrated to the inside
surface of the barrel during maturation. These wastes are washed off
the barrels at the barreling site and disposed of directly. Further
reuse of such wastes has not yet been explored.
395
-------
DRAFT
Bottling - Due to the similarities in bottling operations between grain
distillers and rum manufacturers, the resulting waste loads are assumed
Identical.
Combined Process Flow
All known existing data was collected in order to determine combined
process flows. The ratios were calculated for flow, BOD, and suspended
solids to proof gallons produced and are presented in Table (56). Other
parameters requiring consideration are pH and temperature. pH averages
4.8 and temperature 100° C (212° F) due to the high percentage of
stillage in the waste.
Model Plant
The production of the model plant is 30,000 proof gallons per day, based
on the mean of those plants in Table 56. It is assumed that stillage
is discharged without treatment and that process and cooling water are
separated. Based on these assumptions the raw waste loads for the
model plant are as follows:
Flow (MGD) 0.216
BOD (mg/1) 35,600
SS (mg/1) 6,720
Total KN 1,110
Total P 55.3
Temperature 212° C
pH 4.8
Color 100,000 units
SUBCATEGORY A 25 - BOTTLING AND BLENDING OF BEVERAGE ALCOHOL
Plants in this subcategory exist as an adjunct to those beverage alcohol
producers described in Subcategories A 20, A 21, A 22, A 23, and A 24.
The methodology for determining wastewater characteristics for bottlers,
therefore, was an extension of that used for the abovementioned subcate-
gories.
Process Waste Streams
As described in Section III, these plants may only bottle beverage alcohol
produced in wineries and distilleries, or they may additionally redistill
and rectify purchased liquors in order to manufacture such products as
cocktails and cordials. The wastes involved are those from redistilling,
rectifying, and bottling. In order to demonstrate the general nature of
these wastes, data will be presented from plant 85D10, a large rectifier/
bottler. Although this is not intended to represent the typical wastes
for the entire spectrum of the industry, it does identify unit process
wastes that may be common to other bottler/rectifiers.
Redistilling - Both vodka and gin are products which may be redistilled.
The residue from redistillation constitutes the major waste associated
with this segment of the process. Heads from continuous column distil-
lation and bottoms from batch distillation are collected in a holding tank.
396
-------
DRAFT
TABLE 56
RAW WASTE CHARACTERISTICS
RUM DISTILLERS
Flow Ratio
BOD Ratio
SS Ratio
Plant
85C34
85C38
85C39
85C45
MEAN
(I/proof gal)
25.7
28.8
255.0
378.0
27.3
(7.22 gal/pg)
(kg/proof gal)
0.997
0.922
1.40
0.557
0.969
(2.136 Ib/pg)
(kg/proof gal)
0.149
0.206
0.265
0.110
0.183
(0.392 lb/pg)
*Excludes Plants 85C39 and 85C45 which reported process and
cooling water combined.
397
-------
DRAFT
On an average of once per month 3000 1 (800 gal) of this liquid must be
discharged. The approximate BOD is 245,000 mg/1, the suspended solids only
6.3 mg/1, and the pH 5.6. If discharged at once, this would represent a
shock load of 7400 kg (16,300 Ib) of BOD. The residue from redistillation
can amount to one percent of the input to the still, but this waste does
not necessarily relate to total plant production since some alcohol used
is not redistilled. A correlation may be possible if linked to the vodka
and gin production for each plant.
Rectifying - The types and volumes of wastes from rectifying for plant
85D10 are listed below.
Type Volume I/day
Frame Filter Rinse 5700 (1500 gal/day)
Product Chiller Rinse 600 (160 gal/day)
Vodka Column Rinse 2500 (650 gal/day)
Product Tank, Filter, Line
and Pump Rinse 7200 (1900 gal/day)
Bonded Warehouse Rinse 4000 (1000 gal/day)
Winery Rinse 950 (250 gal/day)
Demineralizer Regeneration 1900 (500 gal/day)
These wastes generally contain only dilute portions of alcohol that have
adhered to surfaces during processing, except for demineralizer regeneration.
Periodically, the demineralizing resins must be recharged by washing with
caustic and acid. These are presently collected and neutralized before
discharge.
Bottling - These wastes consist mainly of filler cleanup and miscellaneous
floor washing. Filler discharge will obviously vary depending on the
number of fillers, number of product changes, and volume used. Glue and
paper labels may also contribute to the load.
Bad Product - A small quantity of bad product is destroyed periodically
due to the product not meeting quality standards or being discontinued.
These are crushed in bottles with the liquid being sewered. This may
amount to as much as 10,000 wine gallons per year, however it may vary
greatly depending upon the amount of new product activity or package
changes that occur.
Combined Process Flow
The combined process flow consists of biodegradable liquids with little
or no suspended solids. The flow may vary from 1900 1 (500 gal) per day,
for those small plants only bottling, up to 40,000 1 (10,000 gal) per day
for large rectifier/bottlers. For the most part these flows will be low
in BOD concentration due to dilution factors. Heads from redistillation
and bad product discharges may, however, be quite concentrated depending
on the method of disposal. There is no existing data available concern-
ing combined process flow.
398
-------
DRAFT
It should be noted that considerable non-contact water may be used in
large rectifying/bottling plants. Compressor and redistillation column
cooling water comprises the vast majority of this flow. Plant 85D10 gave
the following breakdown of water usage:
Percent
Sanitary Waste 5.9
Industrial Waste 5.7
Non-Contact Discharge 67.6
Boiler Water 3.9
In Product 16.9
Total 100.0
Model Plant
For purposes of cost analysis and treatment system design a model plant
must be designed. The following assumptions have been made:
1. Residue from redistillation may amount to one percent of the
input to the still. For plants with redistillation this
waste is assumed to be collected in holding tanks.
2. Bad product may accrue and periodically require disposal. This
product is assumed to be collected and held prior to disposal.
3. Demineralizers may be used, requiring periodic regeneration.
This is assumed to be collected in holding tanks and neutral-
ized.
4. All other process wastes are assumed to be separated from non-
contact water. The process wastes are assumed to result from
washdowns previously itemized and to be biodegradable with low
concentrations of BOD and suspended solids.
Based on these assumptions two model plants have been designed. Plant A
is assumed only to bottle. Plant B is assumed to rectify and bottle.
The raw wasteloads are as follows:
A B
Flow (cu m/day) 4 40
(MGD) 0.001 0.010
SUBCATEGORY A 26 - SOFT DRINK CANNERS
In order to determine the wastewater characteristics for the soft
drink industry (Subcategory A26 and A27) 74 plants were contacted
by phone, eight plants were visited, and five plants were sampled.
In addition, a complete literature search was conducted.
399
-------
DRAFT
Process Waste Streams
The major waste streams associated with soft drink canners are filler
spillage, mixing tank washing, and fill tank and line washings.
Filler Spillage - Due to the type of container and the speed of the
line there is considerable spillage involved in canning operations.
This product waste may be characterized as high in BOD, total solids,
and acidity, and low in suspended solids and pH. Expected ranges are
as follows:
BOD (mg/1) 60,000-80,000
Total Solids (mg/1) 100,000-120,000
Suspended Solids (mg/1) 50-200
Acidity (mg/1) 1,200-3,200
pH 2-3.5
Mixing Tank Washing - Mixing room wastes originate from the small
residue of syrup dumped during flavor changes and the water required
to wash the mixing tanks. Syrup used for carbonated beverages may be
as high as 800,000 mg/1 BOD. When diluted with wash water this waste
has the same character as filler spillage, but it is lower in concentration
and higher in pH.
Fill Tank and Line Washings - These wastes, again, correlate closely
to the number of flavor changes. A small amount of syrup, and water
to flush the filling lines, is the source of waste. The character
of the waste is the same as that from the mixing tanks.
Other Wastes - Additional waste may be created by washing bulk con-
tainers, periodic washing of syrup storage tanks, water treatment and
filtration backwash, and plant cleanup. These are considered to be
minor process discharges. Boiler and compressor cooling water comprise
the majority of the non-contact water.
Combined Process Flow
In order to demonstrate the combined waste characteristics from soft
drink canners, one plant has been selected which conducted twenty-four
hour sampling over a period of more than five days. The results are
presented in Table 57. As expected the BOD concentrations were high,
but the ratio of pounds of BOD to gallons produced was quite low. This
is explained by the low flow discharged in conjunction with a high
volume of production. The pH of the waste was below six, indicating
the presence of low pH product in the waste.
Based on the average of all canners surveyed it is felt that the
following ratios are typical.
400
-------
TABLE 57
DAILY WASTE CHARACTERISTICS
SOFT DRINK CANNING PLANT
Plant 86A27
Flow BOD
-P.
o
Day
1
2
3
4
5
6
Averag
Flow
(MGD)
0.033
0.031
0.036
0.035
0.037
0.031
e 0.034
Ratio BOD Ratio
(gal /I ,000 gal) (mgl) (lb/l,000gal
281
277
305
280
296
253
282
1650
960
1140
1160
790
1480
1197
3.86
2.22
2.89
2.70
1.94
3.13
2.79
SS
1) (mgl)
154
177
118
192
219
376
206
SS
(lb/1,000 gal)
0.36
0.42
0.30
0.45
0.54
0.79
0.48
J*L
5.9
4.3
3.5
2.9
4.6
3.5
(282 1/cu m)
(.335 kg/cu m)
(0.057 kg/cu m)
-------
DRAFT
Flow = 741 1/cu m (741 gal/1000 gal)
BOD = 1.02 kg/cu m (8.51 lb/1000 gal)
SS = 0.123 kg/cu m (1.03 lb/1000 gal)
The pH is expected to vary between 3 and 7 except during periods of
cleanup when alkaline wastes will be discharged. Based on sampling at
Plants 86A32 and 86A29 the effluent appears to be somewhat deficient in
nitrogen but adequate in phosphorus for purposes of treatment. BOD:N
ratios averaged 60:1, while BOD:P ratios were 110:1.
Model Plant
For the purposes of control and treatment technology and cost analysis
a model plant has been designed. The production was set at 309 cu m
(81,500 gal) per day. Based on this production and the ratios listed
above, the raw waste loads for the model plant are as follows:
Flow (MGD) 0.0610
BOD (mg/1) 1380
SS (mg/1) 167
Total KN (mg/1) 23
pH 3 to 7
SUBCATEGORY A 27 - SOFT DRINK BOTTLING/CANNING PLANTS
The methology for determining the wastewater characteristics for this
subcategory was the same as for Subcategory A 26.
Process Waste Streams
The major waste stream associated with bottling plants is the bottling
washer. Mixing tank and filler line washdown is expected to be similar
to that from canning plants as previously discussed.
Bottle Washer - As described in Section III, the sources of pollutants
from the bottle washer and sugar residues from left-over product, sus-
pended solids from labels and material left in bottles, and caustic
carry-over from sprays and oaking tanks. Typical values for prerinse
and final rinse sections of a bottle washer taken at Plant 86A32 were
as follows:
Prerinse Final Rinse
BOD (mg/1) 1130 35
SS (mg/1) 76 28
M Alkalinity (mg/1) 263 206
pH 10.3 10.3
The flow associated with this washer was 230 1/min (60 GPH).
402
-------
DRAFT
Combined Process Flow
The final discharge of any plant with a bottle washer will thus be
higher in flow, pH, and alkalinity than a plant which only cans.
Table 58 itemizes and summarizes the characteristics of plants in this
subcategory. A separate mean has been calculated for three plants which
had conducted extensive monitoring. Many of the other plants had data
collected only from grab samples and flow estimates. For this reason
it is felt that the ratios for these three plants more accurately reflects
actual operating conditions. Based on these means it is felt the following
ratios are typical for this subcategory.
Flow BOD SS
Ratio Ratio Ratio
(1/cu m) (kg/cu m) (kg/cu m)
3540 2.38 0.380
It should be noted that the three plants with the lowest flow ratios
were primarily canners with minor bottle washing or, in the case of
Plant 86A29, a bottler whose bottles were being washed by an outside
agent.
The pH for this subcategory is expected to vary between 5.5 and 12 with
relatively high alkalinity due to the bottle washer. BOD to nitrogen
and phosphorus ratios are expected to remain 60:1 and 110:1, respectively.
Model Plant
For the purposes of control and treatment technology and cost analysis
a model plant has been designed. The production was set at 135 cu m
(35,900 gal) per day. Based on this production and the ratios listed
above, the raw waste loads for the model plant are as follows:
Flow (MGD) 0.126
BOD (mg/1) 660
SS (mg/1) 108
Total KN (mg/1) 11
pH 5.5 to 12
SUBCATEGORY A 28 - BEVERAGE BASE SYRUPS AND/OR CONCENTRATES
As discussed in Section III, it has been determined that the major
individual waste streams generated in the beverage base manufacturing
process are as follows:
1. Washing of mixing tanks and flavor tanks at the end of
each day and between flavor changes.
2. Washing of syrup tank cars, 208 1 (55 gal) drums, and 19 1
(5 gal) containers prior to refilling.
3. General plant cleanup.
-------
DRAFT
TABLE 58
RAW WASTE CHARACTERISTICS
BUBCATEGQRY A27
Plant
86A04
86A06
86A07
86A13
86A16
86A20
86A24
86A25
86A26
86A29
86A32
86A34
86A37
86A38
86A39
86A40
Mean
Flow Ratio
(1/cu m)
1260
1990
4120
4520
6780
4290
9370
5910
6380
169
2260
2540
3090.-
2760
3991
3050
3905
BOD Ratio
(kg/cu m)
0.826
0.257
0.371
2.02
4.68
0.806
1.31
6.74
3.01
0.624
3.00
3.04
1.72
1.11
4.12
1.92
2.22
SS Ratio
(kg/cu m)
0.155
0.031
0.283
0.393
2.29
0.322
0.019
0.066
_
. 0.074
0.335
0.226
0.247
0.190
0.629
0.325
0.372
Mean
(38, 39, 40) 3267 2.38 0.380
-------
DRAFT
Washing of Mixing Tanks
Plant 85S06 generates approximately 76 cu m/day (0.02 MGD) of wastewater
from the washing of six mixing tanks and eleven flavor tanks which feed
the mix tanks. It should be noted that this figure is highly variable
with the daily quantity dependent on the number of flavor changes made
and the number of batches mixed on any given day. The equipment is
commonly washed using automatic or manually operated spray ball devices
mounted within the equipment and the quantity of water used is usually
regulated closely.
Washing of Tank Cars, Drums and Containers
The cleaning of tank cars generally consists of a hot water wash fol-
lowed by a sanitizing rinse. Drums are commonly washed in sealed wash
tanks. Each drum is fitted with one resealable opening at the drum's
equator. The drums are positioned on a rack wih the opening face down.
Hot wash water followed by hot rinse water is injected into the drum.
After draining, the drums are removed. The 19 1 (5 gal) containers are
washed by vertical placement (opening down) in a revolving washer which
regulates the water output into each container.
Plant 87S06 reported the following daily quantities of wastewater from
each of these cleaning operations during a normal day:
1. Tank cars (average eleven) - 57 cu m/day (0.015 MGD)
2. Drums - 303 cu m/day (0.08 MGD)
3. Containers - 7500 I/day (2,000 gal/day)
It is noted that these quantities will vary within the plant and between
plants depending on daily cleaning requirements.
General Plant Cleanup
Wastewater quantities typically generated during cleanup at Plant 87S06,
consisting of pipe line sterilization and floor washing, average 30 cu m/day
(0.008 MGD) and this quantity would not be expected to vary markedly
throughout the industry.
Non-Contact Water
There is a small amount of non-contact machinery cooling water and boiler
blowdown generated in the manufacturing of beverage bases. This non-contact
water is generally discharged into the process waste stream or into storm
sewers.
Total Plant Effluent
The wastewater characteristics of the total plant effluent for five
beverage base plants are summarized in Table 59. The data indicate a
wide range of flow and BOD concentrations but consistently show low
405
-------
o
CTl
TABLE 59
SUMMARY OF WASTEWATER CHARACTERISTICS
SUBCATEGORY A 28
PLANT
CODE
87S06
87S07
87 508
87S09
87S14
FLOW
cu m/day
598
68
125
396
459
cu
1.
0.
1.
m/cu m
05
40
16
BOD
mg/1
1868
5910
3750
1140
3050
SS
kg/cu m
2
1
3
.02
.43
.56
mg/1
32
328
40
162
353
kg/cu m
0.032
0.016
0.36
H £.
mg/1 mg/1
35.1 12.2
-------
DRAFT
suspended solids concentrations as compared to BOD concentrations. The
pollutant ratios were determined based on additional data provided by
Plants 87S06, 87S08, 87S14. Plants 87S06 and 87S14 showed good agreement
in terms of wastewater flow per unit of product produced. However, the
pollutant loadings per unit of product produced are dissimilar. Plant
87S08 which generated roughly 60 percent less flow per unit of product
produced than Plant 87S06 had a BOD pollutant loading 30 percent less and
a suspended solids loading 50 percent less. This indicates a rough cor-
relation of 0.5 between the two plants. The nutrient to BOD ratio
(BOD:N:P) was determined to be 100:3.1:1.1 based on the data obtained
from Plant 87S09. It must be noted that only a limited number of data
points was available in determining the data presented in Table 59.
However, the data do offer sufficient information to allow reasonable
assumptions as to the anticipated characteristics of a model beverage
base manufacturing plant.
Mode! Plant
Based on the above considerations, a hypothetical model plant was developed
for Subcategory A 28 and is illustrated in Figure 133. The plant generates
an average wastewater flow of 379 cu m/day (0.10 M6D) due to washing of
mixing and flavor tanks, washing of tank cars, drums, and containers, and
general plant cleanup. The model plant has the following average charac-
teristics.
Production 379 cu m/day (0.10 MGD)
Flow 379 cu m/day (0.10 MGD)
BOD 2400 mg/1
SS 50 mg/1
pH 8.0
The assumed characteristics would be expected to vary with seasonal pro-
duction demands and the amount of cleanup operations conducted in the
plant on any given day. There is some reason to believe that the waste
stream may be slightly deficient in nitrogen based on the BOD:N:P ratio
for Plant 87S09.
SUBCATEGORY A 30 - INSTANT TEA
Production of instant tea generates wastewater from two sources,
clarifier sludge and cleanup.
Clarifier Sludge
Periodic discharge of tea sludge is the only process wastewater gen-
erated in the processing of instant tea. There is no reliable way to
estimate the quantities of pollutant loadings of the clarifier waste
stream since the discharge is highly variable.
Cleanup Mater
Cleaning of equipment may be done on several different schedules as
indicated in the following:
407
-------
DRAFT
SUGAR (SYRUPS ONLY), ACIDS,
COLORS, FLAVORING EXTRACTS
WATER
TREATMENT
MIXING
TANKS
STRAINING
FILLING
CLEANUP WATER
WASHING (DRUMS,
TANK CARS, 5 GAL.
CONTAINERS)
PLANT
CLEANUP
WATER
WASTEWATER
FIGURE 133
MODEL PLANT FOR
BEVERAGE CONCENTRATE AND SYRUP MANUFACTURING PROCESS
408
-------
DRAFT
1. One plant (99T04) operating 365 days per year implements
cleanup of the entire plant every ten days.
2. Two plants (99T01 and 99T02) generally operating on a five-day per
week basis implement plant cleanup at the end of each week.
All plants contacted did some periodic cleaning of equipment during the
week as needed. All plants contacted also did some hosing of floors in
the plant for cleaning of leaks from connections. Equipment cleanup is
generally done by spray ball devices contained within the equipment which
are operated manually as needed. The cleanup consists of the following
sequential steps: (1) fresh water rinse, (2) caustic wash, (3) fresh
water rinse, (4) acid rinse and (5) fresh water rinse.
Non-Contact Water
A considerable amount of cooling water is generated in the processing
of instant tea. Only one plant contacted (99T04) separated all cooling
water from process water. Two plants, 99T02 and 99T03, provided no
separation of contact and cooling water and no recycling of cooling
water. One plant (99T01) recycled a majority of the cooling water
used in the process and discharged the unrecycled into the wastestream.
Total Plant Effluent
The wastewater characteristics of the instant tea industry are summarized
in Table 60. The two plants (99T02 and 99T04), for which the portion
of total effluent attributable to process water was known, showed good
agreement regarding process waste flows with the values being 49,500 1/KKg
(11,900 gal/ton) and 46,500 1/KKg (11,100 gal/ton), respectively.
Plants 99T01 and 99T03 contained an indeterminant amount of cooling water
which could account for the significant difference in flow.
Pollutants in the waste stream considered of significance in^instant tea
manufacturing are BOD and suspended solids. Plants 99T01 and 99T02
showed good agreement of mass of BOD and suspended solids generated per
unit of product produced. Of the remaining two plants, 99T03 generated
four times the BOD load of the two showing agreement and 99T04 generated
a BOD load which was a factor of four less than the two showing agreement.
A possible explanation of the low BOD and suspended solids loadings
generated by plant 99T04 is that clarified tea sludge, rather than
being added to the wastestream, is centrifuged and the dewatered sludge
is sold as cattlefeed.
Model Plant
Based on the data presented in Table 60 and the preceeding discussion
a hypothetical instant tea manufacturing plant was determined and is
illustrated in Figure 134. The plant operaates 24 hours per day, five
days per week with cleanup of all equipment at the end of each processing
week. Daily wasteflow consists of cleanup of equipment as needed, floor
washdown to clean leakage from equipment connections, and deposition of
409
-------
TABLE 60
SUBCATEGORY A 30 - SUMMARY OF WASTEWATER CHARACTERISTICS
Plant
99T01
99T02
99T03
99T04
Flow
1/KKg
*94,700
49,500
*167,000
46,500
BOD
Kg/KKg
41.1
52.4
196.3
10.0
SS
Kg/KKg
34.7.
38.2
-
5.8
* Values are high due to indeterminant amount of cooling
water in the plant effluent.
-------
DRAFT
TEA LEAVES
TO
S0.1D WASTE
FIGURE 134
MODEL PLANT FOR SUBCATE60RY A 30
INSTANT TEA MANUFACTURING PROCESS
411
-------
DRAFT
clarifier tea sludge into the wastestream. All cooling water is dis-
charged separately from the process waste stream. The characteristics
of the model plant are as follows:
Production 9.1 KKg/day (10 ton/day)
Flow 450 cu m (0.12 MGD)
BOD 1000 mg/1
SS 750 mg/1
pH 6.0
Spent tea leaves from the centrifuge are sold as cattle feed or disposed
as solid waste.
SUBCAT'EGORY c s - COFFEE ROASTING UTILIZING ROASTER WET SCRUBBERS
Roaster Wet Scrubbers
A study conducted at a coffee roasting plant utilizing a once-through
type of wet scrubber reported an effluent with a BOD of 100 to 500 mg/1,
suspended solids of 180 to 240 mg/1 and a flow rate of 2100 liters per
kkg'(508 gallons per ton) of green beans roasted. No data are
currently available on the wastewater characteristics of the recirculating
type of wet scrubbers used on coffee roasters.
Total Processing Effluent
Roaster wet scrubbers are the only source of wastewater from a coffee
roasting plant. Table 61 presents a raw waste summary d~f this wastewater.
Model Plant
The model plant for this subcategory is a coffee roasting plant which
utilizes a once-through type of roaster wet scrubber. The model plant
roasts 30 kkg (33 tons) per day of green coffee beans.
Wastewater - The only source of wastewater from the model plant is the
roaster wet scrubber. Parameters of the wastewater are assumed to be
as follows:
1. Flow rate - average - 0.063 mid (17,000 gpd)
2. BOD - 350 mg/1
3. SS - 200 mg/1
4. pH - 4.0 - 7.2
5. 0.76 - kg BOD per kkg of green beans
6. 0.43 - kg SS per kkg of green beans
7, N - 0 mg/1 (assumed, none suspected)
8. P - 0 mg/1 (assumed, none suspected)
412
-------
DRAFT
TABLE 61
RAW WASTE SUMMARY SUBCATE60RY C 8
COFFEE ROASTING
Parameter
Shift Time Hr/Day
Log Mean
8
Minimum
Maximum
8
Flow Ratio L/kkg 2120
(gal/ton) 508
2030
486
2250
539
5 Day BOD mg/1
Ratio kg/kkg
(Ib/ton)
270
0.51
1.02
113
0.27
0.54
645
0.62
1.50
SS mg/1
Ratio kg/kkg
(Ib/ton)
203
0.43
0.86
180
0.39
0.77
240
0.54
1.08
413-
-------
DRAFT
SUBCATEGORY C 9 - DECAFFEINATION OF COFFEE
Extract Centrifuge and Still Slowdown
The blowdown liquor from the solvent recovery still and the extract
centrifuge are normally disposed of as part of the waste stream from the
decaffeination process. These blowdown liquors are a significant source
of both wastewater strength and volume from the decaffeination process.
They contain high concentrations (quantitative data is not available)
of suspended solids, and to a lesser extent, BOD.
Dewatering Screen
After the extract and the beans have been separated, the beans are washed
and screened before drying. The dewatering screen is the source of the
greatest volume of wastewater in decaffeination plants which employ
this device. Although no data is available to quantitatively define the
characteristics of this wastewater, it is estimated that the strength
of this source of wastewater is less than all others except general
plant cleanup.
Cleanup
In plants which do not utilize a dewatering screen, cleanup is the most
significant source of wastewater volume. In addition, decaffeination
processing plant cleanup is an important source of waste strength. Floors
in the process area are hosed down as needed, usually once or twice a
day. The decaffeinating equipment is thoroughly wet cleaned weekly and
spot cleaned as necessary during the week. Caffeine storage areas are
cleaned periodically and also contribute to the wasteload of the plant.
Total Processing Effluent
The quantity and quality characteristics of wastewater from coffee
decaffeinating plants vary considerably. These variations can usually be
traced to the amount of cleaning required on a given day. In-plant studies
(82) show almost two fold variations in daily flow, three fold variations
in BOD and three fold variations in suspended solids. Table 62 includes
data describing the total processing effluent for this subcategory.
Model Plant
The model plant for this subcategory is a hypothetical plant producing
whole decaffeinated green coffee beans. Decaffeination is accomplished
using the liquid/liquid extraction process. All equipment and floors
are wet cleaned. The model plant is assumed to process 55 kkg (60 tons)
of green beans per day, operating 24 hours per day, six days per week.
Wastewater - Sources of wastewater from the model plant would include
all of the sources listed above with the greatest quantities of waste-
414
-------
DRAFT
TABLE 62
RAW WASTE SUMMARY SUBCATEGORY C 9 - DECAFFEINATION OF COFFEE
Parameter NP Logmean Minimum Maximum
Production kkg/day - 55 -
tons/day 60 -
Flow MLD 2 0.242 0.213 0.275
MGD 0.070 0.062 0.079
Flow Ratio 1/kkg 2 4406 3880 5004
gal/ton 1164 1025 1322
BOD mg/1 2 864 682 1045
kg/kkg 3.8 3.0 4.6
Ib/ton 7.5 6.1 9.2
SS mg/1 2 1590 1182 2114
kg/kkg 7.0 5.2 9.3
Ib/ton 13.9 10.4 18.5
Color index * 1 4.5
* Note: Index 4 is the color normally identified with a U.S. cup of
coffee; e.g., 4.5 is more intense than the color of a cup of
coffee
-------
URAFT
water coming from the dewatering screen in plants which utilize this
device. Lesser quantities of wastewater are generated by general
plant cleanup and extract centrifuge and solvent recovery still blow-
down.
Parameters of the raw wastewater were assumed as follows:
1. Flow rate - average - 0.24 mid (70,000 gpd)
2. BOD - 864 mg/1
3. SS - 1590 mg/1
4. pH - 4.3 to 7.2
5. N - 0 mg/1 (assumed)
6. P - 0 mg/1 (assumed)
7. 3.8 kg BOD per kkg of green coffee processed
8. 7.0 kg SS per kkg of green coffee processed.
SUBCATEGORY C 10 - SOLUBLE COFFEE " - •-
Grounds Disposal
Soluble coffee plants normally dispose of their spent grounds by
incineration or in a sanitary landfill. Most plants hydraalicaily
press the grounds to reduce their moisture content prior to disposal
by either of these two methods. The source of the greatest waste
load in most soluble coffee plants is the process of grounds pressing.
Data compiled during this study indicates that grounds pressing waste-
water may have a BOD of up to 30,000 mg/1. A large amount of color is
also characteristic of this wastewater.
Some soluble coffee processors utilize rotary driers rather than presses
to remove moisture from the spent grounds prior to incineration. In
these plants the only source of wastewater is drainage off of the
grounds during storage. Although this is also a source of waste load
it is a less significant wastewater source than grounds pressing.
Centrifuge
In the soluble coffee process hot water passes through the coffee
grounds to extract the soluble constituents* The solution resulting
from this process also contains suspended materials which must be
removed by centrffugation. A major source of waste -load in nearly all
soluble coffee plants is centrifuge cleaning and blowdown. The con-
centrated sludge and cleanup wastes from the centrifuge are normally
discharged as part of the liquid waste stream from the plant.
416
-------
DRAFT
Extract Concentrator Condensate
Before the extracted soluble coffee materials are converted to a solid
by drying, the liquid extract is usually concentrated. Concentration
is accomplished either by heating or cooling the solution. Whichever
method is utilized, a large volume of condensate is generated. This
condensate is the largest single source of wastewater flow from a
soluble coffee plant.
Cleanup
General cleaning of equipment and floors in a soluble coffee plant is
also a significant source of wastewater generation. Floors are wet
cleaned as necessary during production and thoroughly cleaned weekly.
The extractors and related equipment are self-cleaned during production.
Once a week during general cleanup the extractors are cleaned with a
caustic solution.
Total Processing Effluent
The quantity and quality characteristics of soluble coffee processing
wastewater can vary as a result of cleaning procedares
-------
DRAFT
TABLE 63
RAW WASTE SUMMARY
SUBCATEGORY C 10 - SOLUBLE COFFEE
Parameter
NP
Log Mean
Minimum
Maximum
Production kkg 3
tons
Flow mid 2
mgd
Flow Ratio 1/kkg 2
gal /ton
BOD mg/1 3
kg/kkg
Ib/ton
SS mg/1 3
kg/kkg
Ib/ton
78
86
0.617
0.180
7912
2090
2377
18.8
39.6
1555
12.3
24.7
40
44
0.355
0.102
4505
1190
2136
16.9
33.7
683
5.4
10.8
153
169
1.09
0.315
13930
3680
2940
23.26
46.52
3565
28.2
56.5
Color cpu*
2775
* Cobalt - platinum units
418
-------
DRAFT
4. pH - 4 to 5
5. 18.8 kg BOD per kkg of green coffee processed
6. 12.3 kg SS per kkg of green coffee processed
7. N - 0 mg/1 (assumed)
8. P - 0 mg/1 (assumed)
9. Color - 2775 Cobalt - platinum units
SUBCATEGORY F 1 - TEA BLENDING
The blending of tea has been determined to be a dry process involv-
ing the generation of no wastewater.
BAKERY AND CONFECTIONERY PRODUCTS
SUBCATEGORY C1 - CAKES, PIES. DOUGHNUTS, AND SWEET YEAST GOODS UTILIZING
PAN HASHING
Pan Hashing
The source of the greatest waste load in a cake or pie bakery is the
process of washing pans. Pan washing is almost exclusive to the cake
baking industry whether the cakes are full size or the snack cake variety.
Normally, the pans that are used in baking cakes must be washed after
each use. After cakes have been removed from their baking pan, a thin
layer of cake crumbs usually remains in the pan and is removed in the
pan washing operation. The crumbs are essentially pieces of the cake
and thus have a high organic content. Previous studies ( 7 ) and the
wet sampling associated with this project indicates that the cake pan
wash water may have a BOD of up to 54,000 mg/1.
Efforts are being made by a number of bakeries to decrease the amount of
pan washing required in their cake production. Some bakeries are
washing less than once every use, and at least one bakery has completely
eliminated pan washing in their snack cake production.
Hash Room
A feature of virtually every bakery is a wash room in which portable
equipment is cleaned. Items such as small mixers, mixing vats, reuseable
ingredient containers, and hand utensils are normally dry cleaned as
thoroughly as possible before and/or after being taken to the wash room.
In the wash rooms, these items are thoroughly cleaned using hot water
or steam, with the wastewater being collected by floor drains or sinks.
Cake bakeries may have more than one wash room. Some may be operated
in the manner described above with manual cleaning of equipment. Other
types of wash rooms are essentially large dishwashers. Racks of cake
419
-------
DRAFT
pans and other items are rolled into these units. They are completely
closed and then the equipment inside is washed in a manner similar
to that accomplished by home dishwashing equipment.
Clean-In-Place Equipment
Large equipment handling liquid or semi-solid materials are usually
fitted with clean-in-place (CIP) equipment. This equipment includes the
plumbing, controls, and sewer connections necessary to wash the unit
without moving it to the wash room. During interruptions in processing
caused by changes in variety of product or due to end of shift cleanup,
equipment with CIP is normally dry cleaned as thoroughly as possible
and then wet cleaned using hot water and detergent supplied through the
CIP equipment. Wastewater discharge from CIP equipment is normally
through a direct connection to the plant's plumbing system.
Examples of equipment with CIP include the following:
1. Large mixers for cake batter
2. Piping used to deliver the batter from the mixer to
the depositer.
3. The depositer which fills each cake pan with the proper
amount of batter.
Cleanup
General cleanup of other equipment associated with the baking process
and the plant itself is a relatively minor contributor to the waste
load. Conveyors used for the baking and cooling of cakes and pies
are usually dry cleaned; however, they may be wet..eleaned as frequently
as once a week. Cleanup procedures in most plants stress the dry
cleaning of equipment and the floor spaces around the equipment. However,
some wet cleaning is normally accomplished. Some exceptionally dirty
areas may be hosed down; however, more common practices include the use
of mop and bucket or the vacuum type of wet scrubber for floors.
Total Processing Effluent
The quantity and quality characteristics of wastewater from cake and
pie bakeries can vary considerably. These variations can usually be
traced to operating and cleanup procedures associated with the amount
and type of product being produced and the associated cleaning required.
In-plant studies (83 ) show five fold variations in BOD from one day to
the next and three fold variations in wastewater flow. Table 64 includes
data describing the total processing effluent for this subcategory.
Model Plant
The model plant for this subcategory is a hypothetical bakery producing
a variety of cake and pie items. Production includes both full sized
and snack cakes, full sized pies, and sweet yeast goods. The cakes and
-------
DRAFT
snack cakes are baked in pans wh.ich are washed after each use. The pies
and sweet yeast goods are baked on conveyors or in one-way containers
thus the containers require little or no wet washing. Operating procedures
stress dry cleaning of all equipment prior to their cleaning with water.
Total production at the model plant is assumed to be 135 kkg (150 tons)
per day produced in 24 hours -per day, seven days .per week operation.
Hastewater - Sources of wastewater from the model plant would include all
sources listed above with the greatest strength and quantities of waste
coming from the pan washing equipment and the wash rooms. Lesser
quantities of wastewater are generated by the clean-in-place equipment
and general plant cleanup.
Parameters of the wastewater are assumed as follows:
1. Flow rate - average - 0.45 mid (120,000 gpd)
minimum - 0.20 mid (53,000 gpd)
maximum - 0.60 mid (160,000 gpd)
2. BOD - 28,000 mg/1
3. SS - 5,000 mg/1
4. Oil and Grease - 500 mg/1
5. pH - 6.0 to 7.0
6. N - 2 mg/1 (deficient)
7. P - 20 mg/1 (deficient)
8. Ratio - kg BOD to kkg of product - 94.2
9. Ratio - kg SS to kkg of product - 16.8
10. Ratio - kg 0 & G to kkg of product - 1.7
These parameters generally follow those listed in Table 64 for this
subcategory with the exception of suspended solids. The suspended solids
data reported for Subcategory C 1 appears unrealistically low. This
is particularly true when a comparison is made among suspended solids
data from Subcategories C 1, C 2, C 3, and bakeries which span these
subcategories. Thus, the figure of 5,000 mg/1 was used and is based
on data from a bakery spanning subcategories C 1 and C 3.
SUBCATEGORY C 2 - CAKES, PIES, DOUGHNUTS, AND SWEET YEAST GOODS NOT
UTILIZING PAN HASHING
With the exception of pan washing, the sources of wastewater in bakeries
not utilizing pan washing are identical to those in Subcategory C 1.
The principal sources of wastewater are as follows:
421
-------
UKAH
TAHLE 64 . RAW WASTE SUMMARY
CAKCS, PIES, DOUGHNUTS, AND SWEET YEAST
GOODS UTILIZING PAN WASHING
PARAMETER
PROD KKG/DAY
(TON/DAY)
SHIFT TIME HR/DAY
FLOW VOLUME MGD
FLOW RATE L/StC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TON)
5 OAY BOO MG/L
RATIO KG/KKG
(L3/TON)
TSS MG/L
RATIO KG/KKG
(LB/TON)
NO PLANT LOG MEAN MINIMUM
1 172
190
1 2k. Q
1 0.160
1 7.01
111
1 3530
81+5
1 18000
63.5
127
1 1590
5.63
11.2
MAXIMUM
„
— ~
--
--
--
— —
--
- —
--
--
~~
-.
—
PROCESS COOE(S): 51C75I
422
-------
DRAFT
1. Wash room
2. Clean-in-place equipment
3. Cleanup
See the previous discussion of Subcategory C 1 for a description of these
sources and their effects on the total plant effluent.
Total Processing Effluent
The quantity and quality of wastewater from cake and pie bakeries not
utilizing the pan washing vary considerably. These variations can usually
be traced to operating and cleanup procedures associated with the amount
and type of product being produced. In-plant studies ( 84) show up to
five fold variations in BOD from one day to the next ar.u three fold
variations in the wastewater flow. Table 65 includes data describing
the total processing effluent for this subcategory.
Model Plant
The model plant for this subcategory is a hypothetical bakery primarily
producing cakes and snack cakes. Pies baked in one-way pans are also
produced, but account for only a small percentage of the plant's total
production. All of the items baked in the plant are produced by methods
which completely eliminate pan washing. Operating procedures stress
dry cleaning of all equipment prior to cleaning with water. Total pro-
duction at the model plant is 180 kkg (200 tons) per day produced in 24
hours per day, five days per week operation.
Wastewater - Sources of wastewater from the model plant will include
all sources listed above with the greatest strength and quantities
of waste coming from the wash room. Lesser quantities of wastewater
are generated by the clean-in-place equipment and general plant cleanup.
Parameters of the wastewater are assumed as follows:
1. Flow rate - average - 0.16 mid (43,000 gpd)
minimum - 0.14 mid (37,000 gpd)
maximum - 0.19 mid (51,000 gpd)
2. BOD - 2,190 mg/1
3. SS - 1,020 mg/1
4. Oil & Grease - 685 mg/1
5. pH - 5.0
6. N - 30 mg/1 (deficient)
423
-------
LJRAFT
TABLE 65
RAW WASTE SUMMARY SUBCATEGORY C2
CAKES, PIES, DOUGHNUTS, AND SWEET YEAST GOODS NOT UTILIZING PAN WASHING
Parameter Log Mean Minimum Maximum
Prod kkg/day 180
(ton/day) 200
Shift time hr/day 24
Flow volume MGD 0.043 0.037 0.051
MLD 0.163 0.140 0.193
Flow ratio 1/kkg 897 722 1064
(gal/ton) 215 185 255
5 day BOD mg/1 2190 1890 2540
Ratio kg/kkg 2.0 1.7 2.3
(Ib/ton) 4.0 3.4 4.6
SS mg/1 1020 950 1100
Ratio kg/kkg 0.92 0.86 1.0
(Ib/ton) 1.84 1.72 2.0
0 & G mg/1 685 570 830
Ratio kg/kkg 0.62 0.51 0.75
(Ib/ton) 1.24 1.02 1.50
424
-------
DRAFT
7. P - 15 mg/1 (sufficient)
8. 2.0 kg BOD per kkg of product
9. 0.94 kg SS per kkg of product
10. 0.63 kg 0 & G per kkg of product
SUBCATEGORY C 3 -TREAD AND "BUNS
Mixing Equipment Cleaning
The cleaning of mixing equipment is the largest source of wastewater from
a bread and bun bakery. The cleaning may be done manually or with clean-
in-place (CIP) equipment, which consists of the plumbing, controls, and
sewer connections necessary to wash the equipment automatically. The
mixing equipment is normally cleaned daily by first scraping the walls
of the mixers to remove adhering dough and then washing. The solid
(dry) material is either sold as animal feed or handled as solid waste.
In plants using the continuous mix method, the mixers and dough slurry
tanks are then rinsed daily with water. In plants using the batch mix
method, the mixers are normally wet cleaned once or twice a week.
General Cleanup
General cleaning of floors and utensils is the other important source of
wastewater from bread and bun bakeries. Utensils are normally washed in
a sink in the production area in which they are used. Floors in the
mixing area are generally wet cleaned daily using mops and buckets, hoses,
or scrubbers which vacuum the water and spilled product from the floor
as it is used. Floors throughout the rest of the plant are routinely
cleaned several times a day using brooms and dry vacuum cleaners. Once
or twice a week all of the floors in the plant are wet cleaned with mop
and bucket or vacuum scrubber.
Total Processing Effluent
The quantity and quality of wastewater from bread and bun bakeries varies
considerably. These variations are usually the result of the amount
of cleanup taking place and the training and management of the personnel.
Data from various plants show two fold variations in flow and in BOD from
one day to the next. Table 66 includes data describing the total processing
effluent for this subcategory.
Model Plant
The model plant for this subcategory is a hypothetical bakery producing
bread and buns. Buns are a minor production item. All items are batch
mixed, baked in pans (sometimes with lids), and mechanically packaged
in plastic bags. Operating procedures stress dry cleaning of all equipment
prior to wet cleaning (if required). Total production of the model plant
425
-------
DRAFT
TA9LE 66. RAW WASTE SUMMARY
BRFAO AND 3UNS
PARAMETER
PROD KKG/OAY
(TON/DAY)
SHIFT TIME HR/DAY
FLOW VOLUME MGD
FLOW RATE L/SEC
CGAL/MINI
FLOW RATIO L/KKG
(GAL/TON)
5 DAY 800 MG/L
RATIO KG/KKG
(L8/TON)
TSS MG/L
RATIO KG/KKG
(LB/TON)
NO PLANT
14
14
1**
14
Ik
Ik
13
LOG MEAN
40.8
<*5.0
22.7
0.026
1.32
21.0
2080
<»99
<422
0.877
1.75
21k
0 .«46i+
0.927
MINIMUM
21.4
23.6
16.0
0.01<*
0.561
8.90
952
228
110
0.26^
0.526
5<».«»
0.148
0.296
MAXIMUM
78.1
86.1
24.0
0.051
3.12
49. k
4540
1090
1610
2.92
5.84
844
1.45
2.90
PROCESS COD£(SM 51C53I ,51C54I ,510581 ,510601 ,510611 ,
51C62M ,510631 ,510641 ,51C65M ,51C66M ,510685 ,510721 ,
51C80I ,51C82M
426
-------
DRAFT
is assumed to be 41 kkg (45 tons) per day produced in 24 hours per day,
five days per week.
Wastewater - Sources of wastewater from the model plant would include all
sources listed above with nearly all of the wastewater generated by
cleaning of mixing equipment and floors throughout the plant.
Parameters of the wastewater are assumed to be as follows:
1. Flow - average - 0.10 mid (0.026 mgd)
minimum - 0.053 mid (0.014 mgd)
maximum - 0.19 mid (0.051 mgd)
2. BOD - 422 mg/1
3. SS - 214 mg/1
4. pH - 6.0 to 9.0
5. P - 0 mg/1 (assumed)
6. N - 0 mg/1 (assumed)
7. 0.88 kg BOD per kkg of product
8. 0.46 kg SS per kkg of product
SUBCATEGORY C7 - COOKIE AND CRACKER MANUFACTURING
Wash Room
A feature of virtually every cookie and cracker bakery is a wash room
in which portable equipment is cleaned. The cleaning may be done
manually or the wash room may be essentially a large automatic dish-
washer. Equipment such as small icing mixers, mixing vats, rotary
formers, enrobers, and hand utensils are normally thoroughly scraped
and dry cleaned before and/or after being taken to the wash room.
These items are then cleaned in the wash room using hot water and/or
steam.
The major source of wastewater in the wash room is associated
with the cleaning of icing and enrobing equipment (normally associated
with cookie manufacturing). The quantity of wastewater generated by
the wash room varies greatly, depending on the operating personnel.
Clean-In-Pi ace Equi pment
Large equipment handling liguid and semi-solid materials is usually
fitted with cleanv-in-place (CIP) equipment. This equipment includes
the plumbing, controls, and sewer connections necessary to wash the
unit without moving it to the wash room. During interruptions in
processing caused by changes in the variety of product or due to end
427
-------
of shift cleaning, equipment with CIP is normally dry cleaned as
thoroughly as possible and then wet cleaned using hot water and
detergent supplied through the CIP equipment. Wastewater discharge is
normally through a direct connection to the plant waste plumbing system.
Cleanup
General cleanup of other equipment associated with the baking process
and the plant itself is the least significant contributor to the waste
load. Conveyors used for the baking and cooling of cookies and crackers
are usually dry cleaned; howevers they may be wet cleaned as frequently
as once a week. Cleanup procedures in most plants stress the dry
cleaning of equipment and the floor spaces around the equipment.
However, exceptionally dirty areas may be hosed downa or more commonly
be cleaned with a vacuum type wet scrubber or a mop and bucket.
Total Processing Effluejrt
The quantity and quality of wastewater from cookie and cracker bakeries
can vary considerably. These variations are usually the result of
cleanup procedures associated with the type ©f product being produced
and the training and management of the personnel. Data collected
during this study show six fold variations in BOD from one day to the
next and two fold variations in wastewater flow within a single plant.
Table 67 includes data describing the total processing effluent for
this subcategory.
Model Plant
The model plant for this subcategory is a hypothetical bakery producing
a variety of cookie and cracker items. Production includes crackers,
iced and plain cookies, pretzels and sugar wafers. The model plant
produces cookie items and cracker items in approximately equal
quantitites. All items are batch mixed9 baked on conveyor belts in
tunnel ovens (except sugar wafers which are baked in plates in a special
type of oven) and tumble or shingle stack packaged. Operating pro-
cedures stress dry cleaning of all equipment prior to wet cleanup
(if required). Total production for the model plant is assumed to be
180 kkg (200 tons) per day produced in 24 hours per days five days per
week.
Wastewater - Sources of wastewater from the model plant would include
all sources listed above with the greatest strength and quantities
of waste coming from the wash room. Smaller quantities of wastewater
are generated by the clean-in-place equipment and general plant
cleaning.
Parameters of the wastewater are assumed as follows:
1. Flow rate - average - 0.34 mid (90,000 gpd)
minimum - 0.20 mid (53,000 gpd)
maximum - 0.45 mid (120,000 gpd)
428
-------
DRAFT
TA3LE67 . RAW WASTE SUMMARY
COOKIES ANJ CRACKERS
PARAMETER
PROD KKG/OAY
(TON/DAY)
SHIFT TIME HR/OAY
FLOW VOLUME MGO
FLOW RATE L/SCC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TON»
5 DAY 300 MG/L
RATIO KG/KKG
(L8/TON)
T5S MG/L
RATIO KG/KKG
(LB/TON)
NO PLANT LOG MEAN
8 122
135
8 21.1
8 0.056
8 3.07
-------
UKMM
2. BOD - 1200 mg/1
3. SS - 900 mg/1
4. pH - 6.0 - 8.0
5. Ratio - kg BOD to kkg of product - 2.0
6. Ratio - kg SS to kkg of production - 1.5
SUBCATEGORY D 1 - CANDY AND CONFECTIONERY PRODUCTS EXCLUDING' GLAZED
FRUITS
Of the total number of confectioners contacted 15 were considered to
have reliable historical wastewater data on which to characterize the
industry as a whole. A summary of this data is given in Table 68.
Direct product contact water usage was not observed in any segment of
this subcategory. The primary source of wastewater with the highest
pollutant loading is derived from the periodic or daily clean-up of
the plant. Although washdown is practiced in most parts of the plant
at some time, the largest and most consistent area of washdown water
generation is the candy kitchen. Washdown water in the remainder of
the plant is usually restricted to mopping and wiping. Some machinery
parts and molding pans may be removed to a separate area for cleansing.
In addition, certain molding machines were observed to use a clean-in-
place system. Other areas which may contribute to the total effluent
loading are boiler blowdown, air scrubbers and barometric condensers.
Non-contact cooling water was observed to be either discharged to storm
sewers, surface water, sanitary sewers or was recirculated.
As noted in Table 68, the average flow ratio was 3770 1/kkg (904"gal/
ton). The average BOD was 5.10kg/kkg (10.2 Ib/ton) with a range of
1.69 to 15.4 kg/kkg (3.38 to 30.7 Ib/ton); suspended solids was 0.648
with a range of 0.168 to 2.50 kg/kkg (0.336 to 5.00 Ib/ton). No cor-
relation between suspended solids and BOD was noted due to the sol li-
b-Mi zed carbohydrates characteristically discharged by this industry.
Oil and grease loadings ranged from 0.05 to 0.832 kg/kkg (0.10 to
1.664 Ib/ton) with an average of 0.21 kg/kkg (0.42 Ib/ton) for the six
plants with this data available. Variability of the wasteloading and
flow was significantly influenced by variations in processing, raw
materials, production level, washdown and general housekeeping prac-
tices. Wastewater in all plants visited was discharged to municipal
treatment facilities. Many plants utilized some minor form of pre-
treatment and/or in-plant controls to reduce waste loadings; particu-
larly where oil and grease were of concern. Pretreatment was usually
in the form of a grease trap. One plant, however, was considering
dissolved air flotation as a method of reducing effluent concentra-
ti ons.
430
-------
DRAFT
TABLE 68. RAW WASTE SUMMARY
CANDY AND CONFECTIONERY
PARAMETER
NO. PLANT
bOG MEAN MINIMUM
MAXIMUM
PROD KKG/DAY
(TON/DAY)
15
97.0
30.7
307
SHIFT TIME HR/DAY 14
FLOW VOLUME MGD is
FLOW RATE L/SEC 14
(GAL/MlN)
FLOW RATIO L/KKG is
( GAL/TON )
5 DAY BOD MG/L 1*
RATIO KG/KKG
( LB/TON)
TSS MG/L 15
RATIO KG/K'KG
( LB/TON)
OIL & GREASE MG/L 6
RATIO KG/KKG
( LB/TON)
16.
0.
6.
108
3770
904
11290
5.
10.
172
0.
11.
55.
0.
0.
8
099
80
10
2
648
29
7
2L
42
8
0
11
27
2070
495
523
11
3
54
0
0
13
0
0
.00
.024
.71
.1
.69
.38
.1
. $68
.336
.3
.50
.110
338
0.
27.
429
6880
1650
3200
15.
30.
545
2.
5.
222.
0.
1.
422
0
4
7
50
00
8
84
68
PROCESS CODE (S)t 65*80M, 65*83M, 65*841, 65*85M, 65*861, 65*87M,
65*81NT, 65880M2, 653821, 65883M, 653841, 65885M, 65H81M, *C080I
431
-------
DRAFT
Model Plant
The following information reflects those conditions judged to be appli-
cable to a representative candy and confectionery product plant:
Production = 97 kg/day (107 ton/day)
Effluent Volume = 375 cu m/day (0.099 MGD)
Effluent characteristics:
BOD = 1300 mg/1
SS =170 mg/1
Oil and Grease = 56 mg/1
pH = 7.7
Primary source of wastewater: Washdowns.
Special consideration: Oil and grease.
SUBCATEGORY D 2 - CHEWING GUM
Data from a total of five plants were used to develop the wastewater
characteristics as summarized in Table 69. Three of the data points
contributing to this summary were from plants visited by the con-
tractor. Other data points represent data contributed by the National
Association of Chewing Gum Manufacturers (NACGM). Because the NACGM
included much supplemental processing and water usage information with
the historical data, it was concluded that such data could be reliably
utilized as part of the data base making the necessary wastewater
characterization.
Water used in the manufacture of chewing gum is primarily for air
scrubber systems with lesser quantities being consumed during plant
washdown. No direct finished product contact water use was observed
or indicated. Washdown of the plant is usually restricted to mopping
and wiping in most areas with a separate room used for cleaning various
pieces of equipment. Some miscellaneous water use generally occurs
in cleaning of mixing room floors. Air scrubber water is usually re-
circulated and periodically purged.
The ratio of water use to production averages 4500 1/kkg (1080 gal/ton)
with an expected range of 3300 to 6130 1/kkg (792 to 1470 gal/ton).
This range is due primarily to variations in plant size and different
conditions affecting the performance of the air scrubber systems.
Expected BOD ratios ranged from 1.2 to 13.6 kg/kkg (2.4 to 27.2 Ib/ton)
with an average of 4.04 kg/kkg (8.07 Ib/ton). Suspended solids ranged
from 0.175 to 0.858 kg/kkg (0.351 to 1.71 Ib/ton) with an average of
0.388 kg/kkg (0.774 Ib/ton). Variability of the BOD and SS loadings
could not be rationalized in all cases, but is likely to be influenced
by variable amounts of sugar dust that is subsequently removed by the
air scrubber system. Differences in general cleanup practices are
suspected to account for a significant variation in wastewater pollutant
load.
432
-------
DRAFT
TABLE 69 . RAW WASTE SUMMARY
CHEWING GUM
PARAMETER
P^Oa KKG/OAY
(TON/DAY)
SHIFT TIME HR/OAY
FL3W VOLUME MGO
FLOW RATE L/SEC
(GAL/MIN)
FLDW RATIO L/KKG
(GAL/TON)
5 DAY 600 MG/L
RATIO KG/KKG
(LB/TON*
T5S MG/L
RATIO KG/KKG
(L3/TON)
NO PLANT LOG WEAN
5 70.9
78.2
5 i<+.i+
5 0.085
5 6.7<+
107
1080
5 897
<+.0<+
8.07
0,77^
MINIMUM
J!ii
8.00
0.039
3.39
53.7
3300
792
360
1.20
2.<»0
35.6
0.175
0.351
MAXIMUM
119
132
2^.0
0.183
13. <4
213
613U
2240
13.6
27.2
2^9
0.856
1.71
"ROCESS CODECSIt 67G81I ,67G82I ,67G83I ,67G8<*I ,*CG81S
433
-------
DRAFT
Intake waters are generally obtained from municipal water supplies;
however, some plants do utilize well water for non-contact cooling and
air washer make-up water. All effluents, except from one plant, ob-
served during the study were discharged directly to municipal treatment
systems. Pre-treatment was generally not employed; however, one plant
treated and subsequently spray irrigated its effluent.
Model Chewing Gum Plant
Based on available information, a representative plant for this sub-
category has been selected as follows:
Production: 70.9 kkg/day (78.2 ton/day)
Wastewater flow volume: 322 cu m/day (0.085 MGD)
Wastewater characteristics: BOD = 900 mg/1
SS = 95 mg/1
Oil and Grease = 30 mg/1
pH = 7.5
Primary Sources of Wastewater - Air scrubbers, cleanup operations.
Special Considerations - None.
SUBCATEGORY D 3 - CHEWING GUM BASE
As in the case of Subcategory D 2, data for two of the plants supplied
by NACGM were considered valuable for the reasons mentioned in the pre-
vious subsection. Table 70 summarizes the data from three chewing gum
base manufacturers.
During the production of chewing gum base, water is used for washing of
the natural gums, for contact and non-contact cooling, and for periodic
cleanup. The greatest volume of water is used during the washing oper-
ation with considerably less being used for general cleanup. The ratio
of water used to production would be expected to range from 1030 to
11,200 1/kkg (247 to 2690 gal/ton) with an average of 3400 1/kkg (815
gal/ton). Although the range of water use is great, the total waste-
loading does not reflect the same wide range, suggesting different
approaches to water use to achieve the same degree of product and/or
plant cleaning.
Expected BOD ratios range from 1.11 to 1.90 kg/kkg (2.21 to 3.80 Ib/ton)
with an average of 427 kg/kkg (2.9 Ib/ton); suspended solids from 0.800
to 1.82 kg/kkg (1.60 to 3.63 Ib/ton) and an average of 1.21 kg/kkg (2.41
Ib/ton). The reason for the variability of the wastewater flow cannot
be attributed to specific processing differences between plants but is
most likely due to differences in raw material quality; i.e., the amount
of extraneous material which must be removed.
The pH range (two plants) was from 8.76 to 9.5 with a numerical average
of 9.13. Sodium hydroxide (NaOH) used as a bleaching agent, is the cause
of the above neutral pH. Surges of higher hydroxide ion concentration
would be expected during the bleaching cycle pump and subsequent rinsing
of the product to remove residual NaOH.
434
-------
DRAFT
TABLE 70 RAW WASTE SUMMARY
CHEWING GUM BASE
PARAMETER NO. PLANT
PROD KKG/DAY 3
(TON/DAY)
SHIFT TIME HR/DAY 2
FLOW VOLUME MGD 3
FLOW RATE L/SEC 2
(GAL/TON)
FLOW RATIO L/KKG 3
(GAL/TON)
5 DAY BOD MG/L 3
RATIO KG/KKG
(LB/TON)
TSS MG/L 3
RATIO KG/KKG
(LB/TON)
PH 2
LOG MEAN
1105
0.16
20.0
0.094
8.40
133
3400
815
427
1.45
2.90
355
1 .21
2.41
9.1 3
MINIMUM
98
109
16
0
3
56
•1030
247
162
1
2
74
0
1
8
.7
.0
.030
.58
.8
. 11
.21
.2
.800
.60
.76
MAXIMUM
*12
123
24.8
0.310
19.7
312
11200
2690
1120
1 .90
3.80
1700
1 .82
3.63
9.50
PROCESS CODES(S)« *CG80H, 67B80I, 67B85I
435
-------
UKAH
During periodic cleanup of equipment and processing areas„ various sol-
vents are utilized to remove built-up gum residues. According to a treat-
ment feasibility study (85) prepared for a gum base plant, a maximum of
2000 gallons per week of solvent was used with a yearly average of 24,000.
This consumption was based on a production average of 5000 Ib/day.
Of the three plants used for characterization, two discharged wastewater
to municipal treatment systems and one plant employed its own treatment
system prior to discharge to a local tributary.
Model Gum Base Plant
Production: 105 kkg/day (116 ton/day)
Wastewater flow volume: 356 cu m/day (0.094 MGD)
Wastewater characteristics: BOD = 430 mg/1
SS = 355 mg/1
Oil and Grease = 30 mg/1
pH = 9.1
Primary Sources of Wastewater - Gum base wash water, contact cooling water,
cleanup.
Special Considerations - Bleaching agent (sodium hydroxide), solvents.
SUBCATEGORY D 5 - MILK CHOCOLATE PRODUCTION WITH CONDENSORY PROCESSING
AND SUBCATEGORY D 6 - MILK CHOCOLATE WITHOUT CONDENSORY PROCESSING
As noted in Section III, some producers of chocolate products may also
engage in the condensing of milk for milk chocolate and were, therefore,
segregated for separate consideration. Wastewater characteristics for
Subcategory D 5, Chocolate Production with Milk Condensory, is based on
six data sets which reflect the majority of chocolate and cocoa products
manufactured in the United States. Three data sets were used in charac-
terization of Subcategory D 6, Chocolate Production without Milk Condensory.
These data are summarized on Tables 71 and 72 and are further discussed
herein.
The presence of water is not compatible with the production of cocoa
products; therefore, the open use of water is controlled so as to avoid
entrainment in the product. The major portion of wastewater generation
occurs during the periodic cleaning of holding or mixing tanks, trans-
fer buggies, and molding pans. The production area floors are also cleaned
on a periodic basis, usually preceded by dry collection and then mopping,
and/or using industrial floor sweepers. Cocoa butter may be used as a
cleaning solvent with the later recovery of the cocoa butter and chocolate
material. Washdown water is also generated during the cleaning of the con-
densed milk line and milk receiving areas, Subcategory D 5.
For Subcategory D 5 BOD loadings averaged 7.48 kg/kkg (14.9 Ib/ton) with
an expected range of 8.69 to 25.7 (kg/kkg (4.35 to 12.9 Ib/ton); suspended
solids averaged 1.68 kg/kkg (3.35 lb/ton)s ranging from 1.83 to 3.08 kg/kkg
(1.83 to 6.15 Ib/ton). Oil and grease averaged 0»69 kg/kkg (1.38 Ib/ton)
with an expected range of 0.32 to 1.06 kg/kkg (0.64 to 2.12 Ib/ton) and for
436
-------
DRAFT
TABLE 71 RAW WASTE SUMMARY
CHOCOLATE, WITH MILK CONDENSORY
PARAMETER NO. PLANTS
PROD
(
SHIFT
FLOW
FLOW
FLOW
5 DAY
RATIO
KKG/DAY 5
TON/DAY)
TIME HR/DAY 5
VOLUME MGD 5
RATE L/SEC 5
(GAL/MIN)
RATIO L/KKG 5
(GAL/TON)
BOD MG/L 5
KG/KKG
(LB/TON)
TSS MG/L 5
RATIO
OIL &
RATIO
KG/KKG
(LB/TON)
GREASE MG/L 4
KG/KKG
(LB/TON)
LOG MEAN
333
367
22.
0.
7.
1114
4070
975
18*0
7.
!14.
413
1 .
3.
169.
0.
1 .
4
201
22
48
9
68
35
5
69
38
MINIMUM
117
129
16.
0.
1 .
24.
2310
553
1300
4.
8.
308
0.
1 .
78.
0.
0.
8
077
51
0
35
69
915
83
6
32
64
MAXIMUM
944
1040
24
0
34
546
7170
1720
2600
12
25
553
3
6
260
a
2
.0
.524
.4
.9
.7
.08
. 1
.4
5
.06
. 1
2
PROCESS CODE (S): 66*80I, 66*80W, 66*80W1, 66*8315, 66*83W5
437
-------
UKAH
TABLE 72 RAW WASTE SUMMARY
CHOCOLATE, WITHOUT MILK CONDENSQRY
PARAMETER NO. PLANTS
PROD KKG/DAY 3
(TON/DAY)
SHIFT TIME HR/DAY 3
FLOW VOLUME MGD 3
FLOW RATE L/SEC 3
(GAL/MIN)
FLOW RATIO L/KKG 3
(GAL/TON)
5 DAY BOD MG/L 3
RATIO KG/KKG 3
(LB/TON)
TSS MG/L 3
RATIO KG/KKG
(LB/TON)
OIL & GREASE MG/L i
RATIO KG/KKG
(LB/TON)
LOG MEAN
353
278
13.3
0.243
20.2
320
6560
1570
"Z05
4.63
9.24
229
1 .50
3.01
1.59
1 .06
2. 12
MINIMUM
50.3
55.5
8.00
0. 103
12.6
200
5000
1200
1145
1 . 18
2. 36
81 .8
0.669
1 .23
_
-
—
MAXIMUM
1270
1400
16
0
32
514
8620
2070
3420
18
36
642
3
6
_
-
""
.0
.579
.4
. 1
. 1
.38
.76
PROCESS CODE (S): 66*82M, 66*8314, 66*83W4
438
-------
DRAFT
Subcategory D 6, the BOD averaged 4.63 kg/kkg (9.24 Ib/ton) with an expected
range of 1.18 to 18.1 kg/kkg (2.36 to 36.1 Ib/ton); suspended solids averaged
1.50 kg/kkg (3.01 Ib/ton), ranging from 0.669 to 3.38 kg/kkg (1.34 to 6.76
Ib/ton). Oil and grease for the one plant which analyzed this parameter
was 1.06 kg/kkg (2.12 Ib/ton).
BOD and suspended solids loadings appear to be dependent on the relative
amounts of chocolate products produced, i.e., cocoa, syrup, sweetened, un-
sweetened, and milk chocolate. Of special note is the necessary cleaning
of tanks and product containers of the chocolate syrup line. Oil and
grease variability is due to the efficiency of general operating house-
keeping practices used to minimize entrainment of cocoa butter in the
wastewater. In addition, Subcategory D 5 is influenced by washdown from
the milk condensing process and milk receiving area; the total wasteloading
for any one plant being dependent upon the amount of dry milk and/or con-
densed milk which may come from other sources.
Plants in these subcategories characteristically discharge their wastewater
to municipal treatment systems, usually after some form of preliminary
oil and grease removal. This pretreatment may involve only a grease trap
or, as in the case of one plant, a flotation unit. Non-contact cooling
water may either go to municipal treatment or be discharged to surface
waters; the latter being the situation in the larger plants.
Model Chocolate Plant with Condensory Without Condensory
Production: 360 ton/day 240 ton/day
Wastewater flow volume: 761 cu m/day 920 cu m/day
0.201 MGD 0.243 MGD
Wastewater characteristics:
BOD = 1840 mg/1 705 mg/1
SS =415 mg/1 230 mg/1
Oil and Grease = 170 mg/1 160 mg/1
Primary source of wastewater: Washdowns.
Special considerations: Oil and grease removal.
PET FOOD
SUBCATEGORY B 5 - LOW MEAT CANNED PET FOOD
General Plant Cleanup
Clean up in a low meat canned pet food plant is a continuous, minute-
to-minute process which contributes by far the largest share of both
volume and pollutants to the wastewater stream. Clean up can basically
be divided into two main.types: in-plant housekeeping and end of shift
clean up.
439
-------
DRAFT
Housekeeping - Housekeeping is the most continuous of clean up steps.
The various operations throughout a typical low meat pet food plant
generate considerable amounts of scrap. Included in this would be various
spillages from gravy tanks, can filling, meat thawing, grinding, etc.
Grinders as well as mixing tanks, filler bowls, double seamers, etc.,
also require periodic washdown to comply with in-plant and regulatory
sanitation requirements. All of these individual operations contribute
heavily to the organic waste load. These streams are characterized by
small pieces of grain, starches, blood, meat scraps, and other formulation
ingredients. These waste streams constitute a major portion of the total
plant effluent.
End of Shift Cleanup- End of shift clean up is to some extent similar
to the daily minute-to-minute operations inasmuch as all the floor and
equipment surfaces are thoroughly washed and rinsed. Additionally, however,
the larger cooking kettles are typically "boiled-out" with the aid of
detergents. Pipes may be disassembled and scrubbed with brushes. Large
pieces of equipment such as extruders, grinders, screw conveyors, etc.,
may also receive a final "sanitizing" step. These types of cleaning opera-
tions are usually responsible for peak loadings and probably contribute
an equivalent amount of pollutants as would be experienced by an entire
shift of housekeeping washdowns.
Retort Cooling Water
The only other process contributing to the wastewater stream is retort
cooling water. The water which is used to cool the cans is basically
low load water, typically continuously circulated, although some plants
were observed to discharge this segment directly under NPDES permit.
Some of the plants not only recirculate cooling water but reuse it for
clean up, but this was atypical of the plants visited. No quantitative
data are available to determine its relative proportion in the waste stream.
Model Plant
The model plant is one that produces 159 kkg of finished product generating
0.147 mgd of wastewater. The average BOD loading as shown in Table 73 is
3.55 kg/kkg with a range of 1.62 to 7.82 kg/kkg. The average BOD concen-
tration is 1,130 mg/1 with a range of 497 to 2,560 mg/1. The reason for
the wide variation in concentration is principally due to the various
product styles and types found within this subcategory. The other
flow related parameters follow this same pattern.
SUBCATEGORY B 6 - HIGH MEAT CANNED PET FOOD
General Plant Cleanup
Clean up in a high meat canned pet food plant is a continuous, minute-
to-minute process which contributes by far the largest share of both volume
and pollutants to the wastewater stream. Clean up can basically be divided
into two main types: in-plant housekeeping and end of shift clean up.
440
-------
DRAFT
TA3LE73 . RAW WASTE SUMMARY
LOW MfcAT CANNED PE.T FOOD
PARAMETER
PROD KKG/DAY
(TON/DAY)
SHIFT TIME HR/OAY
FLOW VOLUME MGO
FLOW RATE L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TON)
5 DAY 300 MG/L
RATIO KG/KKG
(L8/TON)
T3S MG/L
RATIO KG/KKG
(L9/TON)
NO PLANT LOG Mt'AN
9 159
175
9 19.8
9 (J.l<*7
8 9.36
1<»8
11 5150
755
11 1130
3i55
7.11
11 a<*5
2.66
5.33
MINIMUM
/6.9
^.3
12.0
0.059
<4,71
7*4.7
1750
kZQ
W7
1.62
3.23
397
1.21
Z.k2
MAXIMUM
328
362
2^4.0
0.371
18.o
295
5670
1360
2560
7.62
15. b
1800
5.b6
11.7
PROCESS CODECS): <*7N6<*S ,^7N5^I ,
-------
Housekeeping - Housekeeping is the most continuous of clean up steps.
The various operations throughout a typical high meat pet food plant gener-
ate considerable amounts of scrap. Included in this would be various
spillages from gravy tanks, can filling, meat thawing, grinding, etc.
Grinders as well as mixing tanks, filler bowls, double seamers, etc.,
also require periodic washdown to comply with in-plant and regulatory
sanitation requirements. All of these individual operations contribute
heavily to the organic waste load. These streams are characterized by
small pieces of meat, fat, starches, blood, and other formulation ingredir
ents. These waste streams constitute a major portion of the total plant
effluent.
End of Shift Cleanup - End of shift clean up is to some extent similar
to the daily minute-to-minute operations inasmuch as all the floor and
equipment surfaces are thoroughly washed and rinsed. Additionally, however,
the larger cooking kettles are typically "boiled-out" with the aid of
detergents. Pipes may be disassembled and scrubbed with brushes. Large
pieces of equipment such as extruders, grinders, screw conveyors, etc.,
may also receive a final "sanitizing" step. These types of cleaning opera-
tions are usually responsible for peak loadings and probably contribute
an equivalent amount of pollutants as would be experienced by an entire
shift of housekeeping washdowns.
Retort Cooling Water
The only other process contributing to the wastewater stream is retort
cooling water. The water which is used to cool the cans is basically
low load water, typically continuously circulated, although some plants
were observed to discharge this segment directly under NPDES permit.
No quantitative data are available to determine its relative proportion
in the waste stream.
Model Plant
The canned high meat pet food subcategory is characterized by several
different product styles as described in Section III. The processing
and meat handling techniques are diverse, and as such the data presented
show extreme ranges for concentrations and loadings for all of the flow-
related parameters.
The model plant is one that produces 167 kkg of finished product generating
0.179 mgd of wastewater. The average BOD loading as shown in Table 74
was 48.6 kg/kkg with a range from 29.2 to 80.8 kg/kkg. The average BOD
concentration was 11,800 mg/1 with a range of 6,910 to 20,200 mg/1. The
reason for the wide variation in concentrations is principally due to
the various product styles found within this subcategory. The other flow-
related parameters follow this same pattern.
442
-------
DRAFT
TABLE 74. RAW WASTE SUMMARY
HIGH MEAT CANNED DOG AND CAT FOOD
PARAMETER
PROD
KKG/DAY
(TON/DAY)
SHIFT TIME HR/DAY
FLOW
FLOW
FLOW
VOLUME MGD
RATE L/SEC
(GAL/MIN)
RATIO L/KKG
(GAL/TON)
5 DAY BOD MG/L
RATI
TSS
RATI
0 KG/KKG
(LB/TON)
MG/L
0 KG/KKG
(LB/TON)
NO PLANT LOG MEAN
3 167
184
2 24
3 0
2 7
124
3 4120
987
3 11,800
48
97
3 9130
37
75
.0
.179
.84
.6
.2
.6
.1
MINUMUM
153
169
--
0.
7.
123
3820
917
6910
29.
58.
2520
11 .
22.
178
78
2
4
1
2
MAXIMUM
182
201
--
0
7
125
4430
1060
20,200
80
162
33,100
127
254
.180
.90
.8
PROCESS CODE(S): 47N79W , 47N79I , 47N63H
443
-------
SUBCATEGORY B 7 - DRY PET FOOD
General Plant Clean Up
Clean up in a dry pet food plant is generally a combination of both dry
and wet methods with each coming at different times within a processing
period.
Wet Clean Up - Wet clean up generally consists of periodic floor washings
along with some "end-of-production" equipment clean.up. At "start-up" of
a production run, some "off-test" material is usually generated. The
excess is generally scooped away, but the floor areas around the extruder/
expander equipment is typically washed. Similarly, in some plants, the
fat application areas were observed to be periodically wet-cleaned to
maintain sanitary conditions. Some plants had pre-blending or tempering
chambers in which water or steam was added to the pre-mixed grains before
the extruders. These chambers were periodicaly scrubbed and rinsed.
The principal components discharged are bits of grain, finished product,
and minute fat coated particles. Volume, however, from these clean up
operations is generally minor relative to non-contact cooling water.
Dry Methods - Dry pet food is essentially a blend of dry ingredients to
which water or steam has been added to facilitate the extruding/expanding
process. As such, most of the periodic, housekeeping type clean up involves
handling dry or semidry materials which have been lodged between pieces
of equipment or have fallen on the floor. Continuous dry clean up is
a necessity for good housekeeping.
Non-Contact Cooling Water
The largest source of water in the manufacture of dry pet food is non-
contact cooling water and steam condensate from the extruder/expander
operation. This water acts as a dilutor for the clean up water, the
results of which are very low waste loads in terms of the various flow-
related parameters.
Model Plant
Dry pet foods are typically manufactured with similar equipment and proces-
sing techniques. As a result, the waste loadings and concentrations (with
the exception of plant 47D61I) show limited and predictable ranges. The
model plant produces 21"! kkg/day of finished product with a resulting
effluent 0.019 mgd. As can be seen from Table 75 the flow ratio is
only 155 1/kkg is an indication of the small amount of waste loads
from these plants.
Average BOD loading was .032 kg/kkg with a range from .011 to .096 kg/kkg.
The average BOD concentration was 202 mg/1 with a range of 51 to 796 mg/1.
The other flow-related parameters follow the same pattern as described
above.
444
-------
DRAFT
TA3LE 75. RAW WASTE SUMMARY
DRY DUG AND CAT FOOD
PARAMETER
P*OD KKS/OAY
(TON/DAY)
SHIFT TIME HR/OAY
FLOW VOLUME MGD
FLOW KATF L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
-------
UKAhl
SUBCATEGORY B 8 - SOFT-MOIST PET FOOD
General Plant Clean Up
Clean up in a soft-moist pet food plant is a function of the type of soft-
moist product style manufactured. The products which call for the direct
use of meat, fish, or poultry generally require more periodic cleaning
than do the grain-based formulations. As is true in all of pet foods,
clean up can be divided into housekeeping and end of shift clean up.
Housekeeping - Housekeeping is the most continuous of.clean up steps.
The various grinding, mixing, extruding, and conveying operations generate
scraps of grain, meat, and finished product. Typically these are disposed
of by dry methods such as scoops, shovels, or brooms. Occasionally the
floors will be washed to remove minute particles which can't be removed
by scraping. These few uses of water contribute a small percentage of
flow and pollutants to the waste stream.
End of Shift Clean Up - End of shift clean up with regards to soft-moist
production is generally end of production dry clean up. At this time,
grinders, augers, mixing tanks, extruders, conveyors, etc., are completely
and thoroughly washed with detergents. A final sanitizing rinse sometimes
follows. This type of cleaning generates a peak flow and loading condition
which is generally responsible for a majority of the flow and almost all
of the pollutants.
Non-Contact Cooling Water
The only other source of water used in the production of soft-moist pet
food is extruder cooling water or condensate from an expander. Flows
vary widely according to the type of process. No quantifying data are
available to further delineate these effluent streams. In some plants,
these non-contact cooling waters were observed to be discharged directly
under NPDES permit.
Model Plant
The model soft-moist pet food plant produces daily 51.4 kkg of finished
product while generating an effluent of 0.017 mgd. As shown in Table 76
average BOD loading is 6.73 kg/kkg with a range from 6.28 to 7.20 kg/kkg.
The average BOD concentration was 4600 mg/1 with a range of 3420 to 6200
mg/1. The reason for the wide range of concentrations is due principally
to the surges of water attributable to the various clean up cycles within
varied time spans. The other flow-related parameters follow the same
pattern as described above.
446
-------
DRAFT
TA9LE 76. RAW WASTE SUMMARY
SOFT MOIST DOG AND CAT FOOD
PARAMETER
PROO KKG/DAY
(TON/DAY)
SHIFT TIME HS/DAY
FLOW VOLUME MGD
FLOW PATE L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TON)
5 DAY BOO MG/L
RATIO KG/KKG
(L3/TON>
TSS MG/L
*<\TIO KG/KKG
(L9/TON)
NO PLANT
2
2
2
2
2
2
2
LOG MEAN
51. «»
56.7
16.0
0.017
1.33
21.1
1^60
350
<*600
6.73
13. ^
1090
1.60
3.19
MINIMUM
2.02
2.23
8.00
—
0.067
1.05
iaio
2
-------
LJKAFT
MISCELLANEOUS AND SPECIALTY PRODUCTS
SUBCATEGORY A 29 - FLAVORING EXTRACTS
As discussed in Section III, it has been determined that a typical flavor
manufacturing plant produces flavoring extracts which are subsequently
combined with other extracts and/or ingredients to produce finished
specific flavors. Natural extracts are produced by vacuum distillation,
solvent extraction, or expression of whole plants, plant parts, or plant
essential oils, while synthetic extracts are produced by the combination
of ethyl alcohol and organic acids. A discussion of the waste streams
which would be expected from the manufacturing of finished specific
flavors is presented below.
Vacuum Distillation
Wastewater generated by the vacuum distillation of essential oils and
plant tissues consists of still bottoms. The still bottoms from dis-
tillation of essential oils would be expected to contain terpenes while
distillation of plant tissues would result in remnant tissue in the
still bottoms.
Solvent Extraction
There is no wastewater generated from the solvent extraction of plant
tissues. All installations participating in the study indicated that
solvents were recovered and that spent plant tissue was hauled to
landfill.
Expression
The expression of essential oils from fruits generally results in the
generation of fruit water and spent fruit tissues. Fruit water becomes
part of the plant waste stream and spent plant tissues are generally
sold for production of pectin (citrus fruit only) or sold as cattle
feed.
Synthetic Flavoring Extracts
The organic synthesis of solvents such as ethyl alcohol, methylene
chloride, benzene, and toluene, with organic acids results in the
production of synthetic flavoring extracts. Based on available in-
formation, there appears to be no wastewater generated in this pro-
cess other than equipment cleanup.
Dehydration
The dehydration of flavoring extracts to produce dry concentrates
generates no process water other than cleanup since all liquid is
released into the atmosphere as vapor.
448
-------
DRAFT
Evaporation
The evaporation of flavoring extracts to produce concentrated flavors
generates no process wastewater other than cleanup since all evapo-
rated liquid is released into the atmosphere as vapor.
Finished Specific Flavor Blending Tanks
The blending of flavoring ingredients to produce finished flavors
generates no wastewater other than cleanup water.
Plant Cleanup
Plant 87E05 reported that organic synthesis tanks were cleaned either
by hot water flushing or steam, while stills and extraction tanks were
steam cleaned. The waste streams from the cleaning of organic syn-
thesis and solvent extraction tanks contain a certain amount of sol-
vents. The cleanup waste stream from the stills would not be expected
to contain toxic solvents unless the flavoring extract distilled had
been initially produced by organic synthesis or solvent extraction.
Plants 87E03 and 87E05 both segregate these three cleanup waste streams,
along with still bottoms, from the remainder of the plant effluent.
The cleanup of finished flavor mixing tanks is generally done between
flavor changes and consists of a detergent wash followed by a final
rinse. Floors 1n the blending tank areas are hosed as needed to remove
spills and leaks from equipment connections.
Non-Contact Water
Non-contact condenser cooling water is generated in the vacuum dis-
tillation process. Boiler blowdown is another source of non-contact
water.
Total Plant Effluent
Based on the above considerations it may be concluded that the quantity
and quality of the wastewater generated from the manufacturing of
finished flavors could be dependent on the following factors:
1. If the flavoring extracts used in the manufacturing of
finished flavors are produced in-house or purchased.
Purchasers of extracts would generally require no dis-
tillation, solvent extraction, expression, or organic
synthesis equipment and consequently, the waste streams
from these processes would be eliminated.
2. The form in which the finished flavors are produced. A
plant producing dry flavor concentrates and/or concentrated
449
-------
flavors might conceivably have a smaller waste flow with a lower
pollutant loading, especially if dehydration equipment is cleaned
without use of water.
Wastewater characteristic data was obtained for two plants during the
course of this study. The average wastewater characteristics of plant
87E02 were determined to be as follows:
Flow 5.7 cu m/day (0.0015 MGD)
BOD 0.017 cu m/cu m
SS 0.0155 cu m/cu m
pH 7.4
The plant's production operations consisted of the following:
1. Production of natural vanilla flavoring from the alcohol
extraction of raw vanilla beans.
2. Production of finished specific flavors from purchased
flavoring extracts.
3. Production of spices by dry grinding and blending.
4. Production of certified colors.
The total wastewater flow was attributable to cleanup operations such
as washing blending tanks between flavor changes. The average flow
from the plant was estimated to be 5.7 cu m/day (0.0015 MGD) with a
range of 0 to 11.4 cu m/day (0 to .003 MGD).
The average wastewater characteristics of plant 87E03 were determined
to be as follows:
Flow 125 cu m/day (0.033 MGD)
BOD 0.56 cu m/cu m
SS 0.054 cu m/cu m
pH 7.1
The production operations at this plant consisted of the following:
1. Production of synthetic flavors by organic synthesis.
2. Purification of essential oils by vacuum distillation to
produce standard extracts.
3. Blending of flavoring materials to produce finished
specific flavors.
The wastewater from the organic synthesis and vacuum distillation pro-
cesses was segregated from the rest of the waste stream, neutralized,
450
-------
DRAFT
and contracted to a private service for ultimate disposal. According
to plant personnel the contracted waste is composed of "soluble organics1
and totals 23 cu m/week (0.006 nig/week). A similar waste generated at
plant 87E05 was reported to be composed of the following constituents:
still bottoms, methylene chloride, methyl ketone, methyl hydroxide,
toluene, benzene, and carbon aromatics.
Model Plant
Based on available information from industry, it appears that plant
87E03 is more typical of the industry than plant 87E02. Therefore,
plant 87E03 was selected as the model plant for Subcategory A 29 and
is illustrated in Figure 135. The major wastestreams generated at the
plant consist of still bottoms, and cleanup of stills, organic synthesis
tanks, and blending tanks. However, the wastestreams from the cleanup
of stills and synthesis tanks as well as still bottoms are segregated
from the remainder of the wastestream. All non-contact water is also
separated from the waste stream.
The wastewater characteristics of the-model plant are as follows:
Flow 125 cu m/day (0.033 MGD)
BOD 1350 mg/1
SS 130 mg/1
pH 7.1
SUBCATEGORY A 31 - BOUILLON
The process description of bouillon manufacturing was presented in Section
III and it was determined that equipment cleanup water constituted the
total wastewater flow from a bouillon manufacturing plant.
Equipment Cleanup
Plant 99Q01 conducts a daily plant cleanup of equipment used in bouillon
processing and this wastewater was found to have the following charac-
teristics:
Flow 114 cu m/day (0.03 MGD)
BOD 4200 mg/1
SS 192 mg/1
FOG 150 mg/1
pH 10.4
Plant 99Q02 which conducts periodic daily plant cleanup and weekly cleanup
of all equipment generated wastewater with the following characteristics:
Flow 720 cu m/day (0.19 MGD)
BOD 1610 mg/1
SS 239 mg/1
FOG 82 mg/1
pH 6.9
451
-------
DRAFT
ESSENTIAL OILS
ORGANIC SOLVENTS,
ORGANIC ACIDS
1
i
CLEANUP WATER
VACUUM
DISTILLATION
ORGANIC
SYNTHESIS
SYNTHETIC
FLAVORS
BOILER
1
NON-CONTACT
BOILER SLOWDOWN
CLEANUP
WATER
NATURAL
FLAVORS
STILL
BOTTOMS
TO PRIVATE
SANITATION SERVICE
CLEANUP WATER
BLENDING
CLEANUP WATER
GENERAL
PLANT CLEANUP
FIGURE 135
MODEL PLANT FOR SUBCATEGORY A 29
FLAVORING EXTRACTS
452
-------
DRAFT
Dry Cleanup
All plants contacted utilized dry cleaning and generated no wastewater in
packaging areas.
Total Plant Effluent
The total plant effluent is attributable to equipment cleanup and conse-
quently the waste characteristics presented for equipment cleanup also
apply for total plant cleanup.
Model Plant
The model plant for this subcategory is a hypothetical plant producing
bouillon products exclusively and is illustrated in Figure 136. The
plant operates 16 hours per day, five days per week with daily equip-
ment cleanup. The wastewater characteristics of the plant are as
follows:
Production: 7.3 kkg/day (8.0 ton/day)
Flow: 151 cu m/day (0.03 MGD)
BOD: 3000 mg/1
SS: 200 mg/1
FOG: 150 mg/1
All cleanup in packaging areas is done with air. A grease trap prior to
discharge from the plant is provided to decrease the fats and oil content
of the wastewater.
SUBCATEGORY A 32 - NON-DAIRY CREAMER
Based on processing information obtained during the course of this
study, the major source of wastewater generated in the manufacturing
of both liquid and powdered non-dairy creamer is determined to be equip-
ment cleanup water. Generally, clean-in-place systems are .used for
equipment cleanup. Minor contributors to wastestream quantity are
hosing of floors and wet scrubber discharge.
Clean-in-Place Systems
The clean-in-place systems used for equipment cleanup in non-dairy
creamer plants generally employ six cleaning cycles consisting of the
following sequential steps: (1) hot water pre-rinse, (2) caustic wash,
(3) chlorine rinse, (4) final rinse, (5) sanitization, and (6) air drying.
The quantity of water used in each of the six cycles is usually fixed
and thus water requirements are minimized. For plants of equal size
there is no indication that the quantity of water necessary for the
cleaning cycles would vary markedly. However, the frequency of cleaning
does vary, causing significant differences in wastewater quantity.
Within the industry three distinct patterns of CIP system cleanup exist:
453
-------
DRAFT
INGREDIENTS
1
MIXING
TANK
CLEANUP
WATER
OVEN
DRYING
CLEANUP
WATER
GRINDING
CLEANUP
WATER
PACKAGING
GREASE
TRAP
PLANT EFFLUENT
FIGURE 136
SUBCATEGORY A 31 MODEL'PLANT'
MODEL PLANT - BOUILLON MANUFACTURING PROCESS
454
-------
DRAFT
(1) cleanup at the end of each day, (2) cleanup at the end of each week
(plants operating 24 hours per day), and (3) cleanup at the end of each
processing cycle (plants which produce creamer on an irregular basis
and for varying lengths of processing). Assuming recycling of caustic
and acid rinse water, a typical plant will use about 7.57 cu m/day
(0.002 MGD) of water for each CIP system cleanup.
General Plant Cleanup
Liquid non-dairy creamer plants generally hose packaging area floors
continuously to remove product spills. Powdered non-dairy creamer
plants periodically hose floors in areas where spills of dry product from
equipment connections occur. The quantity of water used in hosing of floors
is unregulated in both cases.
Wet Scrubber
In the case of powdered non-dairy creamer manufacturing, wet scrubbers
are used over the spray dryers to prevent dispersion of fine particulates
into the atmosphere. The effluent from each scrubber at Plant 99NN01
is approximately 16,000 I/day (4000 gal/day). The pollutant characteristics
were determined to be:
BOD: 4.6 mg/1
SS : 7 mg/1
F&O: 0.1 mg/1
Non-Contact Water
A substantial amount of cooling water is needed in the manufacturing
of non-dairy creamer. Based on plant water intake minus the quantity
of wastewater generated, the quantity of non-contact cooling water and
boiler blowdown for a typical plant would be about 378 cu m/day (0.10 MGD)
One multi-product plant (99NN02) producing liquid creamer recycled cooling
water and only makeup water was needed. A powdered creamer plant (99NN01)
discharged non-contact cooling and boiler blowdown water separately from
the waste stream with the quantity estimated at 454 cu m/day (0.12 MGD).
Total Process Effluent
Plant 99NN01 producing only powdered non-dairy creamer generated a total
process effluent with the following average characteristics:
Flow: 56.8 cu m/day (0.015 MGD)
BOD : 1250 mg/1 (range 1000-15000)
SS : 415 mg/1 (range 355-475)
F&O : 250 mg/1 (range 227-275)
pH : 7.0 (range 6.8-7.2)
Plant 99NN02, producing liquid creamer in a multi-product facility,
generated wastewater with the following average characteristics:
455
-------
URAFT
Flow
BOD
SS
F&O
N
P
1800 cu m/day (0.47 MGD)
3000 mg/1
2200 mg/1
140 mg/1
15 mg/1
8.0 mg/1
Although it is not possible to determine which portion of these pollutants
is specifically attributable to the manufacturing of liquid creamer,
the data are presented to indicate that the wastewater from the plant
is nutrient deficient. This particular plant produces a wide variety of
products, each of which is composed of the same basic ingredients as
liquid creamer but in varying proportions. Therefore, it can be concluded
that the wastewater from a plant producing solely liquid creamer would
also be nutrient deficient.
Model Plant
The model plant developed for Subcategory A32 as illustrated in Figure
137 is a hypothetical plant which would produce either liqid or
powdered non-dairy creamer. The plant operates five days per week
with two eight hour shifts per day. Clean-in-place system cleaning is
conducted periodically as needed and at the end of each day and generates
approximately 7.57 cu m/day (0.002 MGD) of wastewater with recycling of caustic
and acid rinse water. If the plant produced liquid creamer, the only other
wastestreams generated would be hosing of floors in packaging areas and
other general plant cleanup amounting to about 56.8 cu m/day (0.015 MGD).
If the plant produced powdered creamer, two spray dryers would be needed
and therefore two wet scrubbers are necessary. Combined flow from wet
scrubbers would be 30 cu m/day (0.008 MGD). An additional wastewater genera-
tion of 26.4 cu m/day (0.007 MGD) would be generated by hosing of dry product
spills and general plant cleanup. In either case the total plant waste
effluent is approximately 64.3 cu m/day (0.017 MGD).
Non-contact water is discharged separately from the waste stream and amounts
to about 380 cu m/day (0.10 MGD). There is no recycling of the non-contact
cooling or boiler blowdown water. The proposed model plant would have
the following characteristics:
Production: 90 KKg (TOO ton) dry product
or 180 KKg (200 ton) wet product
Flow 64.3 cu m/day (0.017 MGD)
BOD 1100 mg/1
SS 440 mg/1
F&O 265 mg/1
N 5.5 mg/1
P 2.9 mg/1
pH 7.0
456
-------
DRAFT
PLANT
SUP!
WATER
3LY
CORN VEGETABLE
SYRUP OIL
^X^I NGRED I ENTS ^S^
MIXING
CIP
i
^
PASTEURIZATION
- CI^ _J
•
1
HOMOGENIZATION
CIP. JL
-s
POWDERED CREAMER OR | LIQUID CREAMER
f
DRYING
BOXES
/fv
T
HSPRA
DRYE.
I
SHAKE
* COOLB
1
f
:?
«
RS (
SIFTING
•1
CLEANUP
.____. «y|
WET
SCRUBBER
H^.~— -^
FLOOR
HOSING
i
PACKAGING!
1
NOTE t
1
L
i> .
PLATE | CIP -I
COOLER I I
1
HOLDING CIP-J
TANK |
i
CIP
PACKAGING . .£
1
WAREHOUSE
EITHER POWDERED flB. LIQUID CREAMER
TO PTM-ITM ire:r> unT on-vn-i
WASTEWATER
EFFLUENT
FIGURE 137
SUBCATEGORY A32 - MODEL PLANT
'NON-DAIRY CREAMER MANUFACTURING
457
-------
DRAFT
SUBCATEGORY A 33 - YEAST
The use of water in yeast factories includes: (1) feed wort preparation;
(2) fermenter start water; (3) sterilization of molasses and feed wort
tanks, fermenterSs, and piping; (4) separation wash water; (5) cleaning
of separation and dewatering equipment; (6) miscellaneous floor and
equipment cleanup; (7) cooling water; and (8) boiler feed. Considering
strength and volumes, the wastestreams from yeast production can be
ranked in the following order: (1) first separation beer, (2) second
separation beer, (3) third separation beer, (4) filtration water from
yeast dewatering; (5) fermenter and storage tank cleanup water; and
(6) floor and equipment cleanup water.
Table 77 shows the pollutant loads of the above operations at a typical
plant (99Y03) producing 76.5 kkg/day (84.3 ton/day). First separation
beer accounts for 43 percent of the total flow, 78 percent of the BOD9
and 31 percent of suspended solids at this plant. The high strength waste
of combined first and second separation beer account for 92, percent of
the total flow, 90 percent of the BOD and 58 percent of the suspended solids
reported from in-plant sampling. Third separation beer is reused for cold
washing during second separation.
i
Rudolfs and Trubnick (86) reported that first separation beer at a similar
plant (99Y01) producing 82.2 kkg/day (90.6 ton/day) was responsible for
approximately 70 percent of the plant raw waste load. The BOD of spent
beer may vary from 2000 mg/1 to 159000 mg/1. Wide variations in flow also
occur as the result of different water usage by individual plants during
centrifugal separation of yeast from spent nutrients.
Third separation beer was reused in the second separation by 66 percent
of plants supplying data, since it contains only a small portion of plant
waste. First and second separation beer typically account for 50 percent
of the flow and 75 percent of the BOD and SS plants that do not reuse
process water.
Discharges from yeast dewatering consist of water removed from the yeast
cream by rotary vacuum filters and recessed-plate filter presses. Table 78
presents the wastewater characteristics for five composite samples (Plant
99Y03) dewatering operations. Filter discharges, containing varying amounts
of yeast and spent filter aid9 cause substantial daily fluctuations in
strength. Quantities of water discharged depend upon production levels
and the moisture content of the final products but are generally less than
10 percent of plant flow.
Cleanup of fermenters and feed wort storage tanks is normally performed
using hot water and steam between batch operations to prevent bacterial
contamination during fermentation. Molasses storage tanks are cleaned
weekly using clean-in-place systems with hot water and a 3 percent sodium
hydroxide solution. Tank cleanup varies according to cleaning techniques
and equipment, and the age and size of the plant storage facilities, but
458
-------
TABLE 77
YEAST.PLANT 99Y03
UNIT OPERATIONS WASTEWATER CHARACTERISTICS
en
vo
Operation
First Separation
Second Separation
Third Separation
Tank Washdown
Yeast Dewatering
Flow
1,
1,
(cu m/day)
008
132
5290)
79
109
% Total
43
49
—
3
5
6
7
6.
PH
.8
.0
8
5.8-13.6
6
.8
Bod (kg/day)
8,656
1,317
324
571
191
% Total
78
12
3
5
2
SS (kg/day
317
273
142
121
159
j % Total
31
27
14
12
16
TOTAL
2,328
100
11,059
100
1,012
100
(1) Third separation wash reused in second separation.
-------
TABLE 78
YEAST DEWATERING EFFLUENT CHARACTERISTICS
PLANT 99Y03
Day Flow Production COD BOD Ratio SS SS Ratio COD COD Ratio COD/COD
(cu m/day) (kkg) (mg/1) (kg) (kg/kkg) (mg/1) (kg) (kg/kkg) (mg/1) (kg) (kg/kkg) Ratio
1
2
3
4
5
290.0
377.0
492.0
95.4
307.3
103
76.5
85.0
85.4
99.3
480
700
1780
1360
960
140
263
876
130
295
1.4
3.5
10.3
1.5
3.0
960
680
1320
1540
1080
30
256
650
141
332
0.29
3.4
7.6
1.7
3.3
2880
1532
3210
.3085
2410
835
578
1579
294
741
8.1
7.5
18.6
3.4
7.4
0.17
0.46
0.56
0.44
0.40
Average 312.3 89.9 431 3.9 287 3.2 805 9.0 0.41
-------
TABLE 79
WATER USAGE AND WASTEWATER CHARACTERISTICS
YEAST PUNTS RECYCLING SEPARATION WATER-PLANTS 99Y01, 99Y05
VARIABLE
Flow (M6D)
Prod, (ton/day)
BOD (mg/1)
SS (mg/1)
COD (mg/1)
BOD (Ib/day)
COD (Ib/day)
SS (Ib/day)
Lb/ton-BOD
Kg/kkg-BOD
Lb/ton-COD
Kg/kkg-COD
Lb/ton-SS
Kg/kkg-SS
BOD/COD Ratio
Flow Ratio
N
126
126
126
126
1
126
1
126
126
126
1
1
126
126
1
126
MEAN
0.693389
90.408730
6262.944444
1822.230159
14602.000000
36214.451772
91999.535950
10570.723429
401.560349
200.780174
1091.334946
545.667473
116.654232
58.427116
0.461649
7679.939779
STANDARD
"DEVIATION
0.083141
3.742085
1023.914718
999.185341
0.0
7280.916476
0.0
6041.406934
84.916591
42.458296
0.0
0.0
66.615573
33.307787
0.0
958.015079
VARIANCE
0.0069
14.0032
1048401.3489
998371.3466
0.0
53011744.7306
0.0
36498597.7454
7210.8275
1802.7069
0.0
0.0
4437.6346
1109.4086
0.0
91 7792 ;8924
MINIMUM
0.440000
69.200000
3600.000000
420.000000
14602.000000
19527.300000
91999.535950
2593.626000
202.986486
101.493243
1091.334946
545.667473
28.532739
14.266370
0.461649
4695.837780
MAXIMUM
0.910000
96.200000
10800.000000
8900.000000
0.0
68495.760000
91999.535950
52732.055000
794.614385
397.307193
1091.334946
545.667473
584.612583
292.306296
0.461649
9790.979098
COEFFII
CONVAR
11.991
4.139
16.349
54.833
0.0
20.105
0.0
57.152
21.147
21.147
0.0
0.0
57.007
57.007
0.0
12.474
N = Number of data points. Note: Computer calculations for this table show no regard for significant figures.
-------
DRAFT
it typically generates less than 5 percent of the total flow, 5 percent
of the total BOD, and 15 percent of the suspended solids.
Floor and equipment cleaning is performed as needed to maintain bacterio-
logical cleanliness. Hot water and occasional small amounts of detergent
or caustic are used to clean molasses clarifiers9 centrifugal separators,
filters, and packaging equipment. Cleanup effluent is a small part of
combined plant waste.,
Table 79 presents a statistical analysis of combined plant raw effluent
data for plants that reuse third separation beer during second separation.
Although flow was found to range as high as 6800 cu m/day (1.8 MGD) in
a plant not recycling separation water9 the generation of pollutants per
unit of production varied less than 10 percent for similar production
levels. The two largest producers both reported an average of 170 kg/kkg
(340 lg/ton) of BOD9 while suspended solids varied from 50 kg/kkg (100 lb/
ton) to 76 kg/kkg (152 Ib/ton).
Yeast effluents (86)9 composed predominantly of highly putrescible dis-
solved organic waste substances, have a specific yeasty odor that rapidly
becomes unpleasant (87)s a coffee color9 and fairly high turbidity. The
wastewater contains yeast cells, fatty residue., albumens9 and their de-
composition products and carbohydrates. Inorganic compounds include9
phosphates9 large amounts of potassium, and sulphates. These effluents
putrefy easily as sulphates are biologically reduced to sulphides, and
require oxygen for stabilization in much the same manner as domestic
sewage. They are usually acidic since a pH of 4.5 is maintained during
fermentation. The pH of a total plant effluent samples collected during
this study ranged from 4,2 to 7.79 but more typical values were in the
range of 6.0 to 6.8
Since molasses is deficient in nitrogen and phosphorus, ammonia, and
phosphoric acid are both required chemical nutrients added in fermentation.
After yeast growth, the effluents from production are again nutrient
deficient. Analyses (79) of similar spent molasses in the rum industry
found the distillery slope to have a 94 percent phosphorus deficiency and
a 56 percent nitrogen deficiency. Plant 99Y20S operating an oxygen
activated sludge treatment system9 adds ammonia and 227 I/day (60 gal/
day) of 70 percent phosphoric acid before treatment.
Node! Plant
Based on the above discussion, a model plant for Subcategory A33 is
defined as follows:
Production 82 Kkg/day (90,4 ton/day)
Flow 2650 cu m/day (0.7 MGD)
BOD 6300 mg/1
SS 1850 mg/1
It is assumed that the model plant practices reuse of Third separation
spent beer, and that first and second separation beer constitute 50
percent of plant flow and contribute 75 percent of the BOD and suspended
solids of raw waste.
462
-------
DRAFT
SUBCATEGORY A 34 - PEANUT BUTTER WITH JAR WASHING
The uses of water in peanut butter processing plants include: 1) jar
washing, 2) floor and equipment cleanup, 3) cool ing.water, 4) boiler feed,
and 5) vacuum seal water. Peanut butter is immiscible in water, and does not
not require the addition of water to the product during processing. In
fact, bacteriological cleanliness demands special attention to insure that
water does not enter the interior of pumps, piping, and other process
equipment. Water use varies widely for individual plants due to produc-
tion requirements, and dissimilar water conservation and recycling tech-
niques. For example, grinder cooling water may be discharged directly
after use or recirculated through cooling towers. Table 80 presents a
breakdown of water usage per operating day by a plant (99P21) producing
59 to 77 kkg/day (65 to 85 tons/day) and demonstrates that over 98 percent
of all water used does not contact the product.
Sources of polluted wastewater from peanut butter production can be ranked
in the following manner: 1) jar washer discharges, and 2) floor and equip-
ment cleanup discharges. All of the plants surveyed dispose of jar washer
effluent and cleanup related wastewater, mixed with substantial amounts
of non-contact water, to municipal sewer systems.
Jar Washing - In plants employing jar washing to reclaim glass for pack-
aging, the detergent rinse is normally discharged and constitutes the
major process waste stream. Plant 99Y20, producing 10 kkg/day (11 tons/
day) has a jar washer discharge of 680 1 (180 gal) per 500 jars washed,
and a maximum daily discharge of 2040 I/day (541 gal/day). Approximately
6000 jars/month are washed at this plant. Jar washer effluent is a low
volume, high strength waste that produces 10 gm (0.022 Ib) of BOD, 3.8 gm
(0.0081 Ib) of suspended solids, 125 gm (0.0275 Ib) of COD, and 4.5 gm
(0.01 Ib) of fats and oils per 510 gm (1.125 Ib) jar washed. Table 81
shows the calculated results of plant 99P20 jar washer effluent sampling
after correcting flow to account for non-contact water.
Bad product manually scraped from improperly filled or sealed jars is
sold as inedible oil stock. Variations in pollutant loading per unit of
production may be attributed to differences in the number and size of jars
washed, and the method of product removal from reclaimable glass.
The largest plant (99Y01) in the industry, producing 140 to 230 kkg/day
(150 to 250 tons/day), reported BOD concentrations nearly doubled and
suspended solids concentrations tripled while practicing glass reclama-
tion. Waste load data from this plant was not used in selecting a model
plant because the wastewater contained large, undetermined amounts of
non-contact water and resulted from the production of several products.
Floor and Equipment Cleanup - Other than jar washer effluent, floor and
equipment cleanup are the only other sources of process wastewater from
peanut butter production. Production facilities typically operate five
days per week, 24 hours per day. Floors in processing areas are normally
463
-------
DRAFT
TABLE 80
APPROXIMATE WATER USAGE PER OPERATING DAY
FOR PEANUT BUTTER PROCESSING PLANT 99P21
VOLUME
SOURCE LITERS GALLONS
Cooling Towers 37,000 9,700
Cooling of Refrigeration and
Air Compressors 16,000 4,300
Boiler Feed Water 9,400 2,500
Sanitary 16,000 4,200
Cleanup and Miscellaneous 1,100 300
With evaporation loss, estimated
discharge 65,000 17,000
464
-------
DRAFT
TABLE 81
JAR WASHER WASTEWATER CHARACTERISTICS PLANT 99P20
Flow
BOD
BOD Ratio
COD
COD Ratio
SS
SS Ratio
FOG
FOG Ratio
2040 I/day (540 gal/day)
7320 mg/1
1.41 kg/kkg (2.82 Ib/ton)
9150 mg/1
1.77 kg/kkg (3.53 Ib/ton)
2810 mg/1
0.58 kg/kkg (1.15 Ib/ton)
3550 mg/1
0.69 kg/kkg 0-37 Ib/ton)
465
-------
scrubbed daily using a small quantity of water and detergent, and the
water is collected using mops or vacuum equipped floor scrubbers.
Water use for equipment cleanup is typically less than 757 I/day (200
gal/day) and is normally sewered. One plant (99P13) reported an esti-
mate based on hose flow rates of 2710 I/day (715 gal/day) used for
cleanup. Table 82 lists cleanup frequency and quantities of water
used for equipment cleanup by a typical plant (99P21). Periodic equip-
ment cleanup occurring at weekly or less frequent intervals is usually
done using steam hoses in a specially designated area equipped with
grease traps on all drains. Equipment cleanup is performed between
shifts or on weekends, and normally is not done while production pro-
cesses are in operation. Plant cleaning procedures are subject to
occasional revisions due to equipment changes and constantly improved
programs of housekeeping and sanitation. Although no data is available
to document the strength of combined cleanup wastewater, an estimated 6.8
to 14 kg/day (15 to 30 Ib/day) of product is reported lost to sewers. Resid-
ual product clinging to equipment may contain up to three percent added
vegetable oil.
Model Plant
Based on the above discussion of wastewater characteristics, the follow-
ing model plant was defined:
Daily Jar Washer Effluent 2044 I/day (540 gal/day)
Avg. Daily Cleanup Effluent 757 I/day (200 gal/day)
Avg. Daily Flow2801 I/day (740 gal/day)
The model plant assumes separation of all domestic sewage and non-
contact water from the process wastewater. Since strength of cleanup
wastewater is unknown, no determination of combined waste strength can
be made.
SUBCATEGORY A 35 - PEANUT BUTTER WITHOUT JAR WASHING
The uses of water and wastewater characteristics for peanut butter plants
in Subcategory A 35 are identical to those in Subcategory A 34, except
that jar washing is not practiced.
Model Plant
The model plant is defined as follows:
Flow = 757 I/day (200 gal/day)
466
-------
SOURCE
TABLE 82
OCCASIONAL CLEANUP WASTEWATER DISCHARGED-PLANT 99P21
FLOW
1. Warehouse concrete floor scrubber
2. Production building wood floor
scrubber
3. Chunk equipment cleanup
4. Equipment exterior wipe-down
5. Equipment exterior wipe-down
6. Elevator conveyor bucket cleanup
7. Process line piping cleanup
8. Bucket and drip pan cleanup
9. 011 stock drum wash
10. Elevator conveyor bucket cleanup
11. Raw nut elevator conveyor cleanup
DETERGENT
1.1 1 liquid
Concentrate
None
Concentrate
Concentrate
MAna
Nnna "*
U/\nA
9 1 kn Dnrfriar
FREQUENCY
dally
2/week
I/week
I/week
I/week
1 luAalf
1 /mnn+it
1 /mAn+h
9/UOAF*
PER
CLEANUP
(1)
114
95
76
1136
379
1QO
A«C
1C! A
eca
0^3
ftft't
YEARLY
(cu m)
29523
9880
3936
59046
19708
OQ/11
117RR
1O1CQ
COT3
OOOO
1\19
BOO
(mg/1)
37600
28433
2267
3050
11766
COO
(mg/1)
85760
42346
6788
8464
37352
SS
(mg/1)
89800
16600
2880
370
6460
FOG
(mg/1)
189 _
573
1217
126
399
pH
10.8
8.0
6.1
11.5
9.9
-------
DRAFT
SUBCATEGORY A 36 - PECTIN
As described in Section III, there are two methods of manufacturing
pectin; precipitation by alcohol and precipitation by use of aluminum
compounds. The characteristics of each waste stream generated in the
alcohol precipitation process at plant 99K01 are presented in Table 83.
The waste stream characteristics of plant 99K02, which uses aluminum
precipitation in the recovery of pectin, are summarized in Table 84.
Comparison of similar waste streams from the two plants yields the
following observations:
1) The quantity of alcohol still bottoms generated per day
by plant 99K01 is approximately 4.5 times greater than
at plant 99K02. This is attributable to the fact that
more alcohol is used in the process at plant 99K01 and
therefore more still bottoms from the recovery of the
alcohol would be expected.
2) The amount of peel washwater generated at plant 99K02 is
greater than at plant 99K01 which is expected due to the
higher production at the former.
3) The quantity of general plant cleanup water is larger at
plant 99K01 than at plant 99K02 which is probably attrib-
utable to an unknown amount of cooling water included in
the waste stream of the former.
It should be noted that there is no evaporation of pectin solution prior to
precipitation at plant 99K02 and therefore no caustic wash waste stream is
generated. In contrast the pectin mother liquor waste stream at plant
99K02 is not generated at plant 99K01 because this waste stream is ulti-
mately distilled for alcohol recovery at plant 99K01 and as a result be-
comes a portion of the alcohol still bottoms. This observation supports
the previous comparison of still bottom waste streams. Additionally,
press liquor wastewater at plant 99K02 is generated when filter sluice is
pressed to separate water from diatomaceous earth.
The wastewater analysis for the total plant effluent from three plants
(99K01, 99K02 and 99K03) is presented in Table 85. It should be noted
that the alcohol still bottoms and filter sluice waste streams at plants
99K01 and 99K02 were not considered in arriving at the figures presented.
Plants 99K02 and 99K03 showed good agreement between waste flow generated
per unit of product produced. The slightly higher flow figure at plant
99K01 can be partially attributed to an undeterminable amount of non-
contact cooling water in the waste stream.
Model Plant
Based on the information presented above a model plant was chosen for
this subcategory. The plant operates 24 hours per day, 365 days per
468
-------
TABLE 83
WASTEWATER CHARACTERISTICS OF INDIVIDUAL WASTE STREAMS AT PLANT 99K01
to
Wastestream
1. Alcohol still
bottoms
2. Filter sluice
3. Peel washing
4. Evaporator caustic
wash
5. General cleanup,
non-contact cooling
water
Flow
cu m/day
170
223
424
0.0008
fifll
COD
mg/1
17,000
4,050
18,800
1,190
Rnn
TS
mg/1
19,200
4,500
20,800
29,700
Cl
mg/1
9,930
146
37
*( IP&
pH
0.8
6.5
4.5
12.3
*(7 n
Total (excluding
items 1 & 2)
1,105
7,521
7,981
25.3
6.04
* Estimate based on plant intake water.
-------
TABLE 84
WASTEWATER CHARACTERISTICS OF INDIVIDUAL WASTE STREAMS AT PLANT 99K02
Wastestream
1. Alcohol still
bottoms
2. Filter sluice
3. Peel washing (leach)
4. Pectin mother liquor
5. Press liquor
wastewater
6. General Plant
cleanup
Total (Excluding
items 1 & 2) 1,532 8,655 76.7 259.6 4.59
*Estimate based on plant intake water
Flow
cu m/day
37.9
757
662
492
189
189
COD
mg/1
2,800
3,200
14,600
2,150
11,425
2,000
Cl
mg/1
__.-
160
95
38
170
*(20)
N
mg/1
25
235
406
224
__••
pH
____
7.0
4.0
4.1
5.5
*(7.0)
-------
TABLE 85
SUMMARY OF WASTEWATER CHARACTERISTICS
Subcategory A 36 - Pectin
PLANT FLOW COD BOD SS CL pH
cu m/kkg kg/kkg kg/kkg kg/kkg kg/kkg
99K01 955 10,160 21.6 6.04
99K02 844 7,304 *(4,821) 64.7 4.59
99K03 821 3,476 1,753
* Estimate based on BOD:COD ratio of 2:3 at the plant.
-------
UKAI-T
year and has the following characteristics:
Production 1.8 kkg/day (2.0 tons/day)
Flow 1530 cu m/day (0.404 MGD)
BOD 4950 mg/1
SS 2100 mg/1
N 260 mg/1
pH 5.0 (4.6 to 6.0) range
The above characteristics are averages only and would be expected to vary.
Production is dependent on whether rapid set or slow set pectin is pro-
duced and whether the raw material used is dry or wet peel. It is assumed
that still bottoms, pressure filter cake sluice, wet spent peel, and non-
contact water are separated from the process waste stream.
SUBCATEGORY A37 PROCESSING OF ALMOND PASTE
There are currently four known processors of almond paste in the
United States. All four discharge their process wastewater to
municipal facilities. Results of a telephone survey to three plants
and one plant visitation indicate that the production of almond paste
contributes a relatively insignificant wasteload to the total waste-
load of the four multi-product processing plants. The production of
almond paste exists in combination with the production of a large
variety of other products such as nut pastes (i.e., pecan, walnut,
hazel nut, cashew, and apricot kernels), granulated nuts, and nut
toppings. The wastewater characteristics of almond paste processing
are currently unavailable for the following reasons: 1) the multi-
product plants contacted were unable to furnish historical data on
almond paste production alone, with the only available information
being that of the final combined products wasteload, 2) the actual
sampling of the almond paste production line was impractical due to the
combination of wastestreams from other product lines, and 3) produc-
tion data was unobtainable.
The industry has made no future plants for the construction of any
new almond paste processing plants and, as previously mentioned, dis-
charges its wastewaters to municipal facilities. Therefore, the
possibility of a future point source discharge from an installation
primarily engaged in the production of almond paste is minimal. Due
to a lack of information on the industry's product line, production
variability, and wastewater characteristics, the development of ef-
fluent guidelines for almond paste processing at this time is not
feasible.
472
-------
DRAFT
SUBCATEGORY B 1 - FROZEN PREPARED DINNERS
General Plant Clean Up
The wastes generated from these types of processing plants are a direct
function of the various raw ingredients used and subsequent handling steps
involved in transforming these ingredients into finished products. By
far the predominant waste loadings (flow, BOD, SS, COD, and oil and grease)
are generated during clean up. Sanitation requirements are such that
in-process clean up is virtually continuous with one large entire-plant
clean up performed at the end of each operating day.
In-Process Clean Up - The raw ingredients are usually pre-processed elsewhere
and are then further processed, cooked, assembled, packaged, and frozen
at the prepared dinner plants. Consequently, the majority of the wastes
from these types of operations originate from clean up of vats, kettles,
fryers, mixers, and other equipment used in the preparation. Included
in this group would be various spillages from gravy tanks, tray filling,
meat thawing, grinding, etc. In addition, equipment coming into contact
with food must be cleaned every four hours.
End of Shift Clean Up - Because of sanitary requirements, a complete plant
clean up is performed after each shift, and a general plant clean up is
undertaken at the end of each processing day. The floors as well as immov-
able equipment are cleaned, and this operation may involve the disassembling
of the equipment for a thorough cleaning and inspection. Included in
this type of equipment would be pipes, cooking kettles, infra-red cookers,,
extruders, and injectors. The wastes generated typically contain fine
particles and dissolved organics from each of the unit operations; consev
quently the pollutants generated may vary widely from day to day within
a particular plant, depending on the products produced. Contributing
to the waste stream's pollutants are the necessary chemicals and detergents
required to remove the various organic stains and residues from the various
units of processing equipment.
Defrost Water
The prepared dinners are assembled and then individually quick-frozen
and stored in large blast refrigerated warehouses, along with raw ingredients
awaiting movement to the preparation area. Because of the large capacity
of the storage facilities, a considerable volume of wastewater is generated.
The water which is used is basically low load water, typically continuously
circulated, although some plants discharge this segment directly under
NPDES permit.
Model Plant
The subcategory for frozen prepared dinners includes T.V. dinners, meat
pies, and other frozen dinners and entrees. Ingredients usually include
meat, fowl, or fish; vegetables; gravies; and minor additives. In addition,
473
-------
DRAFT
there may be added starches (such as noodles), grains (such as rice),
and a variety of small dessert dishes. The bulk of the wastes generated
originates from clean up of processing equipment.
The model plant is one that produces an average BOD loading of 15.6 kg/kkg
with a range from 9.41 to 25.9 kg/kkg as shown in Table 86 . The average
BOD concentration was 1530 mg/1 with a range of 718 to 3260 mg/1. The
wide range in concentrations was due largely to the product type and style
variations as outlined above. The other flow-related parameters follow
this same pattern.
SUBCATEGORY B 2 - FROZEN BREADED AND BATTERED SPECIALTIES
General
This subcategory has marked similarity to the other frozen specialties
for two important reasons. The first is the multiplicity and variation
of products within the subcategory - breaded fish fillets, shrimp, scallops,
mushrooms, onions, etc. Secondly, a majority of the waste loadings and
flows are a result of the extensive clean ups necessary for adequate sanita-
tion. In addition to plant clean up, a considerable volume of waste can
be generated from thawing and washing operations.
Thawing and Washwater
Thawing produces a substantial waste volume since it is followed by thorough
washing and clean up of equipment and spills. If the shells, heads, and
tails are included in the washwater, they constitute a major organic load
and should be removed as solid waste.
Frozen onion rings are by far the major item in battered and breaded vegeta-
ble specialties and a considerable portion of the wastewater may originate
from the onion washing operation. However, in most plants, they arrive
already washed.
Model Plant
The breaded and battered frozen specialty subcategory is characterized
by extreme ranges due to the various production techniques and raw materials
handled. Wastewater generation results from clean up of equipment and
spills, and juices from the onion slicing and washing operation. The
batter is very high in organic strength, and the clean up wastes are correr
spendingly strong. A process summary is presented in Table 87.
The model plant produces an average BOD loading of 16.2 kg/ kkg with a
range from 8.98 to 29.3 kg/kkg. The average BOD concentration was 1,350
mg/1 with a range of 244 to 7,510 mg/1. The wide range in concentration
was due largely to the product type and style variations as outlined above.
The other flow-related parameters follow this same pattern.
474
-------
DRAFT
TA9LE86 . RAW WASTE SUMMARY
FROZEN PREPARED DINNERS
PARAMETER
PROO KKG/OAY
ITON/OAYJ
SHIFT TIME HR/DAY
FLOW VOLUME MGO
FLOW RATE L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TON)
5 DAY BOO MG/L
RATIO KG/KKG v
(L3/TON)
f5S MG/L
RATIO KG/KKG
(L3/TONI
NO PLANT
5
6
5
5
6
b
6
LOG WEAN
86.3
95.1
24.0
0.253
11.1
176
10200
2*»50
1530
15.6
31.2
1150
11.7
23.5
MINIMUM
25.8
28. k
» M»
0.073
3.13
<»9.7
53*»0
1280
718
9.«tl
1«.8
548
6ol7
12.3
MAXIMUM
289
318
_.
0.897
39.3
623
19500
<+670
3260
25.9
51.8
2^20
22.
-------
UKMt I
TA3LE 87 . RAW WASTE SUMMARY
FROZEN BATTERED AND BREADED SPECIALTIES
PARAMETER
PRDO KKG/OAY
(TON/DAY)
SHIFT TIME HR/DAY
FLOW VOLUME MGO
FLOW RATE L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TON)
5 DAY BOD MG/L
RATIO KG/KKG
(L3/TONI
T3S MG/L
RATIO KG/KKG
tL3/TONl
NO PLANT
2
2
2
2
2 :
2
2
LOG MEAN
9.86
10.9
2^.0
0.032
1.37
21.7
12000
2870
1350
16.2
32. k
15
-------
DRAFT
SUBCATEGORY B 3 - FROZEN BAKERY ITEMS
General Plant Clean Up
The subcategory frozen bakery items includes an assortment of commodities
such as frozen pies, cakes, doughnuts, cheesecakes, sweet rolls, etc.,
utilizing ingredients and techniques as detailed in Section III which
are unique to this subcategory. The majority of pollutant loadings are
the result of clean up of the various mixing, extruding, and forming equip-
ment. The various cleaning techniques, additives, and detergents used
to remove hardened dough, eggs, milk solids, and the like contribute sig-
nificantly to the wastewater loadings.
In-Process Clean Up - The raw ingredients, e.g., butter, sugar, cream,
etc., are purchased in bulk, received, blended under controlled conditions,
further assembled in the final product form, sometimes baked, packaged,
and frozen. In order to maintain sanitary conditions, the frozen bakery
dessert plants must thoroughly clean with hot water all the many mixing
vats, cooking kettles, measuring devices, pumps, piping, etc., which have
come in contact with the ingredients and product. This clean up is contin-
uous during the shift as different products are manufactured.
End of Shift Clean Up - Because of sanitary requirements a complete plant
clean up is performed after each shift, and a general plant clean up is
undertaken at the end of each processing day. The floors as well as im-
movable equipment are cleaned, and this operation may involve the disas-
sembling of the equipment for a thorough cleaning and inspection. Also
included in this type of equipment would be pipes that are cleaned in
place as well as small mobile pieces used in batch preparations. The
wastes generated typically contain fine particles and dissolved organics
from each of the unit operations; consequently the pollutants generated
may vary widely from day to day within a particular plant, depending on
the products produced.
Defrost Water
The dessert items are assembled and then individually quick-frozen and
stored in large blast refrigerated warehouses, along with raw ingredients
awaiting movement to the preparation area. Because of the large capacity
of the storage facilities, a considerable volume of wastewater is generated.
The water which is used is basically low load water, typically contin-
uously circulated, although some plants discharge this segment directly
under NPDES permit.
Model Plant
The model plant for this subcategory would be one manufacturing frozen
dessert items including pies, cakes, pastries, and rolls. The bulk of
the wastes generated originates from clean up of processing equipment.
The model plant has an average BOD loading of 22.4 kg/kkg and the average
BOD concentration was 2,090 mg/1 as shown in Table 88. Average TSS loading
was 13.6 kg/kkg at a concentration of 1,270 mg/1.
d77
-------
DRAFT
TABLE 88. RAW WASTE SUMMARY
FROZEN 3AKOY PRODUCTS
PARAMETER
P^DD KKS/OAY
(TON/DAY)
SHIFT TIME HR/DAY
FLOW VOLUME MGO
FLOW RATE L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TON)
5 DAY BOO MG/L
RATIO KG/KKG
(L8/TON)
TSS MG/L
RATIO KG/KKG
(L3/TON)
NO PLANT LOG MEAN MINIMUM
1 50.9
56.1
1 2^.0
1 0 •!<»<»
1 6.31
100
1 10700
2570
1 2090
22. ^
kU.8
1 1270
13.6
27.2
MAXIMUM
— —
--
--
--
--
_ _
--
..
--
-- .
-_
--
"
PROCESS CODE(Sit 38C59L
478
-------
DRAFT
SUBCATEGORY B 4 - FROZEN TOMATO-CHEESE-STARCH COMBINATIONS
General Plant Clean Up
The processing of frozen tomato-cheese-starch items (frozen pizza, macaroni,
lasagna, ravioli, etc.) involves the combining of preprocessed ingredients
into the final product form. The principal waste generation step is plant
clean up. The cleaning procedures are similar to those described for
the frozen prepared dinner subcategory, and its clean up, with very little
modification, can be applied to the frozen tomato-cheese-starch subcategory.
Defrost Water
Refrigeration water is generally recycled, but, if not recycled, contributes
a significant volume of clean water to the waste stream.
Spillage and Clean Up
The types of pollutants generated by a plant are a direct function of
the various raw ingredients used and the subsequent handling steps involved
in transferring these ingredients into finished product. An efficient
plant can hold its waste ingredients to under one percent of the incoming
ingredient weight, e.g., loss of less than one pound of tomato paste used.
Model Plant
All major ingredients are preprocessed elsewhere and arrive at the manu->
facturing plant in bulk containers. These ingredients include tomato
paste, cheese, flour, milk, oil, noodles, seasonings, and meat. The waste
generated from plant clean up contributes the most significant portion
of the waste stream. A process summary is presented in Table 89.
The model plant is one that produces an average BOD loading of 18.8 kg/kkg.
The average BOD concentration was 239 mg/1. The average SS loading was
14.3 kg/kkg with a concentration of 180 mg/1.
SUBCATEGORY B 9 - PAPRIKA AND CHItT PEPPER~~"
The subcategory paprika and chili pepper consists of wet sampling data
from two plants -- 99C50W and 99C51W. As shown in Table ~§0, average
BOD loading was 8.44 kg/kkg with a range of 6.32 to 11.3 kg/kkg. The
average BOD concentration was 391 mg/1 with a range of concentrations
from 253 to 604 mg/1. SS and flow ratio parameters showed similar
consistencies.
Model Plant
The model plant for Subcategory B 9 was selected to have a flow of
2000 cu m/day (0.5 MGD) with the following characteristics:
BOD 400 mg/1
SS 250 mg/1
pH 6 to 9
N&P Sufficient
-------
DRAFT
TA3LE 89 •• RA>W WASTE SUMMARY
FROZEN TOMATO-CHEESE -STARCH DISHES
PARAMETER
PROD KKG/DAY
(TON/DAY!
SHIFT TIME HR/DAY
FLOW VOLUME MGD
FLOW RATE L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TONI
5 DAY 800 MG/L
RATIO KG/KKG
(LB/TON)
TSS MG/L
RATIO KG/KKG
(L3/TONI
NO PLANT
1
1
1
1
1
1
1
1
LOG MEAN MINIMUM
2.
-------
DRAFT
TABLE 90. RAW WASTE SUMMARY
CHILI PEPPERS AND PAPRIKA
PARAMETER .
PROO KKG/OAY
(TON/DAY)
SHIFT TIME HR/OAY
FLOW VOLUME MOD
FLOW RATF L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TON)
5 DAY 800 MG/L
RATIO KG/KKG
(L8/TON)
TSS MG/L
RATIO KG/KKG
(LB/TON)
NO PLANT
2
Z
2
2
2
2
2
LOG MEAN
104
lib
24.0
0.586
25.7
407
21600
5180
391
8.44
16.9
249
5.38
10.8
MINIMUM
97.8
108
.-
0.486
21.3
337
18700
4470
253
6.32
12.6
229
5.05
10.1
MAXIMUM
111
123
--
0.707
31.0
491
251)00
5990
60<«
11.3
22.5
271
5.7<*
11.5
CODE(S): 99C50W ,99C51W
481
-------
DKAFT
SUBCATEGORY C 4 - EGG PROCESSING
Liquid Egg Processing Equipment Cleaning
According to Siderwicz (88 ), cleaning of liquid egg handling equipment
is the largest source of vvastewater from egg processing plants. Virtually
all egg processors have clean-in-place systems for the cleaning and
sanitizing of their pasteurizing equipment, liquid egg holding tanks
and associated piping. This equipment is normally drained of egg
product as completely as po.-ilple before cleaning. The cleaning
is accomplished in three steps;' pre-rinse* washing, and rinsing. Some
egg processors have reduced their water consumption by recovering the
final rinse water and reusing it in the pre-rinse step of the next
cleaning cycle. The quantity of wastewater and the waste load from this
cleaning process depends on whether the egg product remaining in the
pipes after the pumps are shut off is discharged to the sewer or goes
to inedibles. No data is available to quantitatively define the waste-
water generated by these cleaning processes as opposed to an egg processors
total effluent.
Egg Breaker Wastewater
When a substandard egg is broken, the cup and sometimes the entire
breaking machine must be washed down. Siderwicz (88) indicates that the
washing of the egg breaking equipment is the second largest source of
wastewater flow and the third most important source of wastewater
strength. Schultz (89 ) reports that egg breaker wastewater from a
plant processing 70 kkg (64 tons) of eggs per day has a BOD of 4500
mg/1, a suspended solids concentration of 1000 mg/1 and a flow of 0.024
mid (0.006 mgd).
Egg Washing
Siderwicz (88 ) notes that egg washing is another important source of
wastewater volume and the second most important source of wastewater
strength. Egg washers wash eggs with a recirculating detergent/disin-
fectant solution and then rinses them with potable water. The rinse
water is added to the washer tank and provides a continuous overflow.
Every four hours the washer tank is dumped and refilled with fresh water.
Schultz (89 ) has reported that egg washer wastewater from a plant
processing 70 kkg (74 tons) of eggs per day has a BOD of 1450 mg/1, a
suspended solids concentration of 325 mg/1, and a flow of 0.017 mid
(0.004 mgd).
Plant Cleaning
General cleaning of egg processing plants is also a source of wastewater.
Some eggs fall to the floor during handling and must be scraped up,
mopped up or rinsed into a floor drain. All equipment and floors must
be cleaned periodically. The frequency of general plant cleaning varies
482
-------
DRAFT
from plant to plant, and the waste load varies dramatically, depending
on the housekeeping practices.
Combined Plant Effluent
Total discharge volumes from egg processing plants range from 0.015 to
0.53 mid (0.004 to 0.14 mgd). Total discharge per units of production
varies from 0.9 to 17.8 1 per kg (0.5 to 10 gal per Ib), with remarkable
differences in wastewater discharge for apparently similar operations.
Total production ranges from 4 to 85 kkg per day (4.4 to 94 tons per
day). The data collected indicated no relationship between the total
production per day and the total discharge per unit of production. The
BOD values of the total plant effluent from the plants surveyed ranged
from 1,800 to 8,600 mg/1 and the suspended solids concentrations ranged
from 540 to 1,600 mg/1. Table 91 is a summary of the plant effluent.
Model Plant
The model plant for this subcategory is a hypothetical egg processing
plant which produces frozen, liquid and dried egg products. The eggs
are trucked to the plant in 21 kg cases (30 dozen eggs). After a short
period of refrigerated storage, the eggs are loaded, candled, washed
and broken as described in Section III of this document. The eggs are
then pasteurized and frozen, dried, or sold as liquid egg. Total eggs
broken at the model plant in a 24 hr per day operation (including an
8 hr cleanup shift) is assumed to be 30 kkg per day (33 tons per day).
Wastewater - Sources of wastewater from the model plant would include
all sources listed above. Inedible eggs are recovered and sold or
handled as solid waste to help reduce the waste strength. Total
wastewater flow for the model plant is assumed to be 0.2 mid (0.05 mgd)
and flow per kkg of eggs broken is 6.5 1. Effluent BOD is 3,700 mg/1
and the effluent suspended solids concentration is 850 mg/1. Thus, the
waste load from the model plant will be 23 kg BOD and 5.4 kg SS per kkg
of eggs broken. It is also assumed that this model plant utilizes a
catch basin to remove shells from its waste stream. Some of the in-plant
technology described in Section VII is utilized by the model plant.
SUBCATEGORY C 5 - SHELL EGGS
Egg Washing
Egg washing is the major source of wastewater strength and volume from
shell egg plants. Egg washing machines use a recirculating disinfectant/
detergent solution for washing, which is followed by a potable water
rinse. The rinse water added to the washer tank provides a continuous
overflow. Every four hours the washer tank is dumped and refilled with
fresh water. Schultz (89) reported that continuous overflow from the
egg washer had a BOD of 935 mg/1 and suspended solids of 150 mg/1.
Samples taken from an egg washer tank during this study had BOD values
between 1800 and 3600 mg/1 and suspended solids values between 240
and 1400 mg/1.
483
-------
URAFT
TABLE 91 . RAW WASTE SUMMARY
EGG PROCESSING
PARAMETER
PROO KKG/OAY
(TON/DAY)
SHIFT TIME HR/OAY
FLOW VOLUME MGO
FLOW RATE L/SEC
(GAL/MIN)
FL3W RATIO L/KKG
(GAL/TOM)
5 OAY BOO MG/L
RATIO KG/KKG
(LB/TON)
T3S MG/L
RATIO KG/KKG
(LB/TON)
NO PLANT LOG MEAN
9 27.1
29.9
9 9.56
9 0.032
9 3.*»3
5«».<»
9 <*210
1010
9 3190
13.
-------
TA3LE 92 . RAW WASTE SUMMARY
SHELL EGGS
PARAMETER
P*OD KKG/DAY
(TON/DAY)
iHIFT TIME HR/DAY
FLOW VOLUME MGO
FLOW RATE L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/ TON)
5 DAY 800 MG/4.
RATIO KG/KKG
(L9/TON)
TJS MG/L
*ATIO KG/KKG
(LB/TON)
NO PLANT
7
7
5
7
7
7
7
LOG MEAN
21.7
23.9
8.60
C.CO<4
0.179
2.85
237
56.7
2U70
0.590
1.18
73 <+
0.171
0.3^2
MINIMUM
7.60
8.38
7.50
0.003
0.039
0.630
26.0
6.2<+
1200
0.061
0.121
164
O.Glb
0.031
MAXIMUM
62.0
6d.<4
11.3
0.005
0.817
12.9
2150
516
50*0
5.75
11.5
3230
1.96
3.91
PROCESS CODECS)! <+<»A50W
,'»'»A53L
,<»<4A5&W ,
485
-------
DRAFT
Plant Cleaning
General cleaning of shell egg plants is a significant source of wastewater
generation. Some eggs fall to the floor during handling and must be
scraped up, mopped up, or rinsed into a floor drain. All equipment and
floors must be cleaned perioj^lly. The frequency of general plant
cleaning varies from plant t| plant, and the wasteload varies dramatically,
depending on housekeeping or^ctices.
Total Processing Effluent
The quantities and characteristics of wastewater from shell egg plants
vary considerably. These variations are usually the result of operating
and cleanup procedures, which depend on the training and management
of the personnel. Wastewater flow per unit of production varies from
plant to plant, but is generally consistent within a given plant.
Table 92 includes data describing the total processing effluent for
this subcategory.
Model Plant
The model plant for this subcategory is a hypothetical shell egg plant.
The eggs are trucked to the plant in 21 kg cases (30 dozen eggs). After
a short period of refrigerated storage, the eggs are loaded, washed,
candled, graded, and packaged as described in Section III of this
document.
Operating procedures stress the recovery of inedible eggs for sale as a
component of animal feed or disposal as solid waste. The equipment
and floors are wet cleaned after recovery of the inedible egg product.
Total production at the model plant is assumed to be 12.5 kkg (14 tons)
per day produced' in eight hours per day, five days per week operation.
Wastewater - Sources of wastewater from the model plant include all
of the sources listed above. Inedible eggs are recovered and sold.
Total wastewater volume is assumed to be 0.013 mid (3500 gpd). It
is assumed that this model plant utilizes a catch basin or a large
0.6 cm (0.25 in.), mesh screen, to remove shells from the waste stream.
486
-------
SU8CATEGORY C 6 - MANUFACTURED ICE
The quantity of utter-that 1s wasted 1s the parameter of most concern.
In fragmentary Ice manufacturing the quantity of wastewater discharged
approximates the quantity of Water Incorporated Into the 1'cei The
range IB discharge is relatively narrow and Is not highly operator-
dependent. On the other hand> the quantity of water used and waste-
water discharged from block ice manufacturing has up to 20-fold variations
from plant to plant. These variations are primarily due to water con-
servation practices or lack thereof, and most of the variations in
water use are attributed to discharge of once-through cooling water.
The thrust of this program, however, 1s directed to process water and
the waste load 1n terms of kg of pollutant per kkg of product. There-
fore, the following discussion 1s directed to waste load rather than
discharge volume.
The concentration of pollutants from 1ce manufacturing 1s nominal.
Pollutants, if these constituents should be classified as pollutants,
consist predominately of dissolved solids (salts) with very low
suspended solids, BOD, and nitrogen concentrations. The concentration
of salts and suspended solids in the waste stream Is dependent on the
characteristics of the water supply. The water used in ice manu-
facturing must be potable, but if the water had a relatively high salt
and solids concentration, the concentration of these constituents in
the waste stream will be proportionately high.
The major sources of these pollutants are the following:
1. Water pretreatment, if required to remove suspended
solIds. Predominant treatment methods are lime,
sand filters, and carbon filters.
2. Core pumping. A number of block ice plants pump out
the unfrozen core water prior to complete freezing
of an ice block. This core water has a volume of 10
to 22 liters (3 to 6 gal) per block, and it contains
much of the solids and other impurities found in the
water supply.
3. Can dipping in a block ice plant is a source of a
small amount of salts. Pollutants in the waste stream
from can dipping are primarily brine remaining on the
exterior of the cans when they are removed from the
brine tank. However, prior to can dipping but after
lifting the cans from the brine tank, the cans are
suspended for several minutes to allow most of the
brine to drip back into the brine tanks. Chloride
concentrations in the dip tank are normally below
487
-------
DRAFT
that which would produce a salty taste 1n the water and
the solids are usually dissolved and of relatively low
concentration.
3. Slowdown from fragmentary ice making machines has
approximately twice the concentrations of dissolved
and suspended solids at the water supply.
4. Snow and end pieces generated by crushing, scoring, and
sawing block ice into sized or cube ice contribute re-
latively pure water with virtually no pollutants. Some
plants recycle this water for ice-making and others
discharge it as wastewater.
Total Processing Effluent
The quantity and quality characteristics of wastewater from ice manu-
facturing plants is relatively constant in any particular plant. The
same processes are used repeatedly in both block and fragmentary ice
production. Thus, the only variations in quantity or quality of the
wastewater come from variations in the product mix. Wastewater from
ice manufacturing plants is clean in comparison with other industrial
waste streams. Characteristics of the wastewater is similar.to those of
the water supply with slight to 100 percent increase in chloride and
dissolves solid concentrations. Table 93 includes data describing
the total processing effluent for this subcategory.
Model Plant
The model plant for this subcategory is a hypothetical ice manufactur-
ing plant producing both block and fragmentary ice. The block ice 1s
produced as described in Section III and core water is pumped from
the blocks. Both once- through compressor cooling water and core pump-
ing water are discharged to the waste stream. The fragmentary ice
machine is located in the same building as the block ice facility and
its waste is discharged to the waste stream. Average total productfoit
1s 17.2 kkg per day (19 tons per day). Production ts 24 bow* par
day, five days per week for six months a year.
- Sources of wastewater from the model plant include all
sources listed above.
Parameters of the wastewater are assumed as follows:
1. Flow volurae - average - 0.04 mid
minimum - 0.01 mid
maximum - O.T9 raid
2. BOD - 1.2 rng/1
11,000 gpd)
3,000 gpd)
488
-------
TA3LE 93. RAW WASTE SUMMARY
MANUFACTURED ICE
PARAMETER
PROD KKG/DAY
(TON/DAY)
SHIFT TIME HR/OAY
FLOW VOLUME MGO
FLOW RATE L/SEC
(GAL/MIN)
FLOW RATIO L/KKG
(GAL/TON)
5 DAY 300 MG/L
RATIO KG/KKG
(L8/TON)
TSS MG/L
RATIO KG/KKG
(L3/TON)
NO PLANT
3
3
3
3
3
3
3
LOG MEAN
17.2
19.0
17.3
0.011
0.69<«
11.0
2220
532
1.20
0.004
0.006
5.20
0.012
0.02<+
MINIMUM
10.5
11.6
6.00
0.003
0.138
2.18
619
l
-------
DRAFT
3. SS - 5.2 mg/1
4. 0.004 - kg BOD per kkg of product
5. 0.012 - kg SS per kkg of product
SUBCATEGORY C 12 - SANDWICHES
Cleanup and Total Combined Process Waste
General cleaning is the only source of wastewater generation in most
pre-packaged sandwich plants. General cleaning consists of washing hand
utensils in a sink or dishwasher, wiping off counter tops, and mopping
floors. These procedures are normally employed on a daily basis. Portable
chopping machines used in plants that blend salad-type sandwich fillings
are cleaned daily with a hose. The total volume of process wastewater
from the plants contacted ranged from 400 to 11,000 Ipd (100 to
3000 gpd).
Model Plant
The model plant for this subcategory is a hypothetical plant which
assembles a variety of pre-packaged sandwiches, All of the materials
from which the sandwiches are assembled are processed before delivery
at the sandwich plant. Total production at the model plant is assumed
to be 4.5 kkg (5 tons) per day produced in 8 hours per day, five days
per week.
Wastewater - Sources of wastewater from the model plant include general
cleaning of hand utensils, counter, tops and floors. The wastewater
flow from the model plant is 7,600d (2GOO;,g'al) per day.
Two days of sampling.were conducted at a major producer of pre-packaged
sandwiches. However, the samples were taken by an employee of the plant,
and apparently came from the surface of the grease trap. As a result,
the values obtained were not representative of the plant's wastewater.
SUBCATEGORY D 4 - VINEGAR'~
Wastewater characterization is based on data from four plants engaged in
the production of vinegar from apple products. Although vinegar is also
produced from grape products and purchased ethanol, no historical data
for processors utilizing these raw materials was available. Vinegar from
apple products represents the largest segment of the industry and is a
good representation of the industry as a whole. Table 94 summarizes
the data collected.
Water use in the vinegar plant is primarily in the filtration operation
with lesser amounts consumed for daily plant cleanup. Wooden holding
tanks, when not in use for vinegar storage, are filled with water to
avoid shrinking of the wood; draining of these tanks occurs as necessary.
490
-------
TABLE 94. RAW WASTE SUMMARY
VINEGAR
PARAMETER
PROD CU M/DAY
(1000 GAL/OAY)
SHIFT TIME HR/OAY
FLOW VOLUME MGD
FLOW RATE L/SEC
(GAL/MINI
FLOW RATIO L/CU M
(GAL/1000 GAL)
5 DAY 900 MG/L
RATIO KG/CU M
(L8/1000 GAL)
T3S MG/L
RATIO KG/CU M
(LB/1000 GAIL)
NO PLANT LOG MEAN
k 76.1
20.1
k 10.8
-------
DRAFT
Non-contact cooling water is also used in the vinegar generators and may
or may not be recycled. The ratio of wastewater to production averaged
1170 1/kkg (1170 gal/1000 gal) with a range of 540 to 2550 1/kkg (130
to 610 gal/ton).
The expected ranqe of BOD ratios is from 1.20 to 3.07 kg/cu m (10.0 to
25.6 lb/1000 gal) with an average of 1.92 kg/cu m (16.0 lb/1000 gal);
suspended solids is from 0.317 to 1.36 kg/cu m (2.63 to 11.3 lb/1000
gal) with an average of 0.654 kg/cu m (5.46 lb/1000 gal). The range of
waste loadings is not directly related to any observed differences be-
tween the processors. However, the handling of filter washwater and
storage tank sedimentation can greatly influence the waste loadings.
Of particular importance in the vinegar process is the presence of acetic
acid in the effluent. The arithmetic average pH for three plants with
raw effluent data was 5.17 with a range of 4.59 to 5.50. Surges of waste-
water with lower pH can be expected during the flushing of holding tanks
and cleanup of spillages.
Model Vinegar Plant
Production: 78 cu m/day (20,000 gal/day)
Wastewater flow volume: 90.8 cu m/day (.024 M6D)
Wastewater characteristics: BOD = 1950 mg/1
SS = 660 mg/1
pH = 5.2
Primary source of wastewater: filtration operation, washdowns.
Special consideration: pH adjustment.
SUBCATEGORIES El (MOLASSES. HONEY, GLAZED FRUIT, AND SYRUPS), E 2
(POPCORN). E 3 (PREPARED GELATIN DESSERTS). E 4 (SPICES). E 5 (DE-
HYDRATED SOUP) . AND E 6 (MACARONI. SPAGHETTI, VERMICELLI, AND NOODLES)
The processes associated with Subcategories E 1 through E 6 have been
found to generate little wastewater. What little wastewater that is
generated results from equipment cleanup and floor washing. The volume
generally amounts to less than 4000 I/day (1000 gal/day). The pollu-
tant loading is comparable to that of domestic sewage. The develop-
ment of model plants is not necessary for these subcategories.
SUBCATEGORIES F 2 BAKING POWDER). F 3 (CHICORY)/AND F 4 (BREAD CRUMBS
NOT PRODUCED IN BAKERIESj
As described in Section III, the processes associated with these sub-
categories are dry processes that generate no contact process waste-
water. Therefore, development of model plants is not necessary for
these subcategories.
492
-------
SECTION VI
SELECTION OF POLLUTANT PARAMETERS
WASTEWATER PARAMETERS OF PQLLUTIONAL SIGNIFICANCE
Major wastewater parameters of pollutional significance for the
miscellaneous foods and beverages industry include BOD ( 5-day
20°C), COD, suspended solids, and oil and grease. Minor parameters
of significance include pH, nickel, alkalinity, total dissolved
solids, nutrients (forms of nitrogen and phosphorus), color, chlorides
and temperature. On the basis of all evidence reviewed, there does
not otherwise exist any purely hazardous or toxic pollutants (e.g.,
heavy metals, pesticides) in waste discharged from the miscellaneous
foods and beverages industry.
When land disposal of wastewater is practiced, contribution to ground
water pollution must be prevented. Under land disposal procedures,
all practices should be in general accord with the Environmental
Protection Agency's "Policy on Subsurface Emplacement of Fluids by
Well Injection" with accompanying "Recommended Data Requirements
for Environmental Evaluation of Subsurface Emplacement of Fluids
by Well Injection" ( 90 ).
Significant pollutional parameters for the protection of ground
water from land disposal include BOD, COD, pH, temperature, total
dissolved solids, and nutrients.
RATIONALE FOR SELECTION OF IDENTIFIED PARAMETERS
The rationale for selection of the significant parameters for the
miscellaneous foods and beverages industry is given below:
Organics
Biochemical oxygen demand (BOD) is a semi-quantitative measure
of the biologically degradable organic matter in a wastewater. .For
this reason, in wastewater treatment, it is commonly used as a measure
of treatment efficiency. It is a particularly applicable parameter
for the miscellaneous foods and beverages industry since the wastes
are highly biodegradable with very few exceptions.
The primary disadvantage of the BOD test is the time period required
for analysis (five days is normal) and the considerable amount of care
that must be taken to obtain valid results.
Under proper conditions, the chemical oxygen demand (COD) test can
be used as an alternative to the BOD test. The COD test is widely
used as a means of measuring the total amount of oxygen required for
493
-------
DRAFT
oxidation of organics to carbon dioxide and water by the action of a
strong oxidizing agent Under acid conditions. It differs from the BOD
test in that it is independent of biological assimilability. The major
disadvantage of the COD test is that it does not distinguish between
biologically active and inert organics. The major advantage is that it
can be conducted in a short period of time, or continuously in automatic
analyzers. In many instances, COD data can be correlated to BOD data
and the COD test can then be used as a substitute for the BOD test
where a reliable relationship can be demonstrated to exist. Considerable
difficulties occur with the COD test in the presence of chlorides.
The measurement of total organic carbon (TOC) offers a third alter-
native for an indication o organic concentrations. This test offers
the potentiality of a high degree of reliability and produces results
in a matter of minutes. However, at the present time the equipment
required for the test is relatively expensive, has not been used
extensively to date, and has had little experience in the miscellaneous
foods and beverages industry.
With a few exceptions, the wastewaters generated by the miscellaneous
foods and beverages industry contain relatively high levels of readily
biodegradable organics.
Suspended Solids
Suspended solids serve as a parameter for measuring the efficiency
of wastewater treatment facilities and for the design of such facilities.
Suspended solids concentration in water affect light penetration,
temperature, solubility products, and aquatic life. Upon settling,
solids may blanket organisms or their habitats, either killing the
organism or rendering the habitat unsuitable for occupation. Suspended
solids concentrations greater than 80 mg/1 in fresh water streams have
been reported (91 ) to be detrimental to fisheries.
Suspended solids are a major pollutant parameter for most of the
subcategories discussed in this document. It is relatively minor for
most of the confectionery operations as well as for a few other products
for which carbohydrates are of greater importance.
Oil and Grease
Floating oils may interfere with reaeration and photosynthesis
and prevent respiration of aquatic insects which obtain their oxygen
at the water surface. Free and emulsified oils may interfere with
fish respiration and destroy algae and other plankton. Deposited
oily substances on the bottom of a stream bed may destroy benthic
organics.
494
-------
Oil and grease is a major parameter for the vegetable oil processing
and refining industry, the bakery and confectionery industry, the pet
food industry, and for several of the miscellaneous products.
These oils and greases of animal and vegetable origin should not be
confused with petroleum wastes. The oils and greases generated by
the industries which are subject to this study are readily biodegradable
in both municipal and private treatment systems.
pH is an important criterion for in-process control, odor control,
and bacterial growth retardation. Highly acidic or caustic solutions
can be harmful to aquatic environments and can interfere with water
or wastewater treatment processes. The acceptable range for successful
performance of biological treatment and a healthy fresh water habitat
is between 6.0 and 9.0.
Several of the subcategories discussed in this document require
minor pH adjustment before discharge or biological treatment. It is
perhaps most significant for vinegar which produces an effluent
with high concentrations of acetic acid.
Nickel
Nickel as a pure metal does not consitute a serious threat to
receiving waters; however, many of the salts of nickel are soluble
in water and may be hazardous to aquatic life. Since the acute and
chronic toxicity values of nickel vary widely, the EPA ( 92 ) has
proposed a limiting application factor of 0.02 of the 96 hour IC™
as required to provide adequate protection for aquatic life.
The only known source of nickel in process waste water from the
miscellaneous foods and beverages industry would be attributable to
the edible oils refining industry where small amounts of nickel are
used in the process. The discharge of nickel from edible oil refining
plants has been found to be very insignificant under present operating
practices. Effluent limitation of nickel within technological capa-
bilities and pollution control requirements is justified in a pre-
cautionary sense, due to the potential polluting effects attributable
to this material.
Alkalinity
Alkalinity in water is a measure of hydroxide, carbonate, and bi-
carbonate ions. Its primary significance in water chemistry is its
indication of a water's capacity to neutralize acidic solutions.
In high concentrations, alkalinity can cause problems in water treat-
ment facilities. However, by control of pH, alkalinity is also controlled,
495
-------
UKMr I
Total Dissolved Solids
The quantity of total dissolved solids in wastewater is of little
meaning unless the nature of the solids are defined. In fresh water
supplies, dissolved solids are usually inorganic salts with small
amounts of dissolved organics, and total concentrations may often
be several thousand milligrams per liter.
It is not considered necessary to recommend limits for total
dissolved solids since harmful salts and organics are limited by other
parameters.
Nutrients
Forms of nitrogen and phosphorus act as nutrients for the growth
of aquatic organisms and can lead to advanced eutrophication in surface
water bodies. In water supplies, nitrate nitrogen in excessive con-
centrations can cause methemoglobinemia in human infants and for this
reason has been limited by the United States Public Health Service to
ten milligrams per liter as nitrogen in public water supplies ( 93 ).
Under aerobic conditions ammonia nitrogen is oxidized to nitrite
and ultimately to nitrate nitrogen. Phosphorus compounds are commonly
used to prevent scaling in boilers and orthosphosphate may occur in
boiler blowdowns. The use of phosphate detergents for general cleaning
can contribute phosphates to total wastewater discharges. When applied
to soil, phosphorus normally is fixed by minerals in the soil, and
movement to ground water is precluded.
Color
True water color is a result of substances in solution after
suspended materials have been removed. It may be derived from mineral
or organic sources and may be the result of natural processes as well
as manufacturing processes.
The effect of extreme water color on aquatic life is to limit
light penetration, thereby restricting the photosynthetic zone and
impacting benthos. Otherwise, color may serve as an indirect'indica-
tion of pollution and be aesthetically objectionable.
The production of soluble coffee, tea, rum, and yeast results in a
wastewater with considerable color. The effectiveness of biological
treatment for color removal is questionable. Carbon filters or other
devices may be necessary for color removal in some instances, but present
technology for color removal from these wastewaters is nonexistent.
The acceptable limits of color in navigable waters are highly
dependent on the natural levels of color in the waters and the degree
496
-------
of available dilution. The Environmental Protection Agency ( 92 )
has proposed that acceptable conditions regarding the combined effect
of color and turbidity in water will be met if the water's compen-
sation point is not changed by more than 10 percent from its seasonably
established norm, and if no more than 10 percent of the biomass
of photosynthetic organisms is placed below the compensation point
by such changes.
Chlorides
Chlorides can cause detectable taste in drinking water in salt (sodium,
calcium, magnesium) concentrations greater than about 150 mg/1;
however, the concentrations are not toxic; drinking water standards
are generally based on palatability rather than health requirements.
In the application of wastewater to land, no practical limits can be
recommended by this document since chlorides are generally non-toxic
to crops, although some fruit trees are sensitive to chlorides. A
consideration of crop irrigation with wastewater should take into
account -chloride concentrations.
The operations discussed in this document which discharge significant
chloride concentrations are block ice production, olive oil production,
and pectin production. In the case of block ice production, the
concentrations in the wastewater are within drinking water standards.
The concentrations for olive oil and pectin are considerably higher
and attention must be given to specific discharges.
Temperature
The discharge of heated waters, with inadequate dilution, may result
in serious consequences to aquatic environments. Generally, problems
of heated water are associated with various cooling waters that are
not subject to recommendations in this document. One process stream,
currently discharged in some cases from rum distilling, approaches
the boiling point of water; however, recommended control technology
developed in Section VII would eliminate this problem.
METHODS OF ANALYSIS
During the course of this study a number of wastewater samples were
collected and analyzed at the laboratories of Environmental Science
and Engineering, Inc., Gainesville, Florida. The following outlines
the analytical methods used.
Solids
Total solids was determined by drying an aliquot of sample at
104°C according to EPA methods (EPA, Methods for Chemical Analysis
of Water and Wastes, 1974, p. 270; Standard Methods, pp. 535-536).
497
-------
DKAI-T
Dissolved and suspended solids were determined by glass fiber
filtration and drying at 104°C, (Standard Methods, pp. 535-536).
Volatile solids was determined by combustion at 550°C, (EPA Methods,
1974, p. 272; Standard Methods, p. 536).
pH and Temperature
pH and temperature were determined at the time of sample collection.
Nitrogen and Phosphorus
Total nitrogen was determined by the Kjeldahl digestion procedure
(Standard Methods, p. 469) and total phosphorus by the ascorbic acid
method (Standard Methods, p. 526, 532).
Oil and Grease
Oil and grease was determined gravimetrically by the liquid-
liquid extraction technique with hexane. The procedure is a modif-
ication of the technique described in EPA Methods, pp. 226-228.
BOD was determined by oxygen depletion at 20°C using a membrane
electrode to measure DO (Standard Methods, pp. 489-495; EPA Methods,
1974, pp. 11-12).
COD
COD was determined by dichromate oxidation followed by titration
with ferrous ammonium sulfate (Standard Methods, pp. 495-499; EPA
Methods. 1974, p. 20).
Color
Color was determined colorimetrically on a Klett-Summerson colorimeter
and is reported in chloroplatinate units, a variation of the method given
in EPA Methods, 1974, pp. 36-38 and Standard Methods, pp. 160-162. While
this method is designed for natural waters, the major need for color
analyses has been in the tea and coffee industries where the nature
of the color of the wastewaters approximates that of natural waters.
MH
Ammonia was determined by a selective ion electrode (EPA Methods, 1974,
pp. 165-167).
498
------- |