EPA-600/2-77-184
August 1977
Environmental Protection Technology Series
PROCEEDINGS EIGHTH NATIONAL SYMPOSIUM Ol
FOOD PROCESSING WASTES
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 4526!
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-77-184
August 1977
PROCEEDINGS EIGHTH NATIONAL SYMPOSIUM
ON FOOD PROCESSING WASTES
March 30 - April 1, 1977
Seattle, Washington
by
Food and Wood Products Branch
Industrial Environmental Research Laboratory
Corvallis, Oregon 97330
Co-sponsored by
National Canners Association
American Meat Institute
Southeastern Poultry and Egg Association
Pacific Egg and Poultry Association
Western States Meat Packers Association
Northwest Food Processors Association
National Independent Meat Packers Association
American Frozen Food Institute
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research
Laboratory-Cincinnati, U.S. Environmental Protection Agency, and approved
for publication. Mention of trade names or commercial products does not con-
stitute endorsement or recommendation for use.
11
-------
FOREWORD
When energy and material resources are extracted, processed, converted,
and used, the related pollution impacts on our environment and even on our
health often require that new and increasingly more efficient pollution con-
trol methods be used. The Industrial Environmental Research Laboratory-
Cincinnati (lERL-Ci) assists in developing and demonstrating new and improved
methodologies that will meet these needs both efficiently and economically.
The Eighth National Symposium on Food Processing Wastes was co-sponsored
with the National Canners Association, American Meat Institute, Southeastern
Poultry and Egg Association, Pacific Egg and Poultry Association, Western
States Meat Packers Association, Northwest Food Processors Association,
National Independent Meat Packers Association and American Frozen Food Insti-
tute. The primary purpose of these symposia is the dissemination of the la-
test research, development and demonstration information on process modifica-
tions waste treatment, by-product recovery and water reuse to industry,
consultants and government personnel. Twenty-nine papers are included in
this Proceedings as well as the final registration list.
These symposia will be continued; if you are interested in participating
or wish to receive additional information about the Ninth, contact:
Industrial Pollution Control Division
Industrial Environmental Research Laboratory-Ci
Environmental Protection Agency
Cincinnati, Ohio 45268
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
111
-------
CONTENTS
Foreword ill
THE SHAPE OF THINGS TO COME 1
W. A. Mercer
STATUS OF ERA'S EFFLUENT GUIDELINES FOR THE FOOD INDUSTRY 9
J. Denit, E. H. Forsht
EFFLUENT POLISHING AND WASTEWATER REUSE AT SNOKIST GROWERS CANNERY 20
L. A. Esvelt, H. H. Hart
CONTROL OF ODORS FROM ANAEROBIC LAGOONS TREATING MEATPACKING WASTES 38
J. A. Chittenden, L. E. Orsi, J. L. Witherow
TOMATO CLEANING, WATER RECYCLE AND MUD DEWATERING 62
W. W. Rose
REMOVAL OF SUSPENDED SOLIDS AND ALGAE FROM AERATION LAGOON WASTE 76
WATERS TO MEET 1983 DISCHARGE STANDARDS TO STREAMS
E. R. Ramirez, D. L. Johnson, T. E. Elliott
EFFLUENT GENERATION, ENERGY USE AND COST OF BLANCHING 85
J. L. Bomben
DISSOLVED AIR FLOTATION TREATMENT OF SEAFOOD PROCESSING WASTES— 98
AN ASSESSMENT
D. B. Ertz, J. S. Atwell, E. H. Forsht
COMMERCIAL FEASIBILITY OF RECOVERING TOMATO PEELING RESIDUALS 119
W. G. Schultz, H. J. Neumann, J. E. Schade,
J. P. Morgan, A. M. Katsuyama, H. J. Maagdenberg
WASTE REDUCTION BY PROCESS MODIFICATION IN SWEET CORN PRESERVATION 137
H. Robertson, M. E. Lazar, J. M. Krochta, D. G. Farkas
APPLICATION OF FINE SCREENS IN THE TREATMENT OF FOOD PROCESSING 147
WASTEWATERS
R. C. Neal, A. D. Bubp, R. L. Chaney
PRELIMINARY EVALUATION OF ANAEROBIC SLUDGE DIGESTION FOR THE TUNA 155
PROCESSING INDUSTRY
A. Kissam, H. Barnett, F. Stone, P. Hunter
-------
REUSE OF BRINES IN COMMERCIAL CUCUMBER FERMENTATION 169
R. F. McFeeters, M. Palm'tkar, M. Velting,
N. Fehring, W. Coon
TREATMENT OF PACKINGHOUSE WASTEWATER BY SAND FILTRATION 186
M. L. Rowe
ECONOMIC RETURN ON POLLUTION CONTROL EXPENDITURES FOR THE 197
PICKLED FOOD INDUSTRY
J. G. Meenahan
AN EFFECTIVE WASTEWATER MANAGEMENT PROGRAM FOR A CARROT PROCESSOR 200
G. E. Wilson, J. H. C. Huang
RECOVERY OF SOLUBLE SERUM PROTEINS FROM MEAT INDUSTRY WASTES 211
R. W. Greiling
IMPROVED BIOLOGICAL TREATMENT OF FOOD PROCESSING WASTE WITH 235
TWO-STAGE ABF PROCESS
B. W. Hemphill, R. G. Dunnahoe
SCP FROM FOOD WASTES BY THE DEEP TANK PROCESS * 253
M. L. Jackson, C. C. Shen
EVALUATION OF INSTANT NOODLES PROCESSING WASTE WATER 266
CHARACTERISTICS AND TREATMENT ALTERNATIVES
P. Y. Yang, V. S. Luis, Jr.
POTATO JUICE PROCESSING 284
J. R. Rosenau, L. F. Whitney, J. R. Haight
RECOVERY AND APPLICATIONS OF ORGANIC WASTES FROM THE LOUISIANA 292
SHRIMP CANNING INDUSTRY
S. P. Meyers, B. E. Perkins
TOXIC CHARACTERISTICS OF SOME CANADIAN FRUIT AND VEGETABLE 308
PROCESSING WASTEWATERS
R. N. Dawson, A. Lamb, P. A. Mulyk
REDUCING WASTEWATER FROM CUCUMBER PICKLING PROCESS BY 322
CONTROLLED CULTURE FERMENTATION
L. W. Little, S. J. Dunn, R. Harrison, J. D. Harris
SALMON PROCESSING WASTEWATER TREATMENT 333
P. A. Bissonnette, S. S. Lin, P. B. Liao
FUNGAL CONVERSION OF CARBOHYDRATE WASTES TO ANIMAL FEED PROTEIN— 355
VITAMIN SUPPLEMENTS
B. D. Church, C. M. Widmer
WATER REUSE OF WASTEWATER FROM A POULTRY PROCESSING PLANT 389
J. B. Andelman, J. D. Clise
vi
-------
WATER RECYCLING IN POULTRY PROCESSING: CASE STUDY IN EGYPT 411
A. A. Hamza, S. Saad
THE TREATMENT AND DISPOSAL OF WASTEWATER FROM DIARY PROCESSING 427
PLANTS
J. A. Moore, B. M. Buxton
vii
-------
FROM WHERE I STAND
by
Walter A. Mercer*
INTRODUCTION
The National Canners Association, its officers and members, wish me to express
its appreciation and gratitude for the opportunity, over these eight years,
to be recognized as a co-sponsor of this National Symposium on Food Processing
Wastes. As always, in the past and now, representatives of the Association's
Research Staff are pleased to be present and to participate in the program.
The intensive planning which has gone into each of the past seven symposia
has resulted in a comprehensive coverage of environmental research dealing
with the many facets of water use and wastewater generation in the production
and preservation of foods of all kinds.
I am greatly impressed with the number and importance of the co-sponsors
listed in this Eighth Symposium program. The fate of the future food supplies
of this Nation will be greatly affected by the current and future welfare of
these segments of the food industry. Their ability to survive and their
willingness to gamble with unpredictable situations, depends on the degree
of economic strangulation brought about by emotionally-conceived and hastily-
enforced environmental controls.
As I look backward, from this day, near the end of a career of research efforts
in a broad area of food industry problems, I recall my participation in the
birth and growth of these symposia.
This Eighth Symposium marks a further sincere cooperative, coordinated effort
between industry and government to solve a complex of troublesome environ-
mental problems for a number of segments of the food producing and processing
industry. The listing of co-sponsors of this Symposium are evidence of that
fact.
My experiences in government-industry relations reach back to the days when
the U.S. Public Health Service was responsible for quality and pollution
abatement research. Through the National Institutes of Health and the Bureau
of State Services, I received the first public health research grant ever
awarded to a member of an industry-supported research group.
That was 20 years ago. The principal aims of the 8-year period of research
support were to make possible and promote water conservation practices in food
processing, while protecting and improving, where necessary, the quality and
wholesomeness, and sanitation of the canned and frozen foods produced.
*National Canners Association, Berkeley, California.
-------
This, and other research, established the basic principles of the counter-flow
water reuse systems now in use. Included was demonstration of the effective-
ness of the controlled application of chlorine to maintain sanitary conditions
and the mechanics of more effective application of water in washing and trans-
porting raw foods in preparation for canning or freezing.
During those years, the asparagus, bean, and pea plants of the Pacific North-
west became homes to me—particularly those in the Blue Mountain area of
Eastern Washington and Oregon. Later each year the action shifted back to
California and its fruit and tomato processing plants.
Now we are faced with a water-shortage crisis and the forced realization that
water is a precious commodity whose every drip must be used and re-used now
and for all of the years to come.
During the last few weeks old research reports, water reuse diagrams, and
conservation recommendations have been dusted off—to be searched through
to find if yesterday's answers fit today's problems—or offer a new solution.
Water reuse procedures and waste reduction procedures, developed by yesterday's
research, are in use today and have certainly progressed to a degree of sophi-
stication which has assisted the industry in its pollution abatement problems.
If the canning and freezing industry were required, or desired, to use water
on a one-pass basis, the annual pack of foods would require an estimated
280 billion gallons of fresh water. Because the industry, many years ago,
recognized the need for, and the pollution abatement benefits, of conserving
water, it now reuses about 180 billion of its intake water. The final waste-
water discharge is approximately 100 billion gallons.
Today, then, there is an estimated 64 percent reuse of processing and container
cooling waters in the canning and freezing of foods. This amazing effort to
conserve water and reduce liquid waste flows was made possible by government-
industry cooperation in years of research which developed and demonstrated
the conservation procedures and the chlorination controls necessary for protec-
tion of the quality and cleanliness of the foods and the sanitary condition
of the food handling equipment.
A question to be answered would cite the current estimate of 64 percent reuse
of waters in food canning and freezing and then demand to know why 100 percefit
reuse and zero discharge of wastewaters cannot now be an industry practice.
First, let me say, that the food processing industry urgently needs and anxiously
desires to cooperate in additional industry-government financed research.
This could be the means of enabling the industry to further diminish its
dependence on water and to lessen its contribution to environmental problems.
Such cooperative research proposals, using as a basis the outstanding results
of past cooperative projects, will be proposed. EPA inability or refusal
to provide financial and advisory support for sound research proposals is
inconceivable. At the focal point of the surrounding circumstances is the wel-
fare of the public, to the degree that it depends on food growing, food preser-
vation, and food distribution.
-------
To accomplish all or a part of the needed research requires further develop-
ment of in-plant water-saving and waste-prevention equipment or procedural
changes—and then of major importance a demonstration, for any and all sig-
nificant changes, of economical, technological, and sanitational acceptance
for application in the processing of foods for human and animal consumption.
Extensive re-use of the same water, which approaches closed-loop technology,
must be observed, studied, and approved for use by government agencies who
regulate the wholesomeness and health safety of foods intended for human and
animal consumption.
At this time, recycling of water in contact with food being prepared for pre-
servation cannot come near the closed-loop procedures used in non-food in-
dustries. Therefore, if the Federal Water Pollution Control Act adheres to
its intent to eliminate all discharge of pollutants by 1985, those food pro-
cessors able to survive will be few and necessarily large in terms of finan-
cial capabilities.
Elimination of food processing plants, however small and however significant
to the food production capability of this Nation, has immeasurable and complex
adverse social and political consequences. One must recognize that mandated
achievement of zero-discharge operations will certainly have the potential
for limiting the Nation's food supply and, most importantly, denying an ade-
quate variety of foods to a significant segment of society because the costs
are prohibitive.
For my industry, the only hope of approaching, on a broad front, the bureau-
cratic dream of zero discharge of pollutants would require that tomorrow we
begin that complex, sophisticated research which must be done to establish
the public health safety and aesthetic acceptability of closed-loop recycling
of food processing waters.
Dr. Robert Schaffner, Associate Director for FDA's Bureau of Foods, has made
it quite clear that EPA may have assumed, in their Effluent Limitation
Guidelines, that the food processing industry will practice extensive recycling
of water and the reclamation and reuse of treated wastewaters. However, he
clearly stated that the issuing of future regulations for the safe use of
recycled waters on foods is FDA's responsibility.
Dr. Schaffner also noted that more research and monitoring is needed to deter-
mine the magnitude of the problem of chlorination of recycled waters.
Dr. Schaffner stated that:
"Chloroform and other chloro- or bromo-alkanes seem to be present most oftenm
and a wide variety of other compounds may be found whose origin or formation
is not understood at this time."
This means that research will have to be undertaken to determine whether or
not an extensive water recycle and rechlorination operation in food processing
may lead to the formation of such organic compounds.
-------
Such research should be EPA-sponsored, in a typical food industry operation,
and I do not refer, here, to treated wastewater for reuse as food processing
water.
A PLEA FOR SURVIVAL OF EPA'S FOOD WASTE RESEARCH PROGRAM AT CORVALLIS, OREGON
It is alarming to me, personally, to my associates in food research and pro-
duction, to associations of food processors, and to consumers, were they
adequately informed, to learn that EPA plans to eliminate, for the next fiscal
year, funding for further food-oriented research development and demonstration
projects. Of great and equal concern are the consequences of a proposed
plan of reorganization which eliminates EPA field offices and, thereby,
disbands the teams of EPA scientists to whom we have looked for technical
advice.
Surely, we have interpreted correctly that portion of the report of the Na-
tional Commission on Water Quality which gives recognition to the value of
EPA scientific input into solving the environmental problems of consumer-
dependent, low-profit margin industries in regional areas.
I quote from the Commission's report:
"Finally, the results of many (of the) studies point to a need for technical
advisory services to localities and to small industries. The compliance
requirements are technically, legally, and administratively complex. The
planning requirements and their long-term implications reach beyond the capa-
city of many small or even moderate-sized communities to cope."
I continue to quote:
"The Commission is convinced that such a field advisory service, operating
out of regional offices of EPA would help localities, states, and the Federal
government save money. It would halp to expedite grant applications. It
would help to assure that communities select the best long-range treatment
option, in terms of cost effectiveness and environmental quality. Properly
staffed, it could also be an essential information clearing-house to keep
localities and industries abreast of innovative technologies and economically
and environmentally preferable solutions to their individual treatment pro-
blems."
The Commission recommendation and others in the report, are completely in
line with my industry's experiences with certain EPA scientists in Washington
and most certainly that Food and Wood Products Group in Corvallis, Oregon.
I am certain that such statements could be made for other similarly-oriented
scientists in other field branch offices of the Agency's current organization.
The Food and Wood Products Branch in Corvallis has, over the past several
years, supervised numerous research projects designed to aid the food process-
ing industry's waste management and pollution abatement programs. Since
wastes generated during the processing and production of foods are unique—
exceptionally high in pollutional strength, highly variable in nature, and
frequently seasonal—common technology designed for domestic sewage or other
industrial wastes cannot be transferred for application to the treatment of
food processing wastes.
4
-------
Vitally needed information has resulted from the programs supported by the
Food and Wood Products Branch. This information has been the basis for the
following: (1) process modifications to curtail waste generation, (2) water
conservation and wastewater reclamation procedures to reduce water consumption,
(3) treatment alternatives to more economically meet pollution abatement regula-
tions, and (4) by-product utilization of waste materials, especially to further
the utility of food materials. Although progress has been made in each of
these research areas, much work remains to be done to optimize food processing
operations to meet environmental goals. This work can only be effectively
continued through Federal assistance as administered for the Western States
by the EPA scientists in Corvallis.
Due to restrictive compliance schedules legislatively imposed by the Water
Pollution Control Act of 1972, the food industry must implement waste abate-
ment procedures which will result in the discharge of specified minimal quan-
tities of water pollutants. The economic impact of these pollution controls
is a serious deterent to food production and reasonable consumer costs. Only
with continued and intensified research programs can the maximum benefit-
cost ratio be achieved for capital expenditures related to pollution abatement.
Especially under present economic conditions, consumers and food processors
alike can ill-afford imposition of added costs which will result from dis-
continuities in the current Federal research programs. Every effort must be
made to assure that the vitally needed programs are uninterrupted. Continuity
in the present staff of the Food and Wood Products Branch is prerequisite
to maintaining these programs.
Advantages of the Corvallis Laboratory
In addition to preserving continuity of staff and current programs, continuing
the current staff intact at the Corvallis laboratory offers several advantages.
Among the principal advantages is its proximity to areas in which certain
segments of the food processing industry are concentrated. These centers are
evidenced by the large numbers of fruit and vegetable processing plants in
the Northwest and California, seafood processing in Alaska and along the Pacific
Coast, beet sugar refining plants in the Rocky Mountains, cane sugar plants in
Hawaii, and to lesser degrees, the number of meat, poultry, and dairy product
plants throughout the western states.
Close proximity to such a wide variety of food processing centers has promoted
excellent communication and cooperation between industry and the Food and
Wood Products Branch. Environmental problem areas have thus been more readily
and better defined, thereby encouraging meaningful research programs. Broader
selection of test sites for specific projects has assured the widest possible
application of research results—technical and technological developments have
had applicability well beyond geographic and commodity considerations. The
western states, in which Corvallis is centrally situated, is unique in this
important respect.
The Federal food wastes research program as administered by the Branch in
Corvallis, by necessity relies heavily on extramural resources, primarily
through universities, private research organizations, other governmental
agencies, and trade associations. A large number of such institutions, where
food wastes research programs are actively conducted, is located in the West.
-------
The universities in Washington, Idaho, Oregon, and California are highly ori-
ented toward agricultural production and food processing; the Sea Grant Colleges
in Washington and California, as well as the National Marine Fisheries Services
laboratories in Washington and Alaska, are organized to work on seafood pro-
cessing problems; the U.S. Department of Agriculture's Western Regional
Research Center and the National Canners Association Western Research Laboratory
have worked in close cooperation with the EPA in developing technical informa-
tion and new technology for pollution abatement in fruit, vegetable, and seafood
processing. The proximity of the Corvallis Branch to these institutions has
largely enhanced the implementation of valuable research programs at these
locations.
Constant communication between EPA and industry, and between EPA and extra-
mural research institutions, is essential to the continuation of fruitful
research. Proximity not only encourages maintenance of such communication,
but also promotes frequent personal contact between the Corvallis staff and
these outside parties, thereby offering the staff the opportunity to guide
and monitor research projects at minimal costs to EPA and to provide continuous
development of the staff's expertise in industrial matters. These opportunites
will enhance EPA's ability to assure the promulgation of sound and realistic
environmental regulations, thereby acquiring maximum benefits for the well-
being of both industry and the general public.
The Basis for My Plea
For the food canning industry, we have said in the past and wish to repeat
the following:
"The lofty intent of the American public to protect our environment was imper-
fectly translated into laws by the Congress. Those laws are being imper-
fectly translated into regulations and the regulations themselves are being
imperfectly enforced. In too many cases these successive distractions have
resulted in actions which are unrelated to the basic intent of environmental
protection.
You who have scientific and technical competence are the ones to whom we must
look to keep all of the machinery properly directed. While you may not control
the legal machinery itself, you are the advisors upon whom all must lean."
During this past weekend I was shocked and dismayed to read some of the
strongly worded statements coming from the headquarters office of EPA, and
especially was I disturbed by the Secretary of Interior, who in his Saturday
speech to the National Wildlife Federation coined the phrase:
"Rape, ruin, and run."
The Secretary, of course, was referring to alledged misuses of public lands
for strip mining, oil development, live stock grazing, and for other purposes.
But if EPA abdicates from its responsibility to continue effective funding
for meaningful research in food processing environmental problems, and if the
-------
present Food and Wood Products Branch is disbanded and removed from the Cor-
vallis Laboratories, I shall repeat loudly and often that apt phrase:
"Rape, ruin, and run."
EPA Has Established That the Food Processing Industry is Different
The fruit, vegetable, and seafood processing industry in the United States
is characterized by extremely wide diversities. This, in fact, applies to
all agriculturally-based and agriculturally-dependent industries. The diver-
sities for food canning and freezing plants may be categorized as follows:
processing plants vary more than 3000-fold in size, as measured by the
tons of raw commodity annually processed
the plants are located in wide-spread geographic areas, from the sub-
Arctic regions of Alaska to the temperate climate of Florida and Hawaii.
Plants sites also vary between the extremes of isolated shoreline loca-
tions in Alaska to highly industrialized metropolitan centers
processing plants operate for varied periods, depending upon the avail-
ability of specific raw commodities; salmon plants in Alaska may operate
for just a few weeks, while citrus, potato, and tuna plants may operate
for ten or more months per year. Generally, the industry is highly
seasonal, with most plants in operation for approximately four months
out of the year
product mixes (e.g., peas and corn, peaches and pears, salmon and crab,
etc.) vary widely between plants, especially in the ratio of commodities
processed at each. Thus, waste characteristics, and hence environmental
problems, vary among plants within each category (fruit, vegetable,
seafood).
These diverse factors, to name but a few, compound the environmental problems
facing the food processing industry. Research must therefore be directed not
only to general problem areas, but to the specific problems unique to each
industrial category. Above all, research programs must include thorough
economic analyses, taking the above factors into account, so that cost-effec-
tive alternatives may be developed for solving the various environmental pro-
blems.
The food processing industry is grateful for the assistance that came in
the past—critical problems have been solved. Today's problems are even more
critical. Tomorrow will not come for the food industry as we know and depend
on it today—unless we strive together—industry and government—to find a
way through the economical and techno!gical maze that lies ahead.
I shall be deeply troubled if the future brings circumstances which can only
be fittingly described by the phrase:
"Rape, ruin, and run."
-------
From Where I Stand—I wish my view of the food industry's future, reflected
from the experiences of the past, could be brighter. I know that food proces-
sing, with all of its variations, non-standardized systems, and plant-to-
plant differences, cannot be stuffed into the EPA-fashioned inflexible mold.
There must be a better way. We—EPA and the industry—must find that way
together.
8
-------
STATUS OF EPA'S EFFLUENT
GUIDELINES FOR THE FOOD INDUSTRY
by
J.D. Denit* and E.H. Forsht*
INTRODUCTION
Even after five years of working with the statutes, the Federal Water Pollu-
tion Control Act Amendments(1) or the FWPCA is still considered by many
observers to be one of the most complex and comprehensive measures enacted
by Congress. The basic goal of the Act is to "restore and maintain the
chemical, physical and biological integrity of the Nation's waters." Also
by July 1983, wherever possible, water quality is to be suitable for
recreational contact and for the protection and propagation of fish and
wildlife.
The basic mechanism of achieving these goals includes an effluent control
system to limit discharges of pollutants from point sources into any body of
water. Effluent limitations and standards are established so that
industrial sources of pollution will have to meet increasingly stringent
limitations by the 1977 and 1983 deadlines set in the Act.
By this July, all direct dischargers of pollutants must meet effluent
limitations based upon the best practicable control technology currently
available (BPT).** This technology represents the average of the best
existing waste treatment performance within each industry category or
subcategory. However, the Legislative History provides that where waste
treatment performance is determined to be uniformly inadequate, the
Administrator of EPA may establish effluent limitations on the basis of what
the Agency finds is appropriate for BPT in the industry. In such instances,
the industry may be held responsible for achieving higher levels of control
than any currently in place.
It should be emphasized that determination by EPA that a treatment system
(or systems) constitutes BPT does not necessarily require the purchase and
installation of that particular equipment. The technology which forms the
basis of the effluent limitations are used as reference points only to
*Branch Chief, Food Industries Branch and Chemical Engineer, respectively,
U.S. Environmental Protection Agency, Effluent Guidelines Division (WH-552),
Washington, D.C. 20460.
**It should be noted that discharge permits for individual dischargers
(normally valid for five years) may have been issued at such a time that
they will expire during calendar year 1977. The NPDES permit issuing
authority is responsible for such action as necessary to "bridge" the six
year period before 1983 limitations are required.
-------
document the availability of technology which can meet the necessary levels
of performance. However, the industry may select alternative methods to
meet the effluent limitations. In some plants, for example, implementing
waste and water management practices, utilizing by-product recovery
techniques, or making changes in process operations may be sufficient to
meet the effluent limitations, without the need for end-of-pipe waste
treatment. On the other hand, the mere presence of technology which may be
defined as BPT does not guarantee that the effluent limitations will be met.
It is the industry's responsibility to assure that the appropriate system is
installed in a manner designed and sized to accommodate the waste
characteristics and flow for individual plants. The treatment units must
also be properly operated and monitored to maintain optimum operating
conditions.
By July. 1, 1983 all industrial point sources must meet effluent limitations
based on the best available technology economically achievable (BAT). This
technology represents the very best control and treatment measures
(including in-plant and process changes) that have been developed or are
capable of being developed within each industry category or subcategory.
New industrial point sources must meet effluent standards based on the best
available demonstrated control technology. These standards include an
assessment of what higher levels of pollution control are available through
the use of improved production processes as well as end-of-pipe treatment
techniques.
Again it should be noted that selection of a particular treatment system as
best available or best demonstrated technology, which is used as the basis
for effluent limitations or standards, is not tantamount to a technology
requirement. Other treatment alternatives can be used to meet the
regulations.
CURRENT STATUS OF FOOD INDUSTRY REGULATIONS AND STUDIES
As shown in Table I, the Agency's promulgated regulations cover
approximately 28,000 point sources within the food processing industry.
These facilities discharge nearly 70 million pounds of BODS per day to the
receiving waters or to publicly owned treatment works.
The legal status of these regulations are outlined in Table II. Regulations
covering 88 subcategories of the food industry have been promulgated. Of
these, 32 subcategories are now under judicial review, including 17 subcate-
gories within the grain mills, meat processing and leather tanning
categories which the courts have remanded to the Agency for reconsideration
of various technical and economic issues.
Table III outlines the status of the remaining food industry studies. The
miscellaneous foods category has been divided into the beverages, edible
oils, bakeries and confectioneries, and miscellaneous specialty food
segments. It is the Agency's intention that regulations for beverages and
10
-------
TABLE 1. SELECTED CHARACTERISTICS OF THE FOOD INDUSTRY
Category
Dairy (2)
Grain Mills (3)
Fruits & Veg. (4,5)
Seafood (6,7)
Sugar (8)
Feedlots (9)
Meat Prod. (10,11,12)
Leather (13)
No. of
Dischargers
5,400
210
1,900
1,400
130
10,000
8,700
180
27,920
Flow
MGD
200
150
470
180
3,150
-
5,550
50
9,750
BOD5 Plants meeting
M Ibs./day BPT
4,200
610
5,480
710a
2,210
-
55,400b
710
69,320
201
40%
75%
35%
45%
-
65%
10%
a after screening
b after DAF
-------
TABLE 2. LEGAL STATUS OF THE FOOD INDUSTRY REGULATIONS
ro
Category
Dairy
Grain Mills
Fruits & Veg.
Seafood
Sugar
Feedl ots
Meat
Leather
No. of
Subcategories
Proposed Promulgated
11
10
8
33
8
2
5 10
6
5 88
Current
Litigation
0
1
3
10
2
0
10
6
32
Wi thdrawn Remanded
1
5a
10
6
5 17
a 1983 only
-------
TABLE 3. STATUS OF OTHER FOOD INDUSTRY STUDIES
Miscellaneous Food
Beverages
Edible Oils
Bakeries & Confectioneries
Misc. Specialities
Meat Products
Poultry
Fish Hatcheries
Proposed in near future
May be issued as guidance
Anticipate promulgating regulations,
but only for 1983 and New Source
May be issued as guidance
-------
edible oils will be proposed in the near future. However, current resource
constraints within the Agency have delayed the economic impact studies for
the bakeries and confectioneries, and the miscellaneous specialty food
segments. Regulations for these latter two segments and for fish hatcheries
and farms are not now anticipated. Instead, technical reports, called
Development Documents, will be issued as guidance to permit issuing
authorities pending completion of the requisite economic and cost impact
studies. The regulations for the poultry subcategories of the meat products
category were published in proposed form in the Federal Register on April
24, 1975, however, the Agency anticipates promulgating only the 1983
effluent limitations and new sources and pretreatment standards.
The salient legal issues that have been raised in the judicial review of the
cases cited above included jurisdiction (Federal District Court versus the
U.S. Court of Appeals) and ranges of limitations for individual plants
versus national uniform standards by industrial classes. A recent Supreme
Court decision(14) resolved several legal issues regarding the effluent
limitations and standards promulgated by EPA. The Supreme Court held that
(1) EPA has the authority under Section 301 of the FWPCA "to limit the
discharge of pollutants, through industry-wide regulations setting forth
uniform effluent limitations, provided some allowance is made for variations
in individual plants," and (2) the Courts of Appeals have exclusive
jurisdiction to review effluent limitations promulgated under Section 301
which also includes jurisdiction to review guidelines promulgated under
Section 304. It is anticipated that the above decision will greatly
expedite and eliminate confusion during judicial review procedures.
CHANGING INFLUENCES AND CONSTRAINTS
Changing influences and constraints which may affect the food processing
industry regulations include the energy shortage and other environmental
legislation.
This winter has highlighted the dismal prognostications regarding the
world's energy resources, i. e., our conventional supplies of energy cannot
support current levels of consumption. Perhaps these predictions prompted
Congress to include the statutory requirements in the Act to consider the
energy consumption of the waste treatment technology utilized as the basis
for the effluent limitations and standards.
It is not too surprising that a consideration of minimum energy consumption
actually encourages production efficiency, waste management and water
conservation. Improvements in product and by-product recovery generally
divert energy consumption from non-productive waste handling and treatment
to production of useful end products. Improved waste management and water
conservation practices leads to smaller energy requirements for waste
treatment. Smaller treatment systems require fewer resources for production
and installation. Likewise, pumping smaller volumes of effluents through or
among treatment units requires less energy.
14
-------
At the present time, the Agency is assessing various issues regarding a
strategy to strengthen overall environmental protection under effluent
guidelines through consideration of the impact of the Resource Conservation
and Recovery Act(15) and the Safe Drinking Water Act(16). The Resource
Conservation and Recovery Act provides for R & D and dissemination of
information on promising recovery, disposal, and resource use techniques.
The Agency must also issue criteria for identifying hazardous wastes and
require a permit for any facility which treats, stores, or transports
hazardous wastes. The Safe Drinking Water Act is designed to assure that
water supply systems serving the public meet minimum national standards for
protection of public health. This Act also establishes a joint Federal-
State system for protecting underground sources of drinking water.
One example of the inter-relationship among these Acts revolves around the
"zero discharge" effluent limitations established for various point source
dischargers including some within the food processing industry. In
assessing this requirement in relation to the Safe Drinking Water Act, two
points are particularly pertinent. First, ground water is the drinking water
source for at least 50 percent of the American population. The majority of
this ground water is not treated prior to use as drinking water. Second,
the Safe Drinking Water Act is designed to protect both surface and ground
sources of drinking water from initial contamination wherever possible.
Correspondingly, the Resource Conservation and Recovery Act mandates the
protection of both ground and surface waters from leachates from any solid
waste disposal project. Disposal is defined as "the discharge, deposit,
injection, dumping, spilling, leaking, or placing of any solid waste or
hazardous waste into or on any land or water so that such solid waste or
hazardous waste or any constituent thereof may enter the environment or be
emitted into the air or discharged into any waters, including ground water."
As a practical matter, many dischargers may attempt to meet the "no
discharge" limitations by disposing of untreated wastes into an unlined pond
or lagoon. If the surrounding area is permeated by an aquifer which could
be a drinking water source, potential leachates could create serious
environmental problems. In fact, leaching from holding ponds and lagoons is
the second major srurce of ground water pollution next to septic tanks and
cesspools.
In some cases, particularly those involving hazardous materials, a zero
discharge standard may raise consequent questions as a solution to the
problems posed by water pollution. In other cases "zero discharge" in arid
regions may also complicate the essential balance between beneficial uses
downstream and water conservation efforts which are strongly supported by
states, environmental groups, and farmers.
Even with limitations which allow a discharge of pollutants, one must
consider the implications for sludge disposal and effluent discharge to
ground or surface waters. In fact, the Agency's current efforts include
developing strategies to incorporate a close scrutiny of all intermedia
effects into subsequent reviews of the effluent limitations and standards.
15
-------
RECOVERY, REUSE, AND CONSERVATION
These changing influences and constraints which include other environmental
laws and declining resources, confirm the truth in the old maxim that
recovery, reuse, and conservation are essential to pollution control,
resource utilization and economic efficiency. The food processing industry
in general, as exemplified by many segments of the meat processing industry,
should reorient its efforts toward a "total utilization concept," wherein
much of the current waste materials are viewed as "secondary raw material."
This approach closes the processing cycle so that raw material is used to
the fullest extent possible with the subsequent minimization of
environmental pollution. The implementation of in-plant changes to
accomplish this goal is certainly more logical than spending large amounts
of money to simply treat food processing wastes at the end of the effluent
pipe.
Successful examples of these concepts include:
a) Recovery of animal and fish solid wastes as well as dissolved and
suspended nutrients for animal feeds, tallow, inedible products, or low
grade fertilizers;
b) Recovery of potato, orange, grapefruit, and pineapple peel wastes as
animal feeds, and
c) Recovery of whey solids for use in other food products.
Likewise successful water conservation techniques which may apply to the
food processing industry include:
a) Teaching good water management practices to plant operators;
b) Utilizing dry solids transportation rather than wet fluming;
c) Recycling or reusing water; and
d) Implementing dry clean-up operations prior to water washes.
FUTURE CONCERNS
As a practical matter, the future of the food processing regulations
contains the uncertainties associated with possible legislative amendments
to the FWPCA, and the new administration's approach to environmental issues.
Congress is now considering many proposals from industry, labor,
agriculture, environmental, and government groups for amendments to the
FWPCA. The Environmental Protection Agency has submitted several proposals
which would directly affect industry. These include:
16
-------
a) inserting a new section authorizing the imposition of a fee upon non-
complying point sources. The fee would approximate the economic value of
non-compliance to equalize the competitive positions of the cooperative and
recalcitrant point sources;
b) amending section 509(b) to modify the location for judicial review of a
number of EPA actions under the FWPCA so that petitions for review may be
filed only in the United States Court of Appeals for the District of
Columbia. Such a provision finds precedent in the Clean Air Act and would
prevent "forum shopping" which often results in inconsistent interpretations
of the Act and issues of nationwide importance; and
c) amending section 308 to provide that contract employees of EPA would be
permitted to excercise the functions of an "authorized representative" as
the Congress intended when it enacted section 308.
The Agency has advised against extension of the July 1, 1977 deadline for
industrial point sources, because most of the major industries will be in
compliance by that date. Those industries unable to meet the deadline
because of factors beyond their control will be eligible to receive EPA
Enforcement Compliance Schedule Letters that require compliance by a
specified date. Despite the recommendations of the National Commission on
Water Quality, the Agency sees no need to extend the 1983 deadline since
Section 301(c) provides for economic variances from the BAT requirements
where appropriate.
Based on campaign issues, papers and recent statements, it appears as if
President Carter will play an active role in environmental issues and
legislation. The President has stated that "we must vigorously enforce the
pollution control... laws already on the books."(17) He also believes that
much of the environmental damage which now occurs can be prevented and has
pledged his support for research to find environmentally sound ways to
achieve economic goals without unacceptable pollution damage. The President
also believes that it is not possible to discuss environmental pollution
without considering energy. Several elements of the President's energy
policy relate directly to the environment. As an integral part of energy
conservation, industry needs "to make better use of recycled materials, to
better manage our solid wastes, and to realize the fuel savings which
recycling offers".(17)
SUMMARY
These recent trends in environmental issues suggest that, regardless of the
outcome of individual court cases or the proposals for amendments to
environmental legislation, the industrial sector is expected and must
continue to reduce the levels of pollutants into the environment.
Industry should evaluate every waste treatment alternative which is
technically and economically feasible, not just the technology which forms
the basis for the effluent limitations and standards. The Agency intends to
17
-------
continue working to integrate its efforts to help minimize the intermedia
environmental consequences of its regulations. Correspondingly, the optimum
environmental approach which typically includes by-product recovery, reuse,
recycle, and conservation along with end-of-pipe treatment, is increasingly
consistent with industry's economic self-interest.
REFERENCES
1. Federal Water Pollution Control Act Amendments of 1972, P.L. 92-500,
October 18, 1972, 33 USC 1251, et. seq.
2. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Dairy Product Processing Point Source
Category, U.S. Environmental Protection Agency, Wash., D.C., 20460, May
1974, EPA-440/l-74-021-a.
3. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Grain Processing Segment of the Grain
Mills Point Source Category, U.S. Environmental Protection Agency,
Wash., D.C., 20460, March 1974, EPA-440/l-74-028-a.
4. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Apple, Citrus, and Potatoes Segment of the
Canned and Preserved Fruits and Vegetable Point Source Category, U.S.
Environmental Protection Agency, Wash., D.C., 20460, March 1974, EPA-
440/1-74-027-a.
5. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Fruits, Vegetables and Specialties Segment
of the Canned and Preserved Fruits and Vegetables Point Source Category,
U.S. Environmental Protection Agency, Wash., D.C., 20460, October 1975,
EPA-440/1-75/046.
6. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Catfish, Crab, Shrimp and Tuna Segment of
the Canned and Preserved Seafood Processing Point Source Category, U.S.
Environmental Protection Agency, Wash., D.C., 20460, June 1974, EPA-
440/1-74-020-a.
7. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Fish Meal, Salmon, Bottom Fish, Clam,
OyOter, Sardine, Scallop, Herring and Abolone Segment of the Canned and
Preserved Seafood Processing Point Source Category, U.S. Environmental
Protection Agency, Wash., D.C., 20460, September 1975, EPA-440/1-
75/041a.
18
-------
8. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Beet Sugar Processing Subcategory of the
Sugar Processing Point Source Category, U.S. Environmental Protection
Agency, Wash., D.C., 20460, January 1974, EPA-440/l-74-002-b.
9. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Feedlots Point Source Category, U.S.
Environmental Protection Agency, Wash., D.C., 20460, January 1974, EPA-
440/1-74-004-a.
10. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Red Meat Processing Segment of the Meat
Product and Rendering Processing Point Source Category, U.S. Environ-
mental Protection Agency, Wash., D.C., 20460, February 1974, EPA-440/ 1-
74-012-a.
11. Development Document For Proposed Effluent Limitations Guidelines and
New Source Performance Standards for the Processor Segment of the Meat
Products Point Source Category, U.S. Environmental Protection Agency
Wash., D.C., 20460, August 1974, EPA-440/1-74/031.
12. Development Document for Proposed Effluent Limitations Guidelines and
New Source Performance Standards for the Poultry Segment of the Meat
Product and Rendering Point Source Category, U.S. Environmental Pro-
tection Agency, Wash., D.C., 20460, April 1975, EPA-440/I-75/031-b.
13. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Leather Tanning and Finishing Point Source
Category, U.S. Environmental Protection Agency, Wash., D.C., 20460,
March 1974, EPA-440/l-74-016-a.
14. E.I. duPont de Nemours & Co., et. al. v. Train, Administrator,
Environmental Protection Agency, et. al., 97 S. Ct. 965 (1977).
15. Resource Conservation and Recovery Act of 1976, P.L. 94-580, October 21,
1976, 42 USC 6901, et. seq.
16. Safe Drinking Water Act, P.L. 93-523, December 16, 1974, 42 USC 300 f.,
et. seq.
17. Carter, Jimmy, "My Views on the Environment", EPA Journal, Vol. 3, No.
One, January 1977.
19
-------
EFFLUENT POLISHING AND WASTEWATER REUSE
AT SNOKIST GROWERS CANNERY
by
Larry A. Esvelt*, Herbert H. Hart**
INTRODUCTION
Snokist Growers is a growers cooperative located in Yakima, Washington
which markets fresh and canned fruit to wholesale and retail outlets. The
cooperative owns and operates a cannery to produce canned pears, peaches,
cherries, plums, and apple slices, sauce and rings. A normal processing
season consists of about 9 weeks production of pears at about 250 metric tons
raw fruit per day, a week of peach canning at about 250 metric tons raw fruit
per day, 15 weeks of apple processing at about 100 metric tons of raw fruit
per day, and varying cherry and plum production depending on the season.
In 1968 Snokist Growers Cannery placed a biological wastewater treatment
system in operation which has consistently produced an effluent well within
their State and EPA discharge permit. It was evaluated under an EPA grant
(1,2,3) and was subsequently used as an exemplary waste treatment system
during the development of 1977 best practicable treatment guidelines for food
processing wastewater discharges. In 1973, a low water year, Snokist manage-
ment became concerned about the integrity of their water supply system which
appeared to be diminishing due to ground water levels dropping at their wells
and decided that additional water sources were needed. They then investiga-
ted the feasibility of an additional well and of reclaiming a portion of
their biologically treated process effluent for use in the cannery. No sani-
tary wastes enter the processing wastewater stream, minimizing the potential
for human pathogen presence in the wastewater. The lower cost alternative
was development of a new well supply but word of the consideration of re-
claiming effluent reached EPA officials in charge of evaluating the possibil-
ity of reducing food processing wastewater emissions through reclamation and
recycle in compliance with PL 92-500. They suggested the possible availabil-
ity of R & D funds to provide Snokist with economic incentive to select the
effluent reclamation and reuse alternative rather than developing a new well
water supply. Snokist then applied for an R & D grant and received EPA grant
number S803280 in late 1974. The grant was to partially offset the cost
differential between the two alternatives and to finance evaluation of the
reclamation and reuse of treated processing wastewater. Harold Thompson of
the EPA, Con/all is, OR has served as Project Officer.
*Dr. Larry A. Esvelt, P.E. is principal of Esvelt Environmental Engineering,
Spokane, Washington
**Herbert H. Hart is Pollution Control Manager at Snokist Growers, Yakima,
Washington
20
-------
Objectives for EPA R & D Project S803280 were as follows:
1. Determine the feasibility of treating fruit processing wastewater to
achieve a suitable water quality for in-plant reuse and to develop
operational procedures to ensure consistent performance of the treatment
facility.
2. Determine the feasibility of reusing the treated fruit processing waste-
water for:
a) Equipment cleaning
b) Product cleaning and conveying
c) Boiler feed to produce steam for:
1) Cleaning
2) Exhausting
3) Cooking
4) Blanching
d) Direct contact container cooling
3. Document the reduction of pollutants being discharged to the environment
resulting from the reuse of treated processing wastewater and evaluate
the economics of wastewater reuse for achieving EPA's 1983 effluent
standards.
Facilities for biological effluent polishing were installed in 1975 and
were used to provide reclaimed effluent for reuse during the 1975 and 1976
process seasons. The reclaimed effluent quality was routinely monitored for
chemical and biological quality. Special testing was conducted to ascertain
the level of specific pollutants such as heavy metals, pesticides, halogen-
ated organics and pathogens.
Use of the reclaimed water was routinely applied to floor and gutter
wash and applied to the uses to be evaluated under the project objectives
for varying periods. The results of such uses on product quality was
assessed.
WASTEWATER TREATMENT FACILITIES
Snokist Growers cannery wastewater treatment facilities were constructed
in 1968 to accpmodate a flow up to 2.5 mgd and BOD of 28,000 Ib/day. A
schematic flow diagram for the treatment system is shown on Figure 1. Table
1 describes the system components. The activated sludge system produced a
high quality effluent for discharge to the Yakima River (1,2,3).
Before the 1975 processing season additional facilities were added to
polish the biologically treated effluent by multimedia filtration and dis-
infection. The flow schematic during the 1975 and 1976 processing seasons
is shown on Figure 2. The facilities added for the wastewater reclamation
and reuse are listed in Table 2.
-------
SCREENED WASTE
FROM CANNERY
METERING
NUTRIENT(N, P)
ADDITION
TO RIVER
SLUDGE
REAERATION
BASIN
DISSOLVED AIR
FLOTATION SLUDGE
THICKENER
PRESSURIZATION |
Y
AERATION
BASIN
WASTE FLOW
SLJJDGE FLOW
Figure 1. Snokist Growers Wastewater Treatment System Schematic
Flow Diagram 1968 - 1974.
22
-------
SCREENED WASTE
FROM CANNERY
WASTEWATER
REUSE IN
CANNERY
METERING
.*>
NUTRIENT
X~ADDITION
CHLORINE
MULTI MEDIA
FILTERS
FLOTATION SLUDGE
THICKENER
I WASTE SLUDGE
|
%—=0
WASTE FLOW
SLUDGE FLOW
Figure 2. Snokist Growers Wastewater Treatment System Schematic
Flow Diagram 1975 - 1976.
23
-------
TABLE 1. SNOKIST GROWERS WASTEWATER FACILITIES
Facility
Description
1. Screening
2. Aeration Basin
3. Clarification
4. Sludge Recirculation
5. Sludge Reaeration
6. Sludge Thickening
8 mesh/cm (20 mesh/in) vibrating screens
22,700 cubic meter (6 million gallon) earthen
dike, PVC lined basin with 5 surface aerators
having a total of 292 kw (390 horsepower)
27.5 meter (90 ft.) Diameter, hydraulic
sludge removal, 2.4 m (8 ft.) side water
depth, center feed.
Two variable speed pumps each with 6,600
liter per minute (1750 gallon per minute)
capacity
5,700 cu. meter (1.5 million gallon) basin
with 45 kw (60 horsepower) surface aeration
9.2 meter (30 ft.) Diameter pressurized
recycle flotation sludge thickener
TABLE 2. SNOKIST GROWERS WASTEWATER POLISHING FACILITIES CONSTRUCTED 1975
Facility
Description
1. Filters
2. Turbidity Meter
Two 2.4 meter (8 ft.) diameter by 1.8 meter
(6 ft.) side pressure filters. Area= 4.7 sq.
meters (50 sq. ft.) each. Media = Microflec
MF 177 - 90.5 mm (36 in.) filter: 30% 1.5 sp. gr.
anthracite (3mm); 30% 1.6 sp. gr. anthracite;
30% 2.6 sp. gr.silica sand; 10% 4.0 sp. gr.
garnet sand (0.25 mm) supported on 3 inches of
1-2 mm 4.0 sp. gr. sand and 11 inches graded
silica gravel.
Equipped with pipe underdrain, surface wash,
pneumatically operated automatic valves,
automatic backwash program, flow, headloss
meters and automatic flow control.
Low range, continuous flow - Hach CR
3. Filter and Backwash Pumps Two constant speed 3800 liter (1000 gal)
per min @ 20 meter (66 ft.) TDH/2600 liter
(700 gal) per min. @ 23 meter (75 ft.) TDH
pumps, interchangeable. 22.5 KW
24
-------
TABLE 2. Continued
4. Chemical Feed Pumps
5. Reclaimed Water Pump
6. Chlorinator
Liquid alum storage and automatic stroke
adjustable feed pump. Polymer stroke adjust-
able feed pump. To be used if needed.
Split case 2600 liter (700 gal) per min @ 54
meter (177 ft.) TDH pump. 37 KW.
One 227 kg (500 Tb.) per day chlorinator with
motorized "V-notch" control valve and motorized
vacuum valve for "compound loop" control.
7. Chlorine Residual Analyzers Two wastewater type amperometric continuous
flow analyzers for monitoring and controlling
chlorine residual at the filter effluent and
for monitoring chlorine residual at the re-
claimed water pump inlet.
8. Chlorine Contact Chamber Two hundred twenty seven cu. meter (60,000 gal)
and Backwash Water
Storage
baffled chamber - 11.6 meters (38 ft) X 6.7
meters (22 ft) x 3 meters (10 ft) deep with
6 baffles.
9. Controls and Operation a.
b.
c.
Flow to filters automatically maintained
according to chlorine contact level up to
a preset maximum rate per filter.
Filter backwash initiated by timer, high
head loss across filters or manually.
Chlorine residual automatically maintained
by flow proportioning and residual monitor-
ing and feed rate adjustment.
d. Chemical feed of alum and/or polymers, if
used, paced to filter flow.
e. Reuse pump automatic shutdown at low
contact tank level.
f. Alarms transmitted to wastewater lab and
plant for appropriate action due to follow-
ing:
1) Low or high chlorine residual in-
reclaimed water
2) High turbidity in filtered water
3) Low contact tank/backwash storage level
4) System malfunction
25
-------
STUDY OF WASTEWATER REUSE
The wastewater treatment system performance was monitored for perfor-
mance during the 1974, 1975 and 1976 processing seasons under this study.
Monitoring points included the screened wastewater, aeration basin effluent,
clarifier effluent, filter effluent and chlorinated reclaimed wastewater
reused in the cannery. Snokist Growers tested the wastewater for temperature
pH, COD, BOD, suspended and volatile suspended solids, nitrogen, phosphorus,
turbidity, chlorine residual, hardness, alkalinity, total bacterial plate
count and total and fecal coliform bacteria at various of the monitoring
points. Samples from the reclaimed wastewater, and from Snokist's water
supply for comparison, were analyzed by National Canners Association for
heavy metals, by the Environmental Protection Agency Region X Laboratory
for pesticides and volatile halogenated organics, by Foremost Laboratories
for volatile halogenated organics, by Dohrmann Laboratories for total halo-
genated organics and by Columbia Laboratories for herbicides.
The reclaimed water was used in the cannery for floor and gutter wash
throughout the 1975 and 1976 processing seasons. Trial uses of the reclaimed
water were for equipment washdown, initial product conveying, contact con-
tainer cooling and generation of steam for use in equipment washdown, ex-
hausting, cooking and blanching. The effect of using the reclaimed water
for these purposes was compared with using the regular house water supply.
Total Plate Count of bacteria was used to compare the effect of using the
alternate waters and steam sources for washing peelers and belts on parallel
pear processing lines. Total Plate Count was also used to compare the
reclaimed and house waters for use in bin dump tanks and initial conveying
of peaches and apples. The counts were taken of the water in the dump tank
and of water used to rinse the fruit (a standard number of fruit in a
standard amount of water) before and after contact with the dump and conveying
water. Cans of fruit cooled in the two waters under similar conditions
were stored for a given time period and observed for failure. Product quality
evaluation by grading (USDA Grades) and organoleptic comparison was used to
compare steam from the reclaimed water source with house steam.
26
-------
RESULTS
Snokist Growers cannery process wastewater flow rate during pear and
peach processing averaged from 5,500 to 6,500 cu meters/day (1.5 to 1.7 mgd)
during the 1974, 1975 and 1976 processing seasons and 2,500 to 3,500 cu.
meters/day (0.7 to 0.9 mgd) during apple processing. The untreated waste-
water COD during pear processing averaged over 2000 mg/1 in 1974, about
1800 mg/1 in 1975 and about 1500 mg/1 in 1976. The reduction from 1974 to 1975
resulted from additional water use from recycled effluent for floor and
gutter wash. From 1975 to 1976 water use was cut back slightly and peel and
core solids were removed separately resulting on a lower waste load in the
effluent to the treatment system.
The wastewater is nutrient deficient so nitrogen and phosphate are
added prior to treatment. The aerators maintain over 2 mg/1 dissolved oxygen
in the activated sludge system and the sludge recirculation rate approximate-
ly equals the process wastewater flow rate.
The treatment system performance was monitored over the three seasons.
Effluent emission rates are shown on Tables 3,4, and 5. The tables show that
variations in emission rates occurred from week to week. The tables also
show that the emission rate for suspended solids and COD actually increased
from 1974 to 1975 even though over 30% of the effluent was recirculated for
use in the cannery during 1975 and the amount of wastewater discharged re-
duced by nearly 30% on a per unit of product basis. The biological effluent
was of poorer quality in 1975 than in 1974 and the reason was thought to be
increased chlorine useage in the cannery for cleanup which allowed chlorina-
ted slugs to reach the aeration system at startup each morning.
During the 1976 season additional attention was given to control of in-
plant chlorination which resulted in better biological effluent quality and
reduced emission rates for suspended solids, COD and BOD throughout pear
processing except for one week immediately following Labor Day (September 6).
The biological treatment system was apparently hit by a highly toxic slug of
accumulated chlorinated washdown from belts and equipment at startup on
September 7. The system recovered rapidly when reseeded with the contents
of the small aeration basin.
The emission rates for pollutants jumped sharply in 1976 when pear pro-
cessing ended and apples only processing was initiated. The plant may have
been receiving occasional chlorinated slugs but the principal reason for
effluent deterioration is thought to be the onset of cold weather at about
the same time. From near the end of November 1976 through January 1977 the
aeration basis did not exceed 2° Celcius.
27
-------
TABLE 3.
1974 POLLUTANT EMISSIONS
SNOKIST GROWERS WASTEWATER REUSE PROJECT
PROGRESS REPORT
Week
8/23-24
8/26-31
9/3-7
9/9-14
9/16-20
9/23-28
9/30-10/5
10/7-12
10/14-19
10/21-26
10/29-11/2
11/4-9
11/11-15
11/18-22
11/25-27
12/2-6
12/9-13
Fruit
Processed*
kkg
518 Pr
1616 Pr
1319 Pr
1305 Pr,121 PI
624- Pr,426 PI
1636 Pr,236 A
1600Pr,519A,20Pl
1546 Pr,587 A
1588 Pr,595 A
1618 Pr,592 A
1335 Pr, 538 A
1438 Pr, 576 A
648 A
819 A
234 A
386 A
295 A
COD Emission BOD**
Total kg/ kkg Emission
kq Fruit kg/ kkg
1590
4520
1030/4 da
1025
810
1000
765/5 da
1830
1750
1930
770/3 da
1095
1260
770/4 da
255
4440***
540
3.07
2.80
0.98
0.72
0.77
0.53
0.43
0.86
0.80
0.87
0.69
0.54
1.94
1.18
1.09
11.5
1.83
-
3.00
0.06
0.06
-
0.07
--
-
0.20
0.15
0.13
0.09
0.64
0.09
0.14
0.19
0.16
Suspended Solids
Emission
Total kg kg/ kkg
1050
1430
189/4 da
229
230
450
309/5 da
785
486
825
473
428
474
140/4 da
136
3530***
310
2.03
0.88
0.18
0.16
0.22
0.24
0.17
0.37
0.22
0.37
0.25
0.21
0.73
0.21
0.58
8.63
1.05
* Pr = Pears, Pe = Peaches, PI = Plums, A = Apples
** 1 day or 2 days during week only
***! day high results skewed results
Total Wastewater Flow 8/23 - 12/13/1974 = 460,600 cu.m.
Total Fruit Processed 8/23 - 12/13/1974 = 22,800 kkg
Wastewater Discharge Rate = 20.2 cu.m./kkg (4770 gal/ton)
28
-------
TABLE 4.
1975 POLLUTANT EMISSIONS
SNOKIST GROWERS WASTEWATER REUSE PROJECT
PROGRESS REPORT
Week
8/26-30
9/2-6
9/8-13
9/15-20
9/22-27
9/29-10/4
10/6-11
10/13-18
10/20-25
10/28-11/1
11/3-8
11/10-15
11/17-19
11/20,21,24,25
12/1-5
12/8-12
12/15-18
Fruit
Processed*
kkg
1407
1358
1664
1644
1311
1734
1620
1598
1666
1334
1730
1670
789
Pr
Pr
Pr,
Pr,
Pe,
Pr,
Pr,
Pr
Pr,
Pr,
Pr,
Pr,
Pr,
46 PI
520 PI
280 PI
14 PI
473 A
550 A
498 A
570 A
682 A
397 A
570 A
840 A
881 A
291 A
COD Emission
Total kg/kkg
kg Fruit
1300
880
1940
1700
1330
1230
4240
3320
2590
1190
1990
1130
350
362
1540
1070
455
0
0
I
1
0
0
0
2
2
1
0
0
0
0
0
1
1
1
.92
.65
.13
.79
.84
.70
.03
.08
.17
.65
.87
.48
.30
.64
.83
.21
.56
BOD**
Emission
kg/kkg
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
16
076
22
12
105
070
31
38
14
054
13
115
033
39
25
26
Suspended Solids
Emission
Total kg kg/kkg Fruit
592
346
817
625
630
627
3290
2230
1590
482
829
505
88
125
1035
585
161
0
0
0
0
0
0
1
1
0
0
0
0
0
0
1
0
0
.42
.25
.48
.29
.40
.36
.57
.40
.72
.26
.36
.21
.074
.22
.23
.66
.55
*Pr = Pears = Pe = Peaches, PI = Plums, A = Apples
**1 day per week only
Total Wastewater Flow 8/26-12/19/75 = 557,500 cu.m
Total Wastewater Reused = 180,100 cu.m.
Total Effluent to River = 377,400 cu.m.
Total Fruit Processed 8/26-12/19/75 = 26,500 kkg
Wastewater Flow Rate = 21.0 cu.m./kkg (4970 gal/ton)
Effluent Flow Rate = 14.2 cu.m./kkg (3360 gal/ton)
29
-------
TABLE 5,
1976 SEASON POLLUTANT EMISSIONS
SNOKIST GROWERS WASTEWATER REUSE PROJECT
Fru i t
Processed
Week kkg
8/24-8/28 1529 Pr
8/30-9/4 1760 Pr
9/7-9/11 1520 Pr
9/13-9/18 1622 Pr, 473 PI
9/20-9/25 1601 Pr, 482 PI
9/27-10/2 1490 Pr, 29 Pe, 14
10/4-10/9 1828 Pr, 529 A
10/11-10/16 1808 Pr, 563 A
10/18-10/23 1830 Pr, 551 A
10/25-10/30 1674 Pr, 569 A
11/1-11/6 1744 Pr, 575 A
11/8-11/13 1452 Pr, 459 A
11 /1 5-1 1/20 646 A
11/22-11/24 422 A
11/29-12/3**** 566 A
12/6-12/10*** 714 A
12/13-12/17*** 728 A
12/20-12/22*** 418 A
12/27-12/30*** 549 A
1/3-1/7**** 678 A
1/12-1/14**** 397 A
1/17-1/21**** 741 A
1/24-1/28 & 1/31***958 A
2/23-2/25*** 391 A
2/28-3/4 588 A
3/7-3/10 494 A
COD Emission BOD**
Total Unit Emission
kg kg/kkq kg/ kkg
449
514
3854
742
1020
PI 610
694
772
845
1001
957
1146
584
408
1127
867
602
470
315
573
2797
4460
1938
570
709
238
.326
.292
(5.2 max da)
2.54
.35^
.490
.398
.294
.326
.355
.446
.413
.600
.904
.967
1.99
1.21
.827
1.13
.574
.845
7.05
6.02
2.02
1.46
1.21
.482
.078
.027
.464
.052
.133
.039
.027
.030
.053
.041
.044
.099
.066
.137
.336
.245
.165
.111
.57
.238
.047
Suspended Solids
Emission
Total kg Unit kg/kkg
172
118
1022
196
309
147
211
232
362
295
403
364
155
178
751
606
366
266
149
488
686
1627
1277
439
449
140
.112
.067
(1.09 max da)
.672
.094
.148
.096
.090
.098
.152
.132
.174
.190
.240
.423
1.33
.849
.503
.636
.271
.720
1.73
2.20
1.33
1.12
.764
.283
* Pr = Pears, Pe = Peaches, pi = Plums, A = Apples
** One or two days per week data only
*** Aeration Basin Temperature less than 2°C
**** Aeration Basin partially or completely frozen over
(Continued)
30
-------
TABLE 5.(Continued)
Total Wastewater Flow, cu. m.
Total Wastewater Reused, cu. m.
Proportion Reused
Total Effluent to River, cu. m.
Total Fruit Processed, kkg
Wastewater Flow Rate, cu. m./kkg
Wastewater Flow Rate, gal. /ton
Effluent Flow Rate, cu. m./kkg
Effluent Flow Rate, gal ./ton
8/24-11/13
422,720
159,660
37.8%
263,060
24,102
17.5
4,210
10.9
2,617
11/15-3/10
209,750
60,340
28.8%
149,410
8,290
25.3
6,080
10.9
4,330
Overall proportion reused = 34.8%
Overall Effluent Flow Rate = 12.7 cu. m./kkg
3,060 gal./ton
31
-------
Reclaimed Hater Quality
The reclaimed effluent quality is highly dependent upon the biological
effluent quality. Figure 3 shows the frequency distribution for the biolo-
gical and filtered effluents during the pear processing season of 1974, 1975
and 1976. Generally the filter system is effective at reducing the suspended
solids, but only by about 30 to 50% and then only when the solids load to the
filters is relatively low. Figure 3 shows the relatively poorer effluent and
reclaimed water produced in 1975 compared with the other years.
The chlorination system was effective at disinfecting the reclaimed
water. Figure 4 shows a frequency distribution for chlorine residual and
total and fecal coliform organisms in the reclaimed effluent during the
1976 season. The poorest disinfection results occurred during the week of
biological system upset following Labor Day when chlorine demand exceeded
chlorination capacity. Other poor disinfection results occurred when equip-
ment malfunctions allowed the chlorine residual to fluctuate. Figure 4
indicates that the chlorine residual maintained at 3 mg/l resulted in less
than one total and fecal coliform organism per 100 ml. A chlorine residual
of 2 mg/l resulted in total coliform organism concentration of about 2 per
100 ml or less. Analyses for Salmonella and Staphylococcus organisms were
negative.
Heavy metals analyses were conducted on the reclaimed water and com-
pared with Snokist Growers cannery house water supply (ground water). Iron,
arsenic, cadmium, tin and manganese were consistently below the detectible
limits in both waters. Lead was consistently below detectible limits in the
house water but was detected at 0.01 and 0.02 mg/l in two of five samples
of reclaimed water analyzed in 1976. Mercury was found at from 0.0003 to
0.0009 mg/l in 4 samples of reclaimed water and was not detectable in 4
samples. Mercury was 0.0003 to 0.0010 in three house water samples and
undetected in two samples. One sample of each analyzed on the same day
showed high mercury levels (0.0026 and 0.0017 mg/l in reclaimed and house
water respectively) which is assumed to be analytical error. Zinc concentra-
tions ranged from 0.15 to 0.63 mg/l in the reclaimed water and from less than
0.01 to 0.02 mg/l in the house water.
Pesticide and PCB analyses were performed by the EPA Region X laboratory.
All results were less than the detectable level except PCB's which were de-
tected in the house water on four occasions in 1976 at up to 0.065 micro-
grams per liter.
Analyses for organohalides was conducted by Foremost and Dohrmann
Laboratories in California in 1976. Chloroform in the chlorinated reclaimed
effluent ranged from 1.5 to 25 micrograms per liter while other volatile
organohalides were at or below 1 microgram per liter. Total volatile organic
halides were measured at about 13 mocrograms per liter in the reclaimed
effluent compared to 3 to 7 micrograms per liter measured in the house water.
32
-------
.05
Q
W
Q
Z
LU
100
<:
70
50
40
30
20
10
T
ACTIVATED SLUDQE EFFLUENT
FILTERED EFFLUENT
I I I I I
I I
Figure 3.
5 10 20 30 40 50 60 70 80 90 95 98 99
FREQUENCY LESS THAN,%
Suspended Solids in Biological and Filtered Effluents during
Pear and Peach Processing - 1974, 1975, 1976 Seasons.
Snokist Growers Cannery.
33
-------
X.
0,5
< 4
Q
c7>
LU
oc 3
UJ
z
cr 2
o ^
o
0
10
TOTAL COLIFORM
/FECAL
/ COLIFORM
12
TIME
5 10 20 30 40 50 60 70 80
ORGANISMS LESS THAN 8 CL2
90 95 98 99
GREATER -, "%
Figure 4. Chlorine Residual and Total and Fecal Coliform in Reclaimed
Effluent. Snokist Growers Cannery 1976 Processing Season.
34
-------
Trial Uses of Reclaimed Water
The reclaimed water with house steam and with steam generated from the
reclaimed water was tested for equipment cleaning against house water with
house steam. No differences could be detected in equipment sanitation.
The reclaimed water was used for initial product cleaning and conveying.
Its use in the peach bin dump tank for 3 days was compared to 3 days with
house water. Results are shown on Figure 5. Figure 6 shows the results of
using the reclaimed and house water in the apple dump and initial conveying
area. No differences in product quality resulted from use of the reclaimed
water.
The reclaimed effluent was of equal quality in both hardness and silica
to the house water for consideration as a boiler feed source. It was used
in a portable steam generator to provide steam for comparison with house
steam for cleaning, exhausting, cooking and blanching. As noted above no
differences in cleaning were detected. Pears exhausted with reclaimed water
steam could not be significantly distinguished from the house steam exhaust-
ed controls. Apple sauce cooked with house and reclaimed water steam and
apple slices blanched with the two steams were different but differences
were judged to result from varying degrees of carmelization due to heat and
time differences in the batch processes used. The normal continuous proces-
ses could not be used with the portable steam generator.
The reclaimed water was used for trial runs of direct contact container
cooling. During the 1975 season about 1200 cans were cooled in reclaimed
water and an equal number of controls retained. One hundred cans from each
cooling water were stored at 30° C for 3 months. No differences were appar-
ent on opening. One thousand cans from each cooling water were stored at
13° C for a year. No failures occurred in either set. During 1976 over
3000 cans were cooled in each the reclaimed and house water. Approximately
100 of each were stored at 35° C for 6 months with no failures and no dis-
cernable differences. The additional 3000 from each cooling water have been
stored 6 months at about 18° C with no failures noted.
DISCUSSION
The process effluent being treated and reclaimed by Snokist Growers
Cannery appears to be suitable for use in several areas of the cannery. The
uses include initial product dumping and conveying, washdown of floors,
gutters and equipment, boiler feed and direct contact container cooling. The
greatest beneficial use at Snokist appears to be for container cooling with
the cooling water subsequently used for floor and gutter wash. A 50% reduc-
tion in wastewater discharge and pollutant emission is anticipated when this
use is implemented.
Cost of the filter system, chlorination system, pumps and piping were
approximately $300,000. Operation and maintenance costs are expected to be
about $5,000 per year. Approximately 90 million gallons of effluent may be
reclaimed and reused during a typical processing season for which the allo-
cated cost is estimated at about $0.35 per thousand gallons including
35
-------
DUMP TANK WATER
10
I07-
io6_
io5-
10
TPC/ml
/
RECLAIMED WATER *
PEACHES AFTER DUMP TANK
TPC /
PEACHES BEFORE DUMP TANK
TPC /g
DAY I
DAY 2
DAY 3
10
Figure
5. Total Plate Count in Dump Tank and on Fruit
Using Reclaimed and House Water..- Peaches.
UJ
m
o
z
a.
UJ
a. x
2 z
^ <
Q I-
E
§ I05
o
z
l
-------
amortization. This is not a competitive price with other supply sources in
the Yakima area but the value for meeting 1983 effluent standards is added
justification for considering this type of system in the future.
SUMMARY
The results of a two-year study of reclamation and reuse of treated
process wastewater at Snokist Growers Cannery show the reclaimed water qual-
ity to be suitable for use in the following areas of the fruit cannery:
1) Initial fruit wash and conveying.
2) Washdown of fruit peeling and conveying equipment.
3) Steam generation boiler feed.
4) Floor and gutter wash.
5) Direct contact container cooling.
The consistent suitability for use in these areas is contingent upon
a well disinfected, low suspended solids product from the biological effluent
polishing system which consists of mixed media filtration and chlorination.
The polishing system performance is, however, highly dependant on performance
of the biological treatment system (activated sludge) in producing a low sus-
pended solids effluent.
Reclaimed water reuse during the 1976 process season resulted in approx-
imately 35% reduction in wastewater discharged. Projections for full scale
cooling use indicate a 50% or greater effluent reduction will be achieved.
REFERENCES
1. Aerobic Treatment of Fruit Processing Wastes. FWQA Report DAST-8.
Snokist Growers, Yakima, Washington (1969)
2. Esvelt, L. A. and H. H. Hart. "Treatment of Fruit Processing Waste
by Aeration," JWPCF, 42_:1305 (1970).
3. Esvelt, L. A. "Aerobic Treatment of Liquid Fruit Processing Wastes,"
Proc. First National Symposium on Food Processing Wastes.
EPA, PNWL, Corvallis, OR (1970).
37
-------
CONTROL OF ODORS FROM AN ANAEROBIC LAGOON
TREATING MEAT PACKING WASTES
by
J. A. Chittenden*, L. E. Orsi**, J. L. Witherow***,
and W. J. Wells, Jr.****
INTRODUCTION
The warm, highly concentrated wastes from a meat packing operation are
uniquely suited to the use of anaerobic lagoons to provide a high degree
of pretreatment prior to final aerobic treatment. The advantages of
anaerobic lagoons include minimum design removal efficiencies of BOD,
grease, and suspended solids of 80%. The anaerobic process has minimum
capital and operating costs, is simple to operate, mechanical equipment
is not necessary, and the treatment processes can withstand the shock
loadings common in the food processing industry.
One major disadvantage associated with the anaerobic lagoons is the odors
that result from such a process. This problem has resulted in many
companies seeking other treatment alternatives at considerable penalties
in capital and annual operating costs.
The cause of these odors and a successful method of eliminating the problem
is the subject of this paper. Also discussed is a conceptual process for
recovering a significant amount of wasted energy by utilizing the heating
value of the methane generated by the anaerobic process. The financial
incentive for using the anaerobic process as opposed to a completely aerobic
system for the treatment of meat packing wastes in a proposed packing plant
is also presented.
BACKGROUND
The anaerobic digestion process utilizes bacteria which function in the
absence of free oxygen to break down organic waste. The waste material
is converted through a number of intermediate products to water, gases and
solids of lesser molecular weight. The bacteria use bound oxygen to survive
and obtain it from organic compounds, water and oxides of nitrogen and
sulfur. The gases produced are mainly methane and carbon dioxide which are
odorless. However, reduction of sulfur-containing organic matter and
sulfates produce organic sulfides, occasionally disulfides in the C-t-Cc
range, and hydrogen sulfide (1).
Hydrogen sulfide (H2S) is usually the major cause of objectionable odor
from the anaerobic process. It has an odor characteristic of rotten eggs,
* Texas Amarillo Systems Company, Amarillo, Texas
** Wilson & Co., Oklahoma City, Oklahoma
*** U.S. Environmental Protection Agency, Corvallis, Oregon
**** Bell, Galyardt, & Wells, Omaha, Nebraska
38
-------
which can be detected at very low concentrations of between 1.0 x 10
and 1.0 x 10~4 mg/l in water (2) and 4.7 x 10~4 ppm in air (3). Hydrogen
sulfide is also a toxic gas, having a threshold limit of 10 ppm for indus-
trial exposure with concentration of 20 - 150 ppm causing eye irritation.
A thirty minute exposure to 500 ppm of hydrogen sulfide can result in dizzi-
ness, headache, staggering, loss of consciousness, diarrhea, bronchitis and
broncho-pneumonia. Finally, exposure to 800 -1000 ppm can be fatal in
30 minutes or less (4) . The toxicity of the gas is of concern in confined
structures such as sewer or wetwells, but its odor is of primary importance
to the utilization of lagoons.
The reduction of sulfate to sulfide under anaerobic conditions is well
established. Sulfate is considered to be the source of almost all the sul-
fide in the anaerobic lagoon. Part of the sulfide produced will combine
with metal ions, such as iron, and become insoluble. Most of the sulfide
usually remains soluble as hydrogen sulfide and, at the near neutral pH
in the anaerobic lagoons, partially dissociates into hydrogen and bisul-
fide ions. The sulfide remaining as H2S in solution will escape into the
air until its partial pressure is in equilibrium with the H2S in solution.
Thus, the odor of hydrogen sulfide from an anaerobic lagoon is proportional
to the sulfate in the waste water treated. The source of sulfates is
the water supply. Sulfate concentrations are not thought to be increased
by the meat packing operations. From experience, the meat industry and
state regulatory agencies have not usually considered using anaerobic lagoons
when the sulfate in the water supply exceeded 200 mg/l, and some are reluc-
tant with concentrations over 100 mg/l.
Two laboratory investigations (5) (6) have reported on the theory of sul-
fide production and have developed basic information under controlled condi-
tions. Lawrence, et al., combine the dissociation and gas equilibrium
equations to form equation 1. that can be used to calculate the ratio of
concentrations of soluble sulfides in the water to hydrogen sulfide in the
gas.
[TSSl
=
Where TSS is HS + H2S in water,
H0S is hydrogen sulfide in gas,
I &
H is hydrogen ion concentration in water,
a is the absorption constant, and
K, is the ionization constant.
The Handbook of Chemistry and Physics has values on a and Kj_, at various
temperatures. Data collected during the investigation gave somewhat higher
ratios of (TSS/H2Sg) than was calculated by the equation. From the experi-
mental data, the ratio would be expected to vary from 4 to 8 at the normal
pH of digesters.
39
-------
They operated several 20 liter anaerobic digesters to which known amounts
of sulfate were added to the influent. When 200 and 400 mg/1 of 804-8
was added, the equilibrium soluble sulfide concentrations were 32 mg/1 as
8 and 78 mg/1 as S, respectively. For the digester receiving 400 mg/1
804-8, a figure in the Lawrence paper shows the sulfides in the gas to be
10 mg/1 as S.
Lawrence, et al., conclude that the concentration of hydrogen sulfide in the
digester is related to the concentration of hydrogen sulfide and sulfide
percursers (sulfates) entering with the waste minus the quantity of hydrogen
sulfides expelled with the gas and that the quantity of sulfides in the
gas is related to the solubility of hydrogen sulfide, the pH, and the total
amount of gases produced. They also determined that up to 400 mg/1 of sul-
fide can be precipitated by added iron compounds with no adverse effect on
the anaerobic treatment process. Precipitation of the sulfides eliminates
the hydrogen sulfide odor.
Gloyna and Espino utilized 430 liter pilot units in developing equation
2. to calculate the sulfide production in lagoons.
2. S= = K (S04=)
S
Where S is 24 hour average sulfide concentration in the lagoon,
SO, is the concentration of sulfate ion in the influent, and
K was determined by the investigation to be:
K = .055 + .00012 (Ib.BOD/ac) + .0016 (detention in days)
The test data is reported in Table 1.
TABLE 1. RESULTS FROM LABORATORY LAGOON STUDIES
Test
No.
1
2
3
4
5
6
BOD
Ib/AC/day
136
68
136
136
68
136
804
mg/1
23
23
23
206
200
400
Temp
°C
25
23
26
23
25
26
Detention
days
30
30
15
30
15
30
Sulfide
mg/1
0.432
0.500
1.12
4.29
6.36
8.76
These two investigations indicate the concentration of hydrogen sulfide
in solution in an anaerobic lagoon will be between 2 and 5 percent of the
804 concentration in the untreated wastewater. Using the established
ratio of soluble sulfide to sulfide in the gas the calculated hydrogen
sulfide concentration escaping to the atmosphere will be nearly one percent
of the 804 concentration in the wastewater.
40
-------
-4
Since the threshold odor in air is 10 ppm for hydrogen sulfide, the
odor of hydrogen sulfide would be present with sulfates at 10~2 mg/1 in the
water supply. Experience has shown that a concentration of sulfates up
to 100 mg/1 in the water produces odors that are accepted, thus disper-
sion of the hydrogen sulfide in the atmosphere is a major factor in odor
control.
Models for dispersion of odors were found in the literature (7) (8) (9).
A simplified model to determine maximum horizontal distance of odor travel
is shown in equation 3..
3. C = k
Co
/x\n
Do
Where Co is initial odor concentration
C is odor concentration at X
X is horizontal distance from the source, and
Do is diameter of the odor source.
When using constants (k = 1, n = -1.5) developed by Shirazi, et al. (9)
the necessary distance to reach a threshold odor was 5 miles from an an-
aerobic lagoon receiving 388 mg/1 SO^. (The design of such a lagoon is
proposed in this paper). An attempt to verify these constants with odor
data collected by Minor and Stark (10) was unsuccessful. However, these
facts agree with experience that odor control by dispersion of H2S from
anaerobic lagoons treating high sulfate wastewaters is impracticable and
perhaps unpredictable.
CASE HISTORIES
Two of the first cases of hydrogen sulfide odor problems with an anaerobic
lagoon treating meatpacking wastes occurred at Storm Lake and Harlan, Iowa.
In both cases, the odors were sufficient to initiate nuisance complaints.
Sulfates in both water supplies were quite high; ranging from 75 to 1560
mg/1 at Harlan and from 126 to 690 mg/1 at Storm Lake. In both cases,
the meatpackers switched to another source of water with lower sulfate con-
centrations to reduce the odor problem.
An anaerobic lagoon was constructed in 1965 to treat packinghouse wastewater
near Cherokee, Iowa. The well water analysis showed an average of 90 mg/1
of sulfate. On a winter day, minor odor was detected by one of the authors
20 feet downward from the lagoon. However, there were signs of metal
deterioration, typical of hydrogen sulfide corrosion, on the doors of the
control building and on a nearby cyclone fence. Occasional odors have been
detected by local citizens, but operation or design changes have not been
necessary to control odors.
In 1970 the City of Spencer, Iowa, had an anaerobic lagoon designed for treat-
ment of municipal wastewater consisting of about 90 percent meatpacking wastes.
47
-------
The meatpacker had his own wells, but they were in the same aquifer as the
municipal wells, which had sulfate concentrations of 1030 to 1050 mg/1.
Because of objections by 108 local residents, odor control was incorporated
in the lagoon design. The engineered design included'. (1) Limiting the meat-
packing wastewater to sulfate concentration of less than 200 mg/1 by develop-
ing another ground water supply. (2) Maintaining a scum layer on the
lagoon to reduce the release of H2S from the lagoon surface. (3) Using
submerged inlets and outlets on the anaerobic lagoon to reduce I^S release.
(4) Maintaining a 7.0 pH to reduce release of H2S by reducing the partial
pressure of the gas. (5) Putting the effluent through a degasifier to
remove the H2S and passing the gas through an ozone chamber to oxidize
the H2S. However, because of the odor potential and later imposed ammonia
limits, the anaerobic lagoon was not constructed.
Anaerobic lagoons are used by a meatpacker in Denison, Iowa. The sulfate
concentration in the anaerobic lagoon influent and effluent averaged 332,
and 39 mg/1, respectively. Hydrogen sulfide in the anaerobic lagoon ef-
fluent averaged 4.6 mg/1. The lagoons were covered with a thick layer of
scum, but the H2S odor was strong at the overflow weir and at a small tank
in which the lagoon effluent was aerated. The treatment plant operator
described the odor as a bad situation and aromatic chemicals were used to
mask the odors.
In Texas, two anaerobic lagoons treating meatpacking wastes are known to
have significant odor problems (11). One packer obtains water from the
city of Sweetwater with a measured sulfate content of 218 mg/1. Personnel
from the Extension Service made three odor intensity measurements on the
downwind side of the lagoon which required 31 (Dt) dilution to reach a
threshold odor. The other packer in Maverick County used water from the
Rio Grande River which had a sulfate concentration of 190 mg/1. Odor
measurements made downwind were 31 Dt alongside the lagoon and 2 Dt at
3/4 of a mile from the lagoon. The packer in Maverick County ceased opera-
tion due to financial problems and odor control became unnecessary. The
Extension Service recommended to the other packer conversion of the anaerobic
lagoon to an aerated lagoon.
In Moerewa, New Zealand, anaerobic lagoons are used to treat settled meat-
packing waste (12). The main criticism of the lagoon was reported as the
odor of hydrogen sulfide associated with the gases. The sulfate concen-
tration of the wastewater is not given, but an analyses of the evolved gas
was: hydrogen sulfide at 0.4%, carbon dioxide at 7.0%, methane at 85.0%,
oxygen at 0.6% and nitrogen and others at 7.0%. A record of hydrogen sulfide
in the air close to the lagoons was kept for 8 months. The highest concen-
tration was an isolated occurrence of 0.99 mg/1, but of the total number
of readings at two-hour intervals, 1897 were nil and 389 were positive with
an average concentration of 0.02 mg/1.
The odor control at Moerewa consisted of maintaining a scum cover on the
lagoon. The effectiveness of the cover in reducing the escape of hydrogen
sulfide was established by air samples taken four inches above the water
or scum cover. The concentration of hydrogen sulfide above the scum averaged
0.35 mg/1, compared to the concentrations ranging from 2.0 to 15.0 mg/1
over the scum free areas.
42
-------
In 1973, the city of Greeley, Colorado, Initiated operation of anaerobic
lagoons which received 95% of their wastewater flow from a large meatpacker.
The water supply at the packing plant contained 700 to 800 mg/1 of sulfates
(13). A severe odor problem occurred at the anaerobic lagoon. After one
year's operation, city water was extended to the packer. The sulfate con-
centration in the city water was about 40 mg/1. Six months after the lower
sulfate water was used, odor continued to be a problem. In July 1975, the
State of Colorado Air Pollution Control Agency found the facility in viola-
tion of air quality standards for odors and subsequently issued a cease and
desist order. These air quality standards have a maximum limit for odor
at the property line of 15 Dt. Five other states have similar standards (8).
Several new operation schemes were undertaken to control the odor. To build
up a heavy scum layer on the surface, all of the plant flow was routed to
anaerobic lagoon No. 3. The addition of straw and grease to aid in developing
a cover was planned as a second step. Such a cover was expected to reduce
odor emissions from the lagoon. After the scum completely covered the
surface, measured odors were below the air quality standards for a two
month period.
However, due to an excess build-up of solids in lagoon No. 3, another lagoon
(No. 1) was put into operation on December 17, 1975, and odor levels in
excess of the air quality standards reoccurred. Beginning on January 23,
the primary treatment process at the packing plant was bypassed to increase
the grease level and more rapidly form a cover on lagoon No. 1. On March 1,
the addition of 10 mg/1 of chlorine to the packing plant effluent was ini-
tiated to reduce sulfide odors at the lagoon inlet. In the 30 days after
the initiation of chlorination, the scum layer went from 35 to 40 percent
cover on lagoon No. 1 to a 99% cover. High winds sometimes broke up the
scum layer and temporarily reduced the percent of surface area covered.
The wastewater became anaerobic in 32,500 feet of force main between the
packing plant and the lagoons. Measurement of the hydrogen sulfide concentra-
tion in the wastewater showed up to a 50 mg/1 decrease at the inlet to the
lagoon when chlorination was practiced. The build-up of the cover and
corresponding decrease in the odor dilutions reading are shown in Table 2.
TABLE 2. PERCENT OF SURFACE AREA COVERED (14)
Date
Lagoon
No.
1
Lagoon
No.
3
Dt
March 30
March 29
March 26 99 100 0
March 25* 99 100 7
March 24 99 100 0
March 22 99 100 0
March 19 60** 60** 7
March 18* 95 100 7
March 17 98 100 0
March 11 85** 100** 7
43
-------
TABLE 2. PERCENT OF SURFACE AREA COVERED (14) (Cont.)
Date Lagoon No. 1
March 9*
March 8
March 3
March 2
March 1*
February 25*
February 19*
February 16
February 13
February 4
February 3*
February 2
January 29*
January 28
January 26
* Odor readings by State
** Days of high winds
95
95
35
45
40
40
20
5**
35
10
10
10
5
5
5
officials
Lagoon No. 3
100
100
100
100
100
100
100
60**
100
100
100
100
100
100
100
D
0
7
0
15
15
0
15
15
15
31
170
—
—
—
^^
The overall effectiveness of the scum cover in reducing odor emissions is
summarized in Figure 1. Only those odor measurements made by the State
Agency personnel are shown on the figure.
During March 1975, a series of sulfate measurements was made by the city
on the packing plant effluent and the influent to the treatment plant and
are shown in Table 3. Additional sulfate measurements on the packing plant
effluent averaged 95 and 90 mg/1 in April and May, respectively.
TABLE 3. SULFATE CONCENTRATIONS
Date Packing Plant Treatment Plant
Effluent (mg/1) Influent (mg/1)
March 2
March 3
March 4
March 5
March 8
March 9
March 15
March 16
March 19
March 22
48
110
45
110
150
—
121
—
66
160
71
47
220
80
120
__
120
140
_w
44
-------
m
co
o
33
Z
CO
O
Z
CD
CO
m
o
-n
31
o
r
co
o ro
i
6/26/75 —
6/30/75 —
II 1/75 —
7/ 2/75 —
7/31/75-
IO/ 7/75 —
10/22/75 -
10/30/75-
ll/ 2/75-
11/12/75-
11/13/75
11/18/75-
11/28/75
I2/ 5/75H
12/11/75
12/17/75-
l/ 7/76
1/15/76-
1/29/76-
2/ 3/76 —
2/1 1/76 —
2/19/76 —
2/25/76
3/ 1/76 —
3/ 9/76
3/18/76 —
3/25/76
o
H-
en
I
100% COVER ON LAGOON #3
LAGOONS I ON LINE
PRIMARY TREATMENT
BY-PASSED
COVERS! IS 35-40%
COMPLETE
^-CHLORINE ADDED
>-COVER ^#1 IS 95% COMPLETE
FIGURE I. ODOR LEVELS WITH VARYING OPERATION (8).
45
-------
TABLE 3. SULFATE CONCENTRATIONS (Cont.)
Date Packing Plant Treatment Plant
Effluent (mg/1) Influent (mg/1)
March 23 ~ 205
March 25 88 80
Average 100 120
The city rechecked its testing procedures and reagents for the sulfate
test and found no discrepancies. Tests on city domestic water showed
804 concentrations in the range of 40 - 60 mg/1. An increase of about
50 mg/1 of sulfates by the meat packing process is shown by this data.
This increase is contrary to the commonly accepted belief that sulfate
concentrations are not increased in the meat packing process.
An anaerobic lagoon was used to treat meatpacking wastewater near Ada,
Oklahoma. The water supply had a sulfate concentration of 4.0 mg/1.
During a two-year period, one of the authors inspected the lagoon over
200 times and detected septic odors within a few feet of the lagoon, but
the odor of H2S was not detected. The lagoon did not have a scum cover and
bubbles at the surface showed considerable gas production. The lack of
odor was attributed to the low sulfate concentration in the water (15).
DESIGN CONSIDERATIONS
A study of acceptable methods of controlling anaerobic lagoon odors was
initiated in the course of the design of a new meat packing facility to
be located in Southwestern Arizona. In order to enable the reader to
evaluate the potential of the concepts presented in this paper, the results
of that study are discussed in detail in the remaining pages of this article.
In the proposed location of the meat packing plant, ground water supplies
were limited and of such poor quality that it was concluded that water would
have to be obtained elsewhere.
The only other available source of water was the irrigation canal that
bordered the property. Discussions with the Bureau of Reclamation and
the local Water and Drainage Districts indicated that approval would be
granted for withdrawing the necessary water.
A typical analysis of the water is given in Table 4. Although the water
was somewhat high in salinity, it would be acceptable for potable use after
suspended solids removal and chlorination. Previous experience had shown,
however, that the high sulfates (388 mg/1) would produce unacceptably high
odor emissions from the anaerobic lagoons.
46
-------
TABLE 4. IRRIGATION WATER ANALYSIS
Constituent Concentration tng/1
Silica 9
Calcium 91
Magnesium 32
Sodium 126
Potassium 6
Bicarbonate 172
Carbonate 0
Sulfate 388
Chloride 107
Fluoride 0.5
Nitrate Nil
Dissolved Solids 796
Hardners as CaC03 360
Non-carbonate Hardners 218
as CaC03
Specific Conductance 1.240
mmhos
The meatpacking company retained a consulting firm to study the problem
and make recommendations. In particular, the consultant was given two
assignments:
1) Develop the capital and operating costs of a completely
aerobic waste treatment system and compare those costs with
the capital and operating costs of an anaerobic lagoon system
followed by a minimum of aerobic treatment.
2) Explore the various alternatives for collecting and treating
the gases generated by the anaerobic lagoon so that acceptable
odor emission would result. In particular, the anaerobic lagoon
cover developed for the Wilson & Co. plant at Monmouth, Illinois
was to be evaluated as a potential candidate for the Arizona
lagoon.
The proposed plant location also dictated two other significant design
parameters. First, the final disposal of the waste water would be cropland
irrigation. There was no stream available to accept treated wastes. The
State required treatment of the waste prior to irrigation to a level that
produced a maximum BODs of 100 mg/1. Since the crops could utilize the
nitrogen in the ammonical form, nitrification was not a problem. The second
factor dictated by the proposed site was that aerobic lagoons for final
treatment of the waste were precluded by the soil conditions on the site.
The soil was all sand and cost estimate comparisons of sealing the large
area lagoons to meet acceptable ex-filtration rates versus the cost of
small area aeration basins indicated that extended aeration would be the
47
-------
cheaper alternative.
The proposed plant was to be capable of slaughtering 2880 head of beef
per day and breaking 2000 beef carcasses per day into boxed beef. Drawing
up the experience of similar plants, the raw and treated wastes characteristics
were formulated. Their design values are shown in Table 5.
TABLE 5. WASTE CHARACTERISTICS
Raw Wastes Treated Wastes
Flow - MGD
BOD mg/1
SS mg/1
Grease mg/1
Total K-N mg/1
2.88
1500
1200
900
200
2.88
60
80
10
160
A summary of the study to evaluate the treatment alternative and the com-
parisons of the candidates for a lagoon cover is given in the following
paragraphs.
Completely Aerobic System
A schematic of the proposed completely aerobic system is given in Figure 2.
The first stage aerobic treatment was to be accomplished with two parallel
fixed media towers using redwood media. The influent BOD5, estimated at
1500 mg/1, would be reduced by 65% in the towers. Recirculation around the
towers would maintain an adequate wetting rate to enhance treatment effi-
ciency .
The second stage aerobic treatment chosen was two parallel aeration basins
each containing 250 hp of either slow speed, fixed platform aerators, or
a dispersed air system using static tube diffusers.
The final clarifiers were used to settle the wastes. The settled solids
were returned to the inlet of the redwood towers so that the fixed media
towers would operate in the activated biological media mode. Provisions
were also made to allow sludge return to the extended aeration basin in
order to maintain the MLSS at desirable levels. Approximately 0.2 MGD of
sludge would be wasted to an aerated sludge holding tank with two days
holding capacity.
Approximately 30 acres of storage ponds to provide 20 days of storage would
be provided after the final clarifiers. To avoid ground water contamina-
tion, the lagoons would be sealed.
Effluent from the ponds would be utilized for irrigation.
The aerated sludge from the sludge holding tanks would be pumped directly
to a separate irrigation system.
48
-------
DIRECT-RECYCLE
vo
LIFT
STATION
i
r
I
FIXED
MEDIA
TOWERS
SLUDGE
*'
1
1
1
AERATION
SYSTEM
SLUDGE
RECYCLE HANDLING
FINAL
CLARIFIERS
IRRIGATION
STORAGE
PONDS
TO
"IRRIGATION
T
VHASTE SLUDGE TO
IRRIGATION
FIGURE 2.
COMPLETELY AEROBIC SYSTEM SCHEMATIC.
-------
Anaerobic/Aerobic System
In order to utilize anaerobic lagoons in this application it would be neces-
sary to have an effective gas collection system to prevent objectionable
odorous emissions.
A study was made of the available covers that could be used in this applica-
tion. A cost summary of this study is shown in Table 6.
TABLE 6. COMPARATIVE ANAEROBIC LAGOON COVER COSTS
Type Basin Size No. Basins Est. Cover Cost
Fiberglass Arch
Precast Concrete Double Tees
Floating Fiberglass
Floating Flexible Membrane
50' x 300'
50' x 300'
130' dia.
200' x 300'
9
9
5
2
$1,419,000
600,000
730,000
253,000
The preceeding costs are for the cover only. The cost of concrete, earth-
work, pipe, fittings and protective coatings is not included. Obviously,
the flexible membrane would be the cover of choice. A flow sheet of the
combination anaerobic/aerobic system is shown on Figure 3.
In order to minimize the cover cost, the design of the anaerobic lagoons
was given careful consideration. The loadings were set on the high side
of normal design criteria - 17.5# BOD/1000 ft3. The water depth was to be
maintained at 20 feet rather than the more conventional 15 feet. Finally,
the slopes above water level were maintained at 4:1, while the dikes below
water level were cut to 1:1 slope.
From the anaerobic lagoon, the waste was to be treated aerobically in two
aeration basins operating in the completely mixed activated sludge mode.
Sludge wasting from the two final clarifiers would be to the anaerobic
lagoons. Because of the high anaerobic lagoon BOD removal efficiency (80%),
the aerator horsepower requirements would be 400 horsepower rather than the
500 horsepower required in the completely aerobic system.
The lined storage ponds prior to irrigation would be identical to that
discussed in the completely aerobic system.
Cost Comparison
The estimated costs for the two alternative systems are shown in Table 7.
50
-------
LIFT
STATION
ANAEROBIC
LAGOONS
I WASTE SLUDGE
FINAL
CLARIFIERS
SLUDGE
RECYCLE
IRRIGATION
STORAGE
PONDS
TO
IRRIGATION
FIGURE a
ANAEROBIC/AEROBIC SYSTEM SCHEMATIC.
-------
TABLE 7. CAPITAL COST COMPARISONS—2.88 MGD PLANT
Completely Aerobic Anaerobic/Aerobic
System System
First Stage Aeration
Anaerobic Lagoons
Second Stage Aeration
Sludge Handling Facilities
Irrigation Storage Ponds -
30 ac. lined
Subtotal
Anaerobic Lagoon Cover w/Burner
Total Cost
$1,266,000
-0-
779,000
375,000
314,000
$2,734,000
-0-
$2,734,000
-0-
$ 194,000
688,000
83,000
314,000
$1,279,000
369,000
$1,648,000
From the preceeding table it can be seen that the covered anaerobic lagoons
followed by a mechanical aerobic system would result in an estimated capital
cost savings of over $1,000,000 for a 2.88 MGD plant.
In addition, there would be an appreciable savings in annual operating cost
as shown in Table 8.
TABLE 8. ESTIMATED ANNUAL OPERATING COST COMPARISONS
Completely Aerobic Anaerobic/Aerobic
System System
Power (2.65C/KWH) $155,000 $ 82,000
Labor 22,000 22,000
Maintenance 19,000 9.000
Total $196,000 $113,000
Anaerobic Lagoon Cover Construction Details
Many attempts to construct a successful lagoon cover have been made over
the years and most of them have been notable for their lack of success.
For one thing, earlier covers lacked the ability to withstand u.v. degra-
dation. Lack of permanent bonding of seams was also a common failing.
Another weakness common to earlier systems was the lack of an adequate gas
removal system. This resulted in large pockets of gas producing bubbles
under the film. One system that was inspected had gas bubbles 10 feet high
and 100 feet in diameter.
52
-------
The system at Monmouth, Illinois, that was supplied by Globe Linings, Long
Beach, California, and designed and specified by Messmair, Stanley and
Associates of Rock Island, Illinois, appears to have met and solved these
problems. The significant design and construction details are discussed
below.
The cover material is five ply, 45 mil composite constructed of two nylon
reinforcing screens bonded to three sheets of DuPont Hypalon 45 synthetic
rubber. The estimated cost of the anaerobic lagoon cover is detailed in
Table 9.
Gas removal is of prime importance. To conduct the gases and to act as a
cover support, four inch by twelve inch styrofoam logs were inserted in a
factory-sealed envelope in the liner. A sketch of this concept is shown in
Figure 4. The logs were placed across the width of the lagoon on 20 foot
centers with one row down the length of the lagoons.
Styrofoam - 4" x 12'
Hypalon Cover-
Factory Sealed Envelope
Gas Passageway
FIGURE 4. CROSS SECTION OF COVER FLOAT SYSTEM
The gas generated by the anaerobic action in the lagoon follows the space
between the styrofoam float and the cover out to the edge of the lagoon.
It was feared that in time the solids build-up on the lagoon surface would
tend to fill up these passageways. To date, this has not occurred. Small
bubbles 6 to 12 feet in diameter and 6 to 10 inches high do form, but when
the cover is lifted to this point, the gas leaked out through the passage-
ways and the cover subsided back to the surface.
A sketch of the edge construction details is shown in Figure 5. The use of
the aluminum hold down plates results in a positive seal holding in the
generated gases. The perforated pipe shown in the sketch and a 103 cfm
1 hp blower was used to collect the gases. The blower must run continuously
to prevent gas build-up under the cover.
Provided with the system would be an approved gas incineration system for
disposal of the gas. A schematic of the gas train is shown in Figure 6.
One other problem encountered ws.s the disposal of accumulated storm water.
A portable 3 inch trash pump is used to pump the rain water out of the
pockets in the center of the lagoon.
53
-------
en
V X 2" ALUMINUM
PLATE (16" LONG)
BOLT CAST IN PLACE
CHAMFERED EDGE
6" PUC PIPE W/ %" HOLES ON I2"CTRS
45 MIL. DUPONT HYPALON COVER
CONCRETE CURB
FIGURE 5.
LAGOON COVER INSTALLATION DETAIL.
-------
COVERED
(ANAEROBIC LAGOON [--
CJ1
en
TO
PLANT
BOILER
HX-
COVERED
_J ANAEROBIC LAGOON
^
SEDIMENT 8 DRIP TRAP
METER
-HP BLOWER
T/-CHECK VALVE
_1
FIGURE 6.
GAS RECOVERY SYSTEM SCHEMATIC.
MANOMETER
PRESSURE RELIEF VALVE
a FLAME TRAP
EXPLOSION RELIEF VALVE
DRIP TRAP
GAS BURNER
-------
ENERGY CONSERVATION
In the above discussion, the final disposal of the gas was proposed as in-
cineration. With current energy shortages this viable source of energy
should not be wasted. A study was then made of its possible uses.
TABLE 9. ESTIMATED PROJECT COST GAS RECOVERY AND DISPOSAL
Item Estimated Cost
5 Ply, 45 mil DuPont Hypalon Membrane (126,600 sq. ft.) $253,200
P.C. Concrete Curb-Wall 12,000
Stainless Steel Anchor Bolts and Nuts 3,900
Aluminum Anchor Plate 4,500
6" Perforated P.V.C. Collection Pipe 10,800
Gas Train Piping, Valves, Blower, Meter & Appurtenances 9,700
Concrete Slab and Gas Equipment Shelter 1,500
Total $295,600
10% Contingency 29.600
Total Estimated Construction Cost $325,200
Engineering, Legal and Fiscal 43,600
Total Estimated Project Cost $368,800
TABLE 10. ANAEROBIC LAGOON GAS ANALYSIS
Methane - Vol. % 65-70
Carbon Dioxide - Vol. % 30-35
Nitrogen Trace
Hydrogen Trace
Hydrogen Sulfide mg/1 16
Estimated Heating Value BTU/ft 650
Gas Production -
ft3/#V.S. Destroyed 12-18
Anticipated Production ft /day 278,000
An expected analysis of the digestor gas is shown in Table 10. From this
table it can be calculated that an average of 180 million BTU's per day
would be generated by burning this gas. In addition to direct incinera-
tion, which would waste this energy, two other methods by which this energy
could be utilized were explored.
56
-------
A Supplement To Fossil Fuels Feeding Existing Boilers
In this alternative various firing schedules were examined such that the
waste gases would be stored and used at intervals to fire one or more of
the plant's boilers. In addition to the expense involved in constructing
and maintaining storage facilities, it soon became obvious that, because
of the hydrogen sulfide content of the gas, this was not an acceptable
alternative. If the hydrogen sulfide was not removed from the gas a poten-
tial for corrosion of the boiler stack existed.
The capital costs of the equipment needed to remove the hydrogen sulfide
were estimated to be over $100,000.
Providing a Dedicated Boiler
The next alternative was to provide a separate boiler sized to fire at the
rate of'gas production. The boiler manufacturers state that if the boiler
exhaust was kept above 375°C, corrosion would be no problem. Thus, a conven-
tional carbon steel packaged boiler could be used. The cost of the system
was estimated to total $40,000. The estimated annual costs of operating
the system are in Table 11.
TABLE 11. HEAT RECOVERY ANNUAL COSTS
Labor
Utilities
Maintenance and Upkeep
Depreciation - 12 years
Total
No labor figures were assessed since it was assumed the waste treatment
operator could take care of the gas production and collection facilities,
and the stationary engineer would handle the boiler operation.
The value of the energy thus utilized was assumed to be the incremental
cost of the fossil fuel not consumed as a result of using the gas. These
savings are calculated below.
TABLE 12. HEAT RECOVERY SAVINGS
Basis: Coal Cost = $30/ton
Coal Heating Value = 9000 BTU/#
Cost/106 BTU = $1.67/106 BTU
Gas Generation = 278,000 ft3/day
Heating Value = 650 BTU/#
BTU Recovered @ 85% Eff. = 154 x 106 BTU/day
Gross Annual Savings
@ 365 days/year = $94,000
57
-------
The gross annual savings of $94,000 less the annual cost of $9000 gives a
net savings of $85,000 per year. This savings is equivalent to a pretax
return in investment of 213 percent and it gives a pay back of less than
one year.
In the proposed scheme, a 200 HP fire-tube boiler with its own exhaust stack
would be provided in the boiler room. The boiler would be provided with a
combination boiler to allow he firing of fuel oil of gas production dropped
off. (There is no natural gas available for new installation in the proposed
location of the plant.)
The lagoon cover itself would be used for gas storage. If the gas production
rate exceeded the boiler capacity, the gas incinerator would be fired at
intervals to maintain an acceptable balance.
EMISSION CONTROL
The Enforcement Division of EPA, Region IX, was contacted regarding the
emissions from a boiler or an incinerator operating on anaerobic gas.
The following comments were made:
"In the solution of any environmental problem, the necessary
changes and usage of resources create other environmental
stresses. A common case is where a pollutant in one medium
(water) is placed on another medium (land). A similar trade-
off occurs in this proposed solution for control of odors
from anaerobic lagoons. Specifically, burning of the collected
gas will result in the conversion of the highly odorous
compound, H2S, to a non-odorous compound S02- However, many
discharges of S02 to the atmosphere are limited by regulations
to prevent air pollution. At the proposed location of these
facilities in southwestern Arizona, there are regulations
limiting the emission of S02 from existing boilers which use
fossil fuels. The S02 regulations currently in effect for
southeastern Arizona do not constrain S02 emissions from
existing boilers using gases produced from anaerobic digestion.
However, any new or modified facility would be subject to
preconstruction review regulations of the State and local
agencies. Such a project could also be subject to EPA new
source review regulations depending on its size and location.
Since certain portions of southwestern Arizona are not meeting
the National Ambient Air Quality Standards for S02, any major
new source of S02 located in such areas would be subject to
strict emission limits and offset requirements described in
EPA's Interpretive Ruling of December 21, 1976. Furthermore,
the State or local agency may impose requirements more
restrictive than the minimum required by EPA."
From the above statement it can be seen that prior to an investment decision,
the appropriate authorities should be contacted regarding the emissions
from incinerating the anaerobic gases. If local conditions warrant, S02
removal may be required.
58
-------
The technology for scrubbing flue gases of S02 is well established. One
highly effective process would be the use of a bicarbonate scrubber to
effect a 90% to 95% removal of the S02* The calculated S(>2 emissions from
burning the anaerobic gases at a 16 mg/1 H£S concentration would amount to
2.66# S02 per million BTU's. A 90 to 95% S02 removal efficiency would bring
the 802 emissions well within new source limitations required for fossil
fuel boilers.
Discussions with suppliers of scrubbers to remove 2.66# S02 per million BTU's
from a 200 H.P. boiler indicated that an installed capital cost of $70,000
and an annual operating cost of $10,000 could be routinely achieved. Land
disposal of the waste liquor from the scrubber would be an acceptable disposal
technique.
Based on the economic incentive the anaerobic lagoon presents over a mechanical
system their added costs should not change the management decision to adopt
anaerobic lagoons over a completely aerobic mechanical system.
CONCLUSIONS
Odor problems from anaerobic lagoons treating meat packing wastes are the
result of hydrogen sulfide emissions. The hydrogen sulfide escaping to the
atmosphere will be nearly 1% of the sulfate concentration in the wastewater.
Anaerobic lagoons which treat wastewaters containing 100 mg/1 or more of
sulfate need special design and operation for odor control. Design of sub-
merged inlets and outlets, operation to maintain a complete scum cover, chlo-
rination of the raw wastewater prior to pipeline transport when H2S is pro-
duced, and changing to a lower sulfate water supply have been found to reduce
odor emissions.
Anaerobic lagoon treatment is both cost effective and energy saving for
warm concentrated wastewaters. The major drawback of odor emission, when
the wastewater contains high sulfate concentrations, has been controlled
with a flexible membrane cover and a positive gas removal system. The
floating flexible membrane cover has been shown less expensive than rigid
cover systems.
The anaerobic lagoon has been shown economically advantageous over aerobic
treatment for meat packing wastewater even with the added cost of a flexible
cover and gas removal system.
Although incineration of odorous digestor gas is the common disposal method,
these gases can be used economically in a dedicated boiler for the produc-
tion of steam for process uses and to conserve our fossil fuel resources.
This energy conservation system has a one year payout.
59
-------
REFERENCES
1. Bethea, R. M., "Comparison of Hydrogen Sulfide Analysis Techniques."
Journal of Air Pollution Control Association. Vol. 23, No. ,
p. 710 - 713 (1973).
2. Pomeroy, R. D., and H. Cruse. Hydrogen Sylfide Odor Threshold. Jour.
Amer. Water Works Assn. 61(12):677. 1969.
3. Leonardos, G., D. Kendall, and N. Barnard. Odor Threshold Determina-
tions of 53 Odorant Chemicals. Jour. Air Poll. Control Assn. 19(2):91.
1969.
4. American Mutual Insurance Alliance. Handbook of Hazardous Materials.
Technical Guide No. 7, Chicago, Illinois, p. 60 (1974).
5. Lawrence, A. W., McCarty, P. L., and Guerin, F. "The Effect of Sulfide
on Anaerobic Treatment." Proceedings of 10th Industrial Waste Conference,
Purdue University, p. 343 (1964).
6. Gloyna, E. F., and Espino, E. "Sulfide Production in Waste Stabiliza-
tion Ponds." ASCE Journal of Sanitary Engineering Division, Vol. 95,
ND. SA3, p. 607 (1969).
7. Hogstrom, Ulf. Possibilities of Predicting Odor Frequencies in Ambient
Air from Sensory and Chemical Analyses at the Source. University of
Uppsala, Uppsala, Sweden. 1970.
8. Sweeten, John M. Odor Perception and Measurement. Agricultural Engi-
neering. Texas Agricultural Extension Service, College Station.
May 1975.
9. Shirazi, M. A., L. R. Davis, and K. V. Byram. Effects of Ambient
Turbulence on Buoyant Jets Discharged Into a Flowing Environment. EPA-
National Environmental Research Center, Corvallis, OR. Jan. 1973.
10. Minor, J. Ronald, and Ronald W. Stark. Evaluation of Alternative Ap-
proaches to Control of Odors from Feedlots. Idaho Research Foundation
Inc. University of Idaho, Moscow, Idaho. Dec. 1975.
11. Personal communication with Dr. John M. Sweeten, Texas Agricultural
Extension Service, College Station, Texas.
12. Rand, M. B., and Cooper, D. E. "Development and Operation of a Low Cost
Anaerobic Plant for Meat Wastes." Proceedings of Industrial Waste
Conference, Purdue University, p. 613 - 638 (1966).
13. Wells, J. W., Wells, P. B., and Alleman, D. D. "Treatment Capabilities
of an Extended Aeration System Following Anaerobic Lagoons Treating
Meat Packing Waste." Proceedings of Sixth National Symposium on Food
Processing Waste. EPA 600/9-76-224.
60
-------
14. City of Greeley, Colorado. Progress Report No. 2. Corrective Action
Plan for Lone Tree Wastewater Treatment Plant. April 1976.
15. Witherow, J. L. "Small Meatpackers Waste Treatment Systems." Pro-
ceedings of the 28th Industrial Waste Conference. Purdue University,
Lafayette, Indiana (1973).
61
-------
TOMATO CLEANING, WATER RECYCLE
AND MUD DEWATERING *#
by
Walter W. Rose*
INTRODUCTION
With the advent of mechanical harvesting, food processors noted an in-
crease in clods of dirt and smear soil coming in with the raw products.
The initial response by the industry to this situation was to use more water
to insure that the product was adequately cleaned. In recent years there
have been several external factors imposed on the industry which now makes
it economical to reduce the volume of fresh water and to discharge less
pollutants.
With engineering assistance from a consulting firm, Eutek Inc. , the
National Canners Association put together a project that was addressed to
the problem of cleaning tomatoes with less water, the removal of mud from
the dump tank and the development of a water recycle system.
BACKGROUND
Results of past work has been reported at previous symposia (1) (2). The
major efforts during the previous two years were to demonstrate the use of
rotating rubber discs for cleaning and to develop the water recycle system.
Prior to the full scale demonstration of low water cleaning by rubber discs,
pilot studies had been conducted (3). Studies have also looked into the
energy and economic aspects of operating in various modes and results
have been reported (4) (5).
Results up to the 1976 season did show that mechanical energy, in the form
of soft, rotating rubber discs, effectively clean tomatoes with the use of
minimal amounts of water. In addition to the wiping of surface dirt from
the tomatoes, the spinning discs also removed a significant number of
adhering stems. Physical and chemical treatment was applied to the dump
* National Canners Association, Berkeley, California
** Supported in part by the U. S. Environmental Protection Agency under
Grant Number S-803251
62
-------
tank water and a major portion of the water was recycled back to the dump
tank. Soil could be separated from the water and removed from a thickener
tank.
The major objectives of the 1976 season were to evaluate the cleaning of
tomatoes by a machine of different design than that previously used, to
verify past data and to economically dewater mud which had been separated
from the water. Because of a labor strike at the beginning of the process-
ing season and rains which followed, the accomplishment of the first two
objectives is in doubt. The quality of fruit was poor and caused the
processor to alter his method of operation. Rather than operate independent
systems, it was necessary for the processor to split the product into two
flows, part of the tomatoes went through conventional processing and part
went through the demonstration system.
The emphasis of this report will be in reporting on results obtained from
dewatering mud by a horizontal vacuum belt. The 1976 test results did
show that the cleaner was effective in cleaning tomatoes and that minimal
water was required. Because of problems previously mentioned, data
collected in 1976 can not be directly compared with that obtained the pre-
vious two years.
WATER RECYCLE
Figure 1 is a schematic drawing which illustrates the major components of
the water recycle system and the method of processing the mud. As shown
in figure 1, the key process elements are a solids trapping false bottom, an
ejector for solids transport, a screen with a solid hopper, a soil solids
separating swirl concentrator, a gravity clarifier - thickener and a tube
flocculator. The soil solids passed through the false bottom and were
transported by an ejector to the gravity screen. The screen removed
gross solids sue1! as vines, rocks and other debris. After screening the
water flowed by gravity to the swirl concentrator. Grit, sand and heavy
particles in the incoming water were separated and discharged as a fixed
underflow from the swirl concentrator to a gravity thickener* Overflow
from the thickener could be chemically treated in the tube flocculator and
returned. The swirl concentrator returned approximately 80% of flow back
to the dump tank. Overflow from the thickener tank could also be returned
to the dump tank or be discharged to the sewer, the direction of flow was
determined by a level control in the dump tank.
63
-------
SCREEN
at
SWIRL
CONCENTRATOS
UNDERFLOW
GRIT TUBE
RETURN WATER FROM
SORTING BELTS ^
THICKENER &
STORAGE OF SOLIDS
./PRIMARY
PUMP
OVERFLOW TO SEWER
RETURN WATER FROM SORTING
BIN DUME
STANDPIPE
LEVEL CONTROJ,
MAKE-UP WATEl
FALSE BOTTOM *
RANSPORV
EJECTOR
VACUUM DEWATERING
BELT
SLUDGE
CAKE TO FINAL
. DISPOSAL
VACUUM
SOURCE
FIG. 1 FLOW DIAGRAM OF WATER RECYCLE SYSTEM.
-------
VACUUM BELT DEWATERING UNIT
Prior to 1976, sludge from the thickener tank was periodically withdrawn
for disposal onto agricultural land. Data indicated that the solids content
of the sludge was generally 20 to 25% solids. Higher solids in the mud
would decrease the disposal costs. A low cost, vacuum belt dewatering
system was fabricated prior to the 1976 season and evaluated as a method
of increasing the solids content of mud removed from the thickener tank.
Figure 2 is a schematic of the dewatering unit. The major components
are as follows:
1. Sludge receiving hopper with level control.
2. Vacuum belt with variable speed chain drive.
3. Vacuum source and filtrate withdrawal pump.
4. Filter cake removal and belt cleaning.
Sludge from the thickener tank was delivered to the hopper by gravity. The
feed to the hopper was controlled by a level switch inside the hopper. The
level switch was connected to an air operated sludge level control valve,
installed in a line between the thickener tank and the hopper.
Two types of belting material were used in the dewatering study. One belt
was made of polypropylene monofilament, 6. 4 ounces per square yard and
with a porosity of 125 cfm of air at 20 inches of mercury. The second belt
was made of nylon high twist, 8 ounces per square yard and with a porosity
of 40 to 70 cfm at 20 inches of mercury.
A NASH water sealing vacuum pump, with a 2 hp motor turning at 1950 rpm,
was used to maintain vacuum within the vacuum chamber. During a test,
the pump operated at a vacuum of 36 inches of water and pulled approxi -
mately 10 cfm. of air through the belt. Sludge was dewatered as it moved
over the vacuum chamber. The filtrate from the sludge was continuously
removed from the vacuum chamber by a diaphram pump.
The dewatered sludge cake was scraped off the belt after it passed over the
pull drive mechanism. It was collected in a gondola and eventually trans-
ported to a landfill for disposal. The washing of the vacuum belt was by
means of spray nozzles located immediately upstream of the sludge hopper.
The belt was cleaned before receiving a new layer of sludge. Rinse water
was collected and transported by a diaphram pump to the thickener tank.
65
-------
SLUDGE CONTROL VALVE
xCHAW
DRIVE
ySLUDSE CQNTROL
/LEVEL SWITCH
BELT WASH AND VACUUM FILTRATE
PUMP SEAL WATER
FIG.2 ASSEMBLY OF VACUUM BELT DEWATERING UNIT
-------
The pertinent operational variable of the vacuum belt are as follows:
1. Sludge solids loading rate
2. Type of vacuum belt
3. Vacuum applied
4. Belt speed
5. Cake thickness
6. Type and concentration of chemical for sludge conditioning.
An intensive program evaluation of the sludge dewatering unit was con-
ducted between September 9 and 23. During this time period, the mud
removal and the sludge dewatering units were operated and monitored on
a continuous, round the clock basis. The dewatering operation, except
on September 15 and 22, 1976, was evaluated without chemical coagulation/
flocculation. On those two days various concentrations of a polymer
(Calgon Cat-Floe) were added to the tube flocculator.
After several trial and error efforts, a schedule for the withdrawal of
sludge from the thickener tank was established. Sludge was permitted to
thicken in the tank for 4 to 5 hours prior to dewatering. It was found that
the vacuum belt could operate for approximately 1 hour before the feed
from the thickener tank became too dilute for further dewatering. When
this occurred, the feed would be stopped and the mud permitted to accu-
mulate in the thickener tank. After some 4 to 5 hours the dewatering
system would be started up again and run until the feed became dilute.
DATA COLLECTION AND INTERPRETATION
Information and samples were collected which permitted an analysis of the
performance of the vacuum dewatering system. Data was gathered for vari-
ation in belt speed, feed sludge and sludge cake on the belt, sludge cake
and filtrate production rate, total solids of the feed sludge and effect of
chemical coagulation. The results from this evaluation have been inter-
preted in terms of sludge solids loading rate, dewatering efficiency, drying
factor, sludge volume reduction efficiency and solids recovery efficiency.
Data are presented in table 1 for the dewatering of mud without the use of
chemical coagulants. During the test period the belt speed varied from
1. 3 to 3. 6 ft/min. The thickness of the sludge feed varied from 3/16 to
10/16 inches with an average of 7/16 inch; the sludge cake thickness
67
-------
oo
TABLE 1. RESULTS OF TOMATO MUD DEWATERING
WITH VACUUM BELT UNIT
(Without Chemical Coagulation)
Date
9/9/76
9/9/76
9/9/76
9/10/76
9/10/76
9/10/76
9/10/76
9/10/76
9/11/76
9/11/76
9/11/76
9/11/76
9/12/76
9/13/76
9/13/76
9/13/76
9/14/76
9/14/76
9/14/76
9/23/76
9/23/76
Average
Range
Time
4:2Dp
5:00p
6:30p
2:15p
5:10p
6:05p
10:00p
ll:20p
4:30a
10:00a
4:10p
5:05p
l:10a
12:30p
4:10p
9:30p
l:00a
2:00a
10:00a
12:00a
5:10a
Belt
Speed
(ft/min)
3.6
2.2
1.3
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
1.3
2.2
(1.3-
3.6)
Mud Thickness (in)
Feed
4/16
3/16
4/16
4/16
4/16
9/16
8/16
5/16
6/16
7/16
10/16
8/16
9/16
6/16
8/16
8/16
8/16
6/16
8/16
8/16
7/16
7/16
(3/16-
10/16)
Cake
2/16
2/16
3/16
2/16
2/16
6/16
4/16
3/16
4/16
4/16
8/16
4/16
6/16
4/16
4/16
6/16
7/16
4/16
7/16
6/16
5/16
4/16
(2/16-
8/16)
Sludge Loading Cake Production Total Sol ids Cone. (39
Rate
(gph/ft2 )
8.6
3.9
3.1
8.6
8.2
18.9
16.8
10.7
12.8
14.9
21.3
17.0
19.2
12.8
17.0
17.1
17.0
12.8
17.0
6.1
9.1
13.0
(3.1-
21.3)
Rate
(gph/ft2)
4.3
2.6
2.3
4.3
4.2
12.6
8.4
6.4
8.5
8.5
17.0
8.5
12.8
8.5
8.5
12.8
14,9
8.5
14.9
4.6
6.5
8.6
(2.3-
17.0)
Feed
25.7
29.5
25.9
32.8
16.3
14.5
29.9
16.1
12.0
20.5
30.1
20.2
20.4
38.2
31.4
24.9
28.8
27.2
31.3
45.9
33.4
26.4
(12.0-
38.2)
Cake
46.9
42.5
36.0
45.7
28.3
25.8
50.0
26.0
18.5
38.0
38.6
34.6
32.0
54.4
43.1
33.9
36.7
37.2
59.3
55.6
49.5
39. /
(18.5-
59.3)
Filtrate
1.3
1.1
2.0
1.3
1.9
1.6
1.7
1.8
3.1
1.2*
1.4
1.7
3.9
1.2*
1.9
1.6
1.2
1.2
1.8
8.0
5.5
2.2
(1.1-
8.0)
Filtrate
Production Rate
(gph/ft2)
2.9
4.8
5.2
2.0
4.0
4.2
2.3
3.8
4.0
4.0
3.8
2.9
4.6
4.6
2.7
2.9
2.6
3.5
3.0
2.5
5.0
3.6
(2.0-
5.2)
* Estimated value
-------
ranged from 2/16 to 8/16 inches with an average thickness x>f 4/16 inch.
The sludge loading rate varied between 3.1 and 21. 3 gph/ft , with an
average of 13 gph/ft . The total solids concentrations for the feed sludge,
sludge cake and filtrate ranged from 12. 0 to 38. 2, 18. 5 to 59. 3 and 1.1 to
8. 0 percent by weight respectively with an average of 26. 4, 39. 7 and 2. 2
percent by weight.
The rates of sludge loading and sludge cake production were estimated from
the measured feed sludge thickness and the belt speed. As an illustration,
data was taken from September 9 at 4:20 P.M. and calculated as follows:
Feed sludge = 4/16 = 0.021 ft.
Belt speed =3.6 ft/min
Belt width = 9 in,- 0. 75 ft
Effective belt surface area = 3 ft ^
Sludge surface loading area = 0. 75 x 3. 6 = 2. 7 ft /min.
Therefore the sludge loading rate is:
= 2. 7 ft2/min x 0. 021 ft x 7. 48 gal/ft3 x 60 min/hr
3ft2
= 8. 6 gph/ft2
2
Similarly, the sludge cake production was estimated to be 4. 3 gph/ft
since the cake thickness for this particular run was 2/16 inch.
On September 1 5 and 22, the vacuum belt was used to dewater sludge which
had resulted from the use of chemical flocculation of thickener overflow.
Data for this time period is presented in table 2. In comparing the data
with that in table 1, certain observations can be made:
1. A slight increase in the thickness of sludge feed and cake
was noted during the time coagulation was used.
2. There was an increase in the sludge loading rate and in
sludge production when chemicals were used.
3. There was a significant increase in the total solids
concentration of the sludge feed and cake when chemicals
were used.
69
-------
TABLE 2. RESULTS OF TOMATO MUD DEWATERING
WITH VACUUM BELT UN IT
(With Chemical Coagulation)
Date
9/15/76
9/15/76
9/15/76
9/22/76
9/22/76
9/22/76
Average
Range
Time
l:40a
8:40a
4:00p
5:00p
6:15p
8:00p
Belt
Speed
(ft/min)
3.6
3.6
3.6
3.6
2.2
1.3
Sludge
Mud Thickness (in) Loading
Rate
Feed Cake (gph/ft2)
9/16
7/16
6/16
9/16
8/16
8/16
8/16
(6/16-
9/16)
7/16
5/16
5/16
5/16
4/16
6/16
5/16
(4/16-
7/16)
19.2
16.9
14.5
21.8
10.4
6.1
14.8
( 6.1-
21.8)
Cake
Production
Rrte
-------
4. There was also a significant increase in the total solids concen-
tration of the filtrate when adding chemicals and reasons for
this increase are not known.
DISCUSSION OF RESULTS
The important parameters in evaluating a sludge dewatering unit includes:
sludge solids loading rate, cake solids production rate, dewatering effi-
ciency and solids recovery efficiency. The sludge solids loading rate was
estimated to be the sum of rates of solids collected in the filtrate and
solids in the cake. The former were estimated from the filtrate production
rates and the solids concentrations of the filtrate. The rate of solids pro-
duction in the cake was estimate from the sludge cake production and the
solids concentration of the cake. The specific gravity of the cake was
estimated to be 1. 41.
The sludge dewatering efficiency looks at the efficiency of removing water
from sludge and is defines as:
D.E. =
(filtrate water out)
filtrate cake
water out water out
x 100
The drying factor reflects on an increase of solids concentration of the cake
that is produced as opposed to sludge prior to dewatering. This parameter
is defined as the ratio of solids concentration of the cake to that of the
sludge. The sludge volume reduction efficiency, representing the percent
volume reduction in dewatered sludge is an important factor that considers
the volume of cake being produced and disposed of. Efficiency is defined
as:
x 100
cake vol. , filtrate prod.
prod, rate rate
The final parameter that was evaluated was solids recovery efficiency and
is defined as:
t
cake solids , _„
x 100
cake solids + filtrate solids
71
-------
Table 3 presents data for dewatering with the vacuum belt during the time
period when chemical coagulation was not used. Solids loading rates ranged
from 9. 8 to 117. 3 Ibs/hr - ft2, with an average of 38. 9 lbs/hr-ft2. Cake
O
production ranged from 8. 9 to 116. 8 Ibs/hr -ft with an average of 38. 3
Ibs/hr -ft . The dewatering efficiency was between 17 to 70 percent with
an average of 37 percent. The drying factor ranged between 1.21 and
1. 89 with an average of 1. 53. The sludge volume reduction efficiencies
ranged from 17 to 49 percent with an average of 30 percent. The solids
recovery efficiencies ranged from 91 to 99 percent with an average of
97 percent.
The effect of chemical treatment of thickener overflow on sludge dewatering
was investigated for six runs that took place on 9/15 and 9/22. The
chemical used was a cationic polymer (Calgon Cat-Floe). The chemical
dosage for the three runs made on 9/15 was 15 mg/1; and for 9/22 it was
20 mg/1. Results for these conditions and the ranges are given in table 4.
In comparing the performance of the vacuum belt as shown in table 3 and
4, it was noted that a significantly higher sludge solids loading rate and
cake solids production rate were obtained with chemical coagulation than
without. The average solids loading rate with chemical usage was 76. 2
lbs/hr-ft2 as opposed to 38. 9 lbs/hr-ft2 without chemical use. Similarly,
the average solids production rate with chemical addition was 75. 2 Ib/hr-
ft , in contrast to 38. 3 Ib/hr-ft without chemical.
It has been estimated that the unit cost for the vacuum belt dewatering unit
is $2 per ton of dry solids. This is in contrast to the $12 to $29/ton of dry
solids for the dewatering of sewage sludge by conventional rotary drum
vacuum filters, filter presses or centrifuges. The unit cost of $2/ton is
estimated on the following assumptions:
1. Tomato production rate - 120 tons/hr.
2. Solids production rate - 15 Ibs soil/ton tomatoes.
3. Period of operation - 60 days.
4. Initial capital cost - $5, 000 with a 5 year life and straight
line depreciation.
5. Manpower - 2 man hours/day at $10/hr.
6. Total power - 3 HP.
72
-------
TABLE 3. PERFORMANCE EVALUATION OF TOMATO MUD
DEWATERING WITH VACUUM BELT UNIT
(Without Chemical Coagulation)
Date
9/9/76
9/9/76
9/9/76
9/10/76
9/10/76
9/10/76
9/10/76
9/10/76
9/11/76
9/11/76
9/11/76
9/11/76
9/12/76
9/13/76
9/13/76
9/13/76
9/14/76
9/14/76
9/14/76
9/23/76
9/23/76
Average
Range
Time
4:20p
5:00p
6:30p
2:15p
5:10p
6:05p
10:00p
ll:20p
4:30a
10:00a
4:10p
5:05p
l:10a
12:30p
4:10p
9:30p
l:00a
2:00a
IftOOa
12:00p
5:10a
Sludge Solids
Loading Rate
(Ib/hr/ft2)
24.0
12.9
9.8
23.1
12.6
32.9
51.2
17.2
15.8
35.7
72.4
31.7
44.2
58.8
42.0
46.3
59.4
34.7
117.3
34.3
44.1
38.9
(9.8-
117.3)
Cake Solids
Production Rate
(Ib/hr/ft2)
23.7
12.5
8.9
22.9
12.0
32.3
50.9
16.6
14.8
35.3
72.0
31.3
42.7
58.3
41.6
45.9
59.1
34.3
116.8
32.6
38.8
38.3
(89-
116.8)
Dewatering
Eff iciency
(%)
47
70
70
38
52
27
27
40
33
36
20
29
29
44
29
27
17
36
17
44
50
37
(17-
70)
Sludge Volume Solids
Reduction Recovery
Drying Factor
1.82
1.44
1.39
1.39
1.74
1.78
1.67
1.61
1.54
1.85
1.28
1.71
1.57
1.42
1.37
1.36
1.27
1.37
1.89
1.21
1.48
1.53
(1.21-
1.89)
(%)
40
35
31
32
49
25
21
37
32
32
18
25
26
35
26
18
15
29
17
35
43
30
(17
49)
(%)
79
97
91
99
95
98
99
97
93
99
99
99
97
99
99
99
99
99
99
95
94
97
(91-
99)
-------
TABLE 4. PERFORMANCE EVALUATION OF TOMATO MUD
DEWATERING WITH VACUUM BELT UNIT
(With Chemical Coagulation)
Date
9/15/76
9/15/76
9/15/76
9/22/76
9/22/76
9/22/76
Average
Range
Time
l:40a
8:40a
4:00p
5:00p
6:15p
8:00p
Sludge Solids
Loading Rate
(Ib/hr/ft2)
108.4
86.5
93.6
92.9
40.5
35.4
76.2
( 35.4-
108.4)
Cake Solids
Production Rate
(Ib/hr/ft2)
107.4
86.1
93.1
92.1
38,2
34.5
75.2
( 34.5-
107.4)
Dewatering
Efficiency
(%)
28
26
25
31
62
44
36
(26-
62)
Drying Factor
1.36
1.56
1.34
1.52
1.27
1.45
1.42
(1.27-
1.56)
Sludge Volume
Reduction
(%)
21
20
18
22
51
35
28
(18-
51)
Solids
Recovery
(%)
99
99
99
99
94
97
98
(94-
99)
-------
Daring the course of this project the newly developed vacuum belt dewater-
ing unit has only been evaluated on the dewatering of sludge from tomato
dump tank water. Its performance on sewage sludge is unknown and makes
a comparison with other existing sludge dewatering devices rather difficult.
What has been evaluated appears to be an efficient and cost effective device
for dewatering mud.
REFERENCES
1. Rose, W. W., Katsuyama, A. M., and Wilson, G. E. Tomato
cleaning and water recycle. Proceedings of the Sixth National
Symposium on Food Processing Wastes EPA 600/2-76-224 (1975)
2. Wilson, G. E. , Rose, W. W. , and Huang, J. Y. C. Tomato flume
water recycle with off-line mud removal. Proceedings of the
Seventh National Symposium on Food Processing Wastes EPA
600/2-76-304(1976)
3. Krochta, J. M., Graham, R. P., and Rose, W. W. Cleaning of
tomatoes using rotating rubber discs. Food Technology 28(12):26
(1974)
4. Carroad, P. A., Krochta, J. M. , and Rose, W. W. Water/energy
trade-off in cleaning tomatoes. Food Technology 30(3):24 (1976)
5. Carroad, P. A. , and Rose, W. W. Water recycle improves disc
cleaning of tomatoes. Food Technology 31(3):92 (1977)
75
-------
REMOVAL OF SUSPENDED SOLIDS AND ALGAE
FROM AEROBIC LAGOON EFFLUENT
TO MEET PROPOSED 1983 DISCHARGE STANDARDS TO STREAMS
by
Ernest R. Ramirez*, D. L. Johnson**, and T. E. Elliott**
ABSTRACT
A new technology has been applied to the removal of algae from aerobic
lagoons. This involves a two-step process which employs electrocoagulation
together with a specially designed dissolved air flotation basin. The
performance of this system is very effective in removal of algae from
aerobic lagoon effluent.
INTRODUCTION
A Swift Hog Packinghouse Plant located in Moultrie, Georgia processes
approximately 2,000 hogs per day. These are dressed and cut in a one-shift
operation. Water used to carry out this operation is approximately
650,000 gallons per day. Following good in-plant waste control practices
plus extensive mechanical pretreatment, this plant employs an anaerobic
lagoon with an area of approximately 1-1/2 acres and a depth of 14 feet.
The anaerobic lagoon effluent is directed to a 19-acre aerobic lagoon.
Details on construction and design factors of both the anaerobic and
aerobic lagoon have been described in the literature (1).
In spite of the anaerobic and aerobic lagoons, the plant determined it
would not be able to meet either 1977 nor 1983 effluent limitations with
the original treatment system. Effluent from the aerobic lagoon is
discharged to Okapilco Creek.
A fundamental problem with the lagoon system at Moultrie, as in other
systems of its type, is the growth of algae in the aerobic lagoon.
The problem of algae in aerobic lagoons has been studied elsewhere and is
discussed in (2) (4). Basically, algae can be described as hydrophilic bio-
colloids with apparent negative surface charges. Oftentimes, algae are
* Swift & Company, Research & Development Center, Oak Brook, Illinois 60521
** Swift Fresh Meats Company, 115 West Jackson Blvd., Chicago, Illinois 60604
76
-------
particulates 3 to 15 microns in size and their specific gravity is approx-
imately that of the water. Destabilization of algae suspensions has been
accomplished with lime, alum, magnesium ions, ferric sulfate, and many
synthetic polyelectrolytes (3) (5). pH is an important factor with regard
to any chemical treatment for removal of algae from water.
ALTERNATE SOLUTIONS TO MEET EFFLUENT LIMITATIONS
The solution to the problem was approached from many directions. These
included the following: (a) modified in-plant operations and improved
pretreatment to diminish the problem, (b) consider the possibility of
working with the municipal plant for treatment of all Swift wastewater,
(c) use of aerobic lagoon effluent as an irrigation source, (d) employ
algae destruct chambers of the Chem Pure Inc. type, (e) use of dissolved
air flotation as described in the literature, (f) consider algae-eating
fish in the aerobic lagoon, (g) use of extended aeration and sand filter
tertiary treatment, (h) use services of algae harvesting facilities as
described by the WRAP System, (i) physical-chemical treatment using
LectroClear process, (j) Sweco concentrate, and (k) Biological Water
Purification reed sand filter system.
Of the above alternate approaches to the wastewater treatment problem
at Moultrie, our Engineering Department settled on the following three
combined with improved pretreatment as being the more efficient and
reliable, when considering the strict standards of the 1983 Effluent
Limitations: (a) extended aeration and sand filter tertiary treatment,
(b) water reclamation by algae harvesting [WRAP System], (c) physical-
chemical treatment with LectroClear.
COST EVALUATION OF TREATMENT ALTERNATIVES
The three selected alternatives mentioned above were then fully evaluated
in terms of Capital Equipment and Cost and Operating and Maintenance
Cost. A cash flow of the three processes indicated the physical-chemical
treatment, using the LectroClear principle, was the most economical. At
the completion of this cost analysis study, plans were then laid for the
design and installation of a LectroClear System capable of meeting proposed
1983 effluent limitations. These limitations, together with the limitations
for 1977, are given in Tables 1 and 2.
77
-------
TABLE 1. MOULTRIE - EFFLUENT LIMITATIONS
(WATER QUALITY LIMITED STREAM)
JULY 1, 1977 - TUNE 30. 1979
Kg/Day (Lbs./Day)
Daily Daily
Avg. Max.
Effluent
BOD5 *
TSS **
Oil & Grease ***
Fecal Coliform
Ammonia
(Nitrogen)****
pH
Settleable Solids
* Based on stream water quality model.
** Based on 1977 Meat Packing Guidelines
*** Based on 1977 Meat Packing Guidelines
****Based on 1983 Meat Packing Guidelines
25 (54)
74 (163)
24 (52)
10 (22)
50 (108)
148 (326)
48 (104)
20 (44)
Basis 0.65 Mgd
Daily Daily
Avg. Max.
Other Units
10
30
10
20 ppm
60 ppm
20 ppm
Max. 400 org./
100 ml.
8 ppm
6.0 - 9.0
Not to exceed
0.1 ml/1
max. Lwk 654 M#Lwk x .25#/M#Lwk.
max. Lwk 654 M#Lwk x .08#/M#Lwk.
(4 ppm) (8.345) f-650 Mgd.)
TABLE 2. MOULTRIE - PROPOSED EFFLUENT LIMITATIONS
(WATER QUALITY LIMITED STREAM)
JULY 1, 1983
Effluent
Kg /Day (Lbs./Day)
Daily Daily
Avg. Max.
Basis 0.65 Mgd
Daily Daily
Avg . Max .
Other Units
BOD5
TSS
Oil & Grease
Ammonia
(Nitrogen)
Fecal Coliform
PH
Settleable Solids
Dissolved Oxygen
26
46
22
52
92
44
5-10 ppm
9-18 ppm
10 mg/1
4-8 ppm
Max. 400 Org/
100ml
6.0 - 9.0
Not to exceed
0.1 ml/1
6 ppm.
78
-------
DECISION TO GO TO ELECTROCOAGULATION-DISSOLVED AIR FLOTATION
The plant was especially interested in selecting a low Capital Equipment
figure together with a low Operating Cost figure. Stipulations put forth
by the plant and included in this design were the following: (a) system
should have the capability of the aerobic lagoon effluent in a single-shift
(8 hours) five day operation. This stipulation automatically increased
the capacity of the unit by threefold, (b) space was not a premium item
under the condition of this project, (c) labor allotted to the wastewater
treatment system would be held to a minimum.
Annual Operating and Maintenance Cost of the LectroClear installation
based on a design of 1,500 gallons per minute is an estimated maximum
$75,000 per year. Annual Fixed Cost and Depreciation Cost are an esti-
mated $25,000 per year based on a straight line 15 year depreciation.
Total Annual Operating Costs and Fixed and Depreciation Costs are an
estimated maximum $100,000 per year for the specific plant at Moultrie,
Georgia.
DESIGN FACTORS
The design of the LectroClear unit installed in Moultrie, Georgia was
based on the following parameters: (a) a single design would be made
which would meet both 1977 and 1983 Effluent Limitations. The primary
difference in the quality of the treated wastewater would be the amount
of metal coagulant employed [alum].
Beaker Scale tests on the removal of algae from Moultrie wastewater
showed the following: (a) optimum reults were achieved when the com-
bination of electrocoagulation and dissolved air flotation were employed,
(b) energy input in the electrocoagulation cell would require not more
than 2 ampere-minutes per gallon of wastewater treated, (c) 50% dissolved
air recirculation using pressures between 50 to 75 psig would be adequate,
(d) to meet the 1977 Standards, approximately 80 -150 ppm of alum would
be required, while to meet the proposed 1983 Limitations approximately
150-250 ppm of c.lum would be used, (e) size of the basin would be 64- feet
long, 15-feet wide, and 5-feet deep, (f) dwell time in the electrocoagu-
lation cell would be 2-1/2 minutes when operated at 1,500 gpm. Details
on mechanisms and design of electrocoagulation cells have been reported
elsewhere (6) (7) (8) (9) (10). Photographs of the electrocoagulation
cell are shown in Figures 1 and 2. Photographs of the overall installation,
showing both flotation basin and electrocoagulation cell, are shown in
Figures 3 and 4.
79
-------
'
Fig. 1 Electrocoagulation Cell.
Bottom section contains
electrodes.
Fig. 2 Inside view of bottom of
Electrocoagulation Cell.
Note close spacing between
electrodes.
Rectifier, powering the electrocoagulation cell, is capable of 24 Volts DC
and 3,000 Amperes. Pump employed for the 50% dissolved air recycle
employs 50 Hp.
RESULTS
Installation of all basic hardware was completed in January 1976.
Several one-week trial runs were carried out to evaluate capabilities of
the installation and also to debug the operation.
Performance of the LectroClear installation during various runs is shown
in Table 3.
-------
—. - -
Fig. 3. Flotation Basin. Electro-
coagulation Cell is at the
left. Note skimming storage
in front of basin.
Fig. 4. Rear View of Flotation
Basin. Tank is pressure
chamber for DAF.
TABLE 3. WATER ANALYSES OF MOULTRIE-LECTROCLEAR TESTS
Sample
Lagoon
LC Effluent*
Lagoon
LC Effluent*
Lagoon
LC Effluent*
Lagoon
LC Effluent*
Lagoon
LC Effluent*
Lagoon
LC Effluent*
Date**
9/20/76
9/20/76
9/21/76
9/21/76
9/22/76
9/22/76
9/23/76
9/23/76
9/24/76
9/24/76
1/4/77
1/4/77
BOD
(mg/1)
8
!:
21.6
3.5
17.3
2.7
20
5
!
5
41
5
TSS
(mg/D
53
5.5
64
4.5
55
2.0
44
:.
47
1
56
9
TKN
pH
(mg/1)
33.
26.
47.
12.
24.
9.
23.
13.
24.
14.
20.
13.
0
5
9
3
6
;
0
0
0
0
0
0
7.
6.
.-
__
-
:.
6.
1.
7.
^
.
:
'\
-
-
>.
':
'
'
FOG Fecal
(mg/1) Coliform (mg/1)
___
5.1
5.1
4.0
2.8
0
* LectroClear Effluent - Treatment is 150 ppm Alum, 2 ppm Anionic Polymer
1.2 Ampere-minutes in Electrocoagulation Cell, 50% recycled DAF,
all samples are daily composite tests.
-------
Solids content of the skimmings are given in Table 4.
TABLE 4. MOULTRIE -
ALGAE SLUDGE ANALYSES LECTROCLEAR BASIN
(WET BASIS)
Date
3/31/76
3/31/76
3/31/76
3/31/76
3/31/76
4/1/76
4/2/76
8/21/76
9/21/76
Total Solids, %
3.9
4.0
4.2
3.6
3.9
4.9
7.8
5.4
7.0
TKN, %
___
0.245
0.25
0.27
0.28
0.28
Equiv. Protein, % Aluminum, %
— _ _ _
_L « Do ""•"" """
1.56 0.44
1.69
The analyses of both as generated skimmings and skimmings on a dry basis are
given in Table 5.
TABLE 5. MOULTRIE SKIMMINGS ANALYSIS (DRY BASIS)
Sample
Comp. 1-5
Date
3/31/76
Phosphorus, %
1.71
TKN, %
7.88
Equiv. Protein, %
49»4
K20
0.70
It is anticipated that approximately 3,000 gallons per day of algae skim-
mings will be generated and removed from the LectroClear basin.
Due to the fertilizer value of the skimmings, the local fanners were
interested in obtaining possession of the skimmings. In the final
analyses, a contract will be made with a local farmer whereby he would
guarantee the removal of skimmings on a daily basis. Figure 5 shows
farmer removing the first few runs of skimmings at the LectroClear
wastewater treatment facility.
82
-------
Fig. 5. Farmer removes skimmings from skimmings holding tank.
DISCUSSION
The results show that removal of algae at the Moultrie plant is a
proportional function of alum dosage used. To meet proposed 1983 dis-
charge limitations to the stream, alum addition of about 150-250 ppm
will be required. The 1977 discharge limitations can be met with
approximately 80-150 ppm of alum addition.
The data clearly demonstrate that the combination of electrocoagulation
and recycle dissolved air flotation technologies can be effectively used
to remove suspended solids from wastewaters. The unique sequential two-
step operation employing electrocoagulation first, provides an ideal
approach to destabilizing the hydrophilic biocolloid in the first-step,
while recycle dissolved air flotation is used to remove the last traces
of suspended material from the wastewater.
This particular installation was used to separate algae from wastewater.
The concept, however, can be effectively used for handling any type of
wastewater where suspended material, coagulated material (blood or
albumen), or precipitated material (heavy metals) are present. It
follows that this technology can be applied to all wastewaters where
suspended, dispersed or emulsified foreign pollutants are present.
In summary, a new concept has been established for utilizing electro-
coagulation and dissolved air flotation in a sequential two-step
operation. The effectiveness of this technology surpasses the per-
formance of processes heretofore industrially employed.
83
-------
REFERENCES
1. Sollo, F. W. 15th Industrial Waste Conference, Purdue University,
Lafayette, Ind. (1960).
2. Friedman, A. A., Peaks, D. A., and Nichols, R. L. Algae separation
from oxidation pond effluents. Journal Water Poll. Control Fed.,
Pg. Ill (1977).
3. Caldwell, D. H., et al. Upgrading lagoons. Transfer Seminar Program,
USEPA, Washington, D. C. (August 1973).
4. Bare, W. F. Ranee, et al. Algae removal rising dissolved air flotation.
Journal Water Pollution Control Fed., 47, 153 (1975).
5. Folkman, Yais and Wachs, Alberta M. Removal of algae from stabiliza-
tion pond effluents by lime treatment.
6. Ramirez, E. R. Electrocoagulation of meat processing wastewater.
WWEMA Industrial Conference (April 3, 1974).
7. Beck, E. C., Giannini, A. P., and Ramirez, E. R. Electrocoagulation
clarifies food wastewater. Food Technology, Vol. 28, No. 2, Pg. 18,
(1974).
8. Ramirez, Ernest R. Electrocoagulation clarifies food wastewater.
Deeds & Data, WPCF (April 1975).
9. Ramirez, E. R., Johnson, D. L. and Clemens, 0. A. Direct Comparison
in physiochemical treatment of packinghouse wastewater between dis-
solved air and electroflotation. 31st Annual Purdue Industrial Waste
Conference, West Lafayette, Ind. (May 1976).
10. Ramirez, E. R. and Clemens, 0. A. Electrocoagulation techniques for
primary treatment of several different types of wastewater. 49th
Annual Conference, W.P.C.F., Minneapolis, Minn. (October 4, 1976).
84
-------
EFFLUENT GENERATION, ENERGY USE AND COST OF BLANCHING
by
John L. Bomben*
INTRODUCTION
The wastewater produced by the blanching of vegetables for freezing or can-
ning is a significant fraction of the total wasteload of the vegetable
processing industry(l). Since the beginning of this industry, the basic
design of blanchers has received little attention, but because of national
concerns about wastewater and energy, changes in the design of conventional
blanching systems are now being tested and evaluated.
This paper describes the characteristics of conventional blanchers and new
blancher designs, the effluent generation and energy use of blanchers, and
the cost of blanching with four different blanchers [water, steam (hydrosta-
tic and vibratory spiral), and hot gas].
Material and Heat Balances
Figure 1 shows a material balance for blanching and cooling vegetables.
(Table 1 lists the definitions of the symbols used in this paper). Total
Solids of the vegetables is used as the unit for the material balance. In
the effluent, Total Solids can be expressed as the equivalent Biological
Oxygen Demand (BOD), Chemical Oxygen Demand (COD) or Total Organic Carbon
(TOC) and Suspended Solids (SS), which are all common units for measuring the
organic load of an effluent. The initial weight of raw vegetables changes
as the vegetables are blanched and cooled because of changes in water and
solids content; the yields (Y and Y ) represent these changes. Reducing
the wasteload of blanching ana cooling translates into reducing the hydraulic
load (E and E ) and the organic load (TS and TS ) of the effluents. If
air is used to cool the vegetables, then E = 0, and the cooling effluent is
eliminated, but since much of the cooling is provided by evaporation of water,
the yield after air cooling will be less than that after water cooling.
Frozen vegetables are sold by weight, and thus the reduction in yield is a
product loss. Bomben et al., (5) have reported on material balances measured
with seven different vegetables using either air or water cooling.
Figure 2 shows a heat balance for blanching and cooling. Using C = 1 kcal/
kg°C (1 BTU/lb°F) AH = 539 kcal/kg (970 BTU/lb) and QL = 0, the theoretical
steam required to heat the vegetables to an enzyme inactivation temperature
can be calculated to be 134 kg of steam/kkg of raw vegetables (268 Ib steam/
ton raw vegetables). The magnitude of Q determines the efficiency of an
actual blancher; the smaller it is, the Higher the efficiency.
*USDA, ARS, Western Regional Research Center, Berkeley, California
85
-------
RAW
VEGETABLES
WV, (TS)R
\
STEAM
OR
, HOT WATER
BLANCHER
i
BLANCHED
BLANCHER
EFFLUENT
EB, (TS)EB
VEGETABLES
YBWV, (TS)B
\
COLD WATER
AND/OR
, AMBIENT AIR
COOLER
i
COOLED
VEGETABLES
COOLER
EFFLUENT
EC, (TS)EC
, (TS)C
WV (TS)R = YCYBWV(TS)C + EB(TS)EB + EC(TS)EC
BLANCHER COOLER
BOD BOD
& &
SS SS
Figure 1. Material balance for blanching and cooling.
-------
00
Wy
RAW
VEGETABLES
M = 16°C
OH
\
OH "Q
t
BLANCHEF
L
YB wv
BLANCHED
VEGETABLES
t2 = 88°C
= WVC (t2-ti)+QL
IF QL = O & C =
OH _ 131 kg
1 keal/kg°C :
i
ww
•«,=
1
COOLER
\
, ww
•*.
16°C
YCWV
COOLED
VEGETABLES
t3 = 27°C
= 27°C
STEAM
Wy kkg RAW VEGET.
Figure 2. Heat balance for blanching and cooling.
-------
TABLE 1. NOMENCLATURE
BOD = Biological Oxygen Demand, kg/kkg
C = heat capacity of vegetables or water, kcal/kg°C
E = hydraulic wasteload of blancher, 1/kkg of raw vegetables
E-, = hydraulic wasteload of cooler, 1/kkg of raw vegetables
AH = heat of vaporization for water, kcal/kg
Q = heat added to blancher, kcal/kg of raw vegetable
QT = heat lost from blancher, kcal/kg of raw vegetable
SS = Suspended Solids, kg/kkg
(TS) = Total Solids in blanched vegetables, %
(TS) = Total Solids in cooled vegetable, %
- = Total Solids in blancher effluent, %
= Total Solids in cooler effluent, %
(TS) = Total Solids in raw vegetable, %
t, = temperature of raw vegetables,°C
t_ = temperature of blanched vegetables, C
t, * temperature of cooled vegetables, C
tw. = temperature of inlet cooling water
tw = temperature of outlet cooling water
W = feed rate of raw vegetables, kkg/hr
W = water evaporated from vegetable, kkg/hr
Y v=Pyield of blanched vegetables, %
Y., = yield of cooled vegetables, %
\s
On the basis of a heat balance using the temperatures shown in Figure 2,
5500 1 of water/ kkg of raw vegetables (1320 gal/ton) are required for cool-
ing vegetables after blanching. For air cooling the heat balance is more
complicated because both heat of evaporation and sensible heat are trans-
ferred. If heat were transferred solely by evaporation, then for the temper-
atures in Figure 2, C = 1 kcal/kg and AH = 556 kcal/kg(1000 BTU/lb):
Wevap , YBC (t - t3) = Q
Wv AH + C (t2 - t3)
Since Y usually equals approximately 95% (5), air cooling gives approxi-
mately a 10% reduction in the yield of vegetables. This loss can be
counteracted by spraying the vegetables with water or blancher condensate,
but in practice the weight loss and water addition do not balance, and
there is a reduction of yield and some production of effluent(5).
Conventional Water and Steam Blanchers
Although conventional blanchers can be found in many forms, they can be put
into two basic categories water and steam blanchers. Table 2 shows the
characteristics of wastewater from conventional steam and water blanchers.
Blanchers are operated at varying conditions, and as a result, the charac-
teristics of the wastewater can be quite different at different times in the
-------
TABLE 2. WASTELOAD AND ENERGY USE IN CONVENTIONAL BLANCHING '
Hydraulic Load BOD SS Steam Efficiency
(1/kkg)(klTkkg) (kiJkkg) (%)
Hot Water 60
Snap Beans 124 0.69 0.13
Lima Beans 821 0.65
Peas 384 3.0
Steam
Snap Beans 125 0.55 0.02
Lima Beans 238 3.5
Peas 313 4.3
(1) Data taken from Rails & Mercer(4), Lund(5), Lazar & Rasmussen(6), and
Bomben(7).
same plant and from plant to plant. Generally, water blanchers produce a
higher hydraulic load, but their organic load can be larger or smaller than
that of steam blanchers. The amount of make-up water added to a water blan-
cher is not a well-controlled variable and is probably the cause of the large
difference in wasteloads reported in the literature. One plant operates a
water blancher with no overflow(8), but since in that particular case
(artichokes) citric acid is added to the blancher water, this has to be
considered a special case, whose applicability to other blanched vegetables
has not been investigated.
Data on the steam consumed by conventional blanchers are very limited, and
typical values are shown in Table 2, where efficiencies are based on the
theoretical steam consumption of 134 kg of steam/kkg of raw vegetables.
Whereas the steam efficiency for the steam blancher was a measured value(7),
the one for the water blancher was estimated(6). Generally, it appears
reasonable to assume that water blanchers consume less steam than steam
blanchers, but measurements of the actual efficiency of a water blancher
are not available, or at least not well-known. The low efficiency of a
steam blancher is attributable to the large losses of steam at the feed
and discharge ends and the large uninsulated surface area.
New Blancher Designs
The Vibratory Spiral Blancher-Cooler is a design which departs markedly from
that of conventional steam blanchers(9). Vibratory spiral conveyors are used
to reduce the size of the steam blancher as compared to a water blancher,
thereby reducing floor space and reducing surface area for heat loss. In
89
-------
addition, the spiral conveyor gives a simple means of sealing the feed and
discharge of the blancher with vegetables thereby reducing this large heat
loss associated with steam blanching. In addition to the blancher, this
design also uses a vibratory spiral conveyor as an air cooler, where conden-
sate from the blancher is sprayed on the vegetables as they are cooled.
Thus, the only effluent for both blanching and cooling is the unevaporated,
unabsorbed liquid leaving the cooler. This design has been tested in a
large-scale pilot plant(7 and 9), and Table 3 shows the waste load and the
steam efficiency obtained with that unit.
TABLE 3. WASTELOAD AND STEAM EFFICIENCY OF VIBRATORY BLANCHER-COOLER
Hydraulic BOD SS
Load
(1/kkg) (kg/kkg) (kg/kkg)
Snap Beans 27.0 0.53 0.084
Lima Beans 27.9 0.90 0.54
Brussels Sprouts 15 0.43 0.073
Cauliflower <3
Broccoli 11 0.25 0.091
Steam Efficiency = 85% (Theoretical steam requirement = 134 kg/kkg of
vegetables)
Hot-Gas Blanching uses the products of combustion along with steam as a heat
transfer medium(4). A belt conveyor moves the vegetable through the hot
gases and steam. This concept has been tested to the pilot plant stage, and
Table 4 shows the waste load and energy requirements obtained with the pilot
plant.
The Hydrostatic Steam Blancher uses water to seal the feed and discharge of
a conventional steam blancher, thereby reducing the escape of steam from the
blancher(lO). Some of these blanchers are operating in freezing plants in
California, and are reported to use 0.5 Ib of steam per Ib of vegetable(ll).
No data are available on the wasteload produced by this system, but it proba-
bly is approximately the same as that of a conventional steam blancher and
flume as reported by Bomben et al.(5) and listed in Table 5.
90
-------
TABLE 4. WASTELOAD AND STEAM EFFICIENCY OF HOT-GAS BLANCHING
Snap Beans
Corn-on-cob
Beets
Spinach
Peas
Hydraulic
Load
(1/kkg)
0.06
58
93
0.66
14
BOD
(kg/kkg)
<0.01
0.86
0.63
0.01
0.12
Steatn efficiency = 56% (includes natural gas and electrical power for
blower and is based on a theoretical steam consump-
tion = 134 kg/kkg).
TABLE 5. WASTELOAD OF STEAM BLANCHING INCLUDING WATER COOLING
Snap Beans
Lima Beans
Peas
Steam Blanching Effluent
Hydraulic , »x
Load WD
(1/kkg) (kgTkkg)
150 0.92
113 1.1
191 2.7
Water Cooling Effluent
Hydraulic ,„*
Load BODU'
(TTkkg) (kgTkkg)
4930 1.6
4960 3.4
4960 2.9
(1) Data taken from Bomben, et. al.(5)
(2) Chemical Oxygen Demand was multiplied by 0.6 to get BOD(7)
Comparison__of Waste Water and Energy Use
Table 6 compare's the wastewater produced and the energy used for the blan-
ching of snap beans by different methods. Wastewater characteristics are
reported as hydraulic wasteload and organic wasteload (BOD & SS) and steam
consumptions are reported as efficiency with respect to the theoretical value
of 134 kg/kkg. In the case of Hot-Gas Blanching, the energy of the natural
gas and the circulating blower are included in the calculation of the effi-
ciency.
Hot-Gas Blanching produces the lowest wasteload. Its energy efficiency is
an improvement over steam blanching, and it approaches that of water blan-
ching. Despite its low wasteload, Hot-Gas Blanching is at a disadvantage
91
-------
TABLE 6. WASTELOAD AND STEAM EFFICIENCY OF DIFFERENT METHODS OF BLANCHING
SNAP BEANS
Conventional Water
Conventional Steam
Hot-Gas
Hydrostatic Steam
Vibratory Spiral
Hydraulic
Load
(1/kkg)
124
125
0.06
—
27
BOD
(kg/kkg)
0.69
0.55
<0.01
—
0.53
SS
(kg/kkg)
0.13
0.02
<0.01
—
0.084
Steam
Efficiency
(Z)
60
5
56
27
85
Blancher-Cooler
because it depends on increasingly scarce natural gas or liquified petroleum
gas.
Since its effluent is the same as a conventional steam blancher and a cooling
flume, the Hydrostatic Steam Blancher makes no reduction in wasteload. Its
steam efficiency is a considerable improvement over that of a conventional
steam blancher, but it is less than that of a conventional water blancher.
The wasteload for the Vibratory Spiral Blancher-Cooler includes both blan-
ching and cooling, whereas the other wasteloads shown in Table 6 do not
include cooling. In preparing vegetables for freezing, flume cooling is
generally used after blanching, and flume cooling has a larger hydraulic
wasteload than blanching(5). The Vibratory Spiral Blancher-Cooler reduces
both of these wasteloads to low levels. If only the Vibratory Spiral Blan-
cher is used, as in canning, then its wasteload would be about the same as
that of conventional steam blanching. The Vibratory Spiral Blancher-Cooler
has the highest steam efficiency of the blanchers shown in Table 5, thus
demonstrating the effectiveness of its seals against steam leaks and its
double wall insulated construction.
COST ANALYSIS
Table 7 shows the capital cost of four blanchers. Since conventional steam
blanchers are usually custom-fabricated, purchase costs could not be accu-
rately obtained; therefore, it was decided not to include this type of
blancher in the cost analysis. Equipment purchase costs are based on
equipment manufacturers' price quotations for a capacity of 4.5 kkg/hr
(5 tons/hr) of snap beans (2.0 minute blanching time). Other items in esti-
mating direct fixed capital are taken as percentages of equipment purchase
cost(12). Floor space is valued at $270/m ($25/ft ). The costs of the
water blancher (reel type) and Hot-Gas Blancher include $5,000 for a flume
cooler. The Vibratory Spiral Blancher-Cooler and the Hydrostatic Steam
Blancher have the cooler as an integral part of the blancher.
92
-------
TABLE 7. CAPITAL INVESTMENT FOR BLANCHERS AND COOLERS
1.
2.
3.
4.
5.
Vibratory
Spiral
Blancher-
Cooler
Equipment purchase $107,587
cost
Delivery 5,379
Installation 21,517
Floor space 4,050
Indirect costs 34,633
$173,166
1 . Equip-
ment Mfg.
2. 5% of 1
3. 20% of 1
4. 15 ra at
$270/m
5. 25% (1+2+
3+4)
Water
Blancher
$16,373
819
3,275
1,719
5,546
$27,732
1 . Equip-
ment Mfg.
+ $5,000
for flume
2. 5% of 1
3. 20% of; 1
4. 6.4 m ^at
$270/m
5. 25% (1+2+
3+4)
Hydrostatic
Steam
Blancher
$ 87,446
4,372
17,489
9,000
29,577
$147,883
1 . Equip-
ment Mfg.
2. 5% of 1
3. 20% of 1
4. 33 m at
$270/m
5. 25% (1+2+
3+4)
Hot-Gas
Blancher
$126,900
6,345
25,380
25,022
45,873
$229,520
1. Rails &
Mercer(4)
correc-
ted by
Marshall
and
Stevens
equip-
ment
cost
index(12)
+ $5,000
for
flume
2. 5% of 1
3. 20% of 1
4. 93 m2 at
$270/m
5. 25% (1+
+2+3+4)
Table 8 shows the labor, fixed and variable costs for the four blanchers, and
it shows the basis used for calculating these costs. Since the open mesh
conveyor belts used in the Hydrostatic and Hot-Gas Blanchers probably require
more maintenance than the conveying systems used in the Vibratory Spiral or
water blancher, higher maintenance costs were used for the former. There are
no data available for the wasteload of Hydrostatic Steam Blanchers; there-
fore, its wastewater cost was calculated from the wasteload of conventional
steam blanching with flume cooling (Table 5). The water use and wasteload
for flume cooling were added to those of water blanching and Hot-Gas Blan-
ching to get an overall cost of water and wastewater for these systems.
-------
TABLE 8. COST OF BLANCHING AND COOLING FOR FREEZING ($/kkg)
Vibratory Spiral Water
Blancher-Cooler Blancher
1.
2.
3.
4.
5.
6.
7.
.8.
9.
Operating labor
Supervision,
fringes benefits,
laboratory, etc.
Maintenance
Depreciation
Insurance, Taxes,
other expenses
Steam
Electricity
Water
Wastewater
0.63
0.41
0.69
1.37
1.10
0.68
0.04
0.00
0.01
0.63
0.41
0.11
0.22
0.18
0.96
0.01
0.16
0.15
Hydrostatic Hot-Gas
Steam Blancher Blancher
0.63
0.41
1.17
1.17
0.94
2.15
0.02
0.17
0.14
0.63
0.41
1.82
1.82
1.46
0-77
0.36
0.16
0.12
$4.93
$2.83
$6.80
$7.55
Annual production - 4.5 kkg/hr x 14 hrs/day x 200 days/yr = 12,600 kkg/yr
1. 1/4 man/shift for operation and 1/4 man/shift for cleaning (2 shifts/day)
with average hourly wage = $5/hr: 2 (2 + 2) $5/(14 x 4.5) = $0.63/kkg =
$0.57/ton.
2. Supervision, fringe benefits, laboratory, supplies, etc. = 65% of opera-
ting labor.
3. Maintenance = 5% of direct fixed capital/yr for Vibratory Spiral and
water blanchers and 10% of direct fixed capital for Hydrostatic Steam and
Hot-Gas blanchers.
4. Depreciation = 10% of direct fixed capital/yr.
5. Insurance, taxes and other fixed expenses = 8% of direct fixed capital/
yr.
6. Steam = $4.30/kkg of steam ($1.95/1,000 Ib steam). Steam cost for Hot-
Gas blancher includes cost of gas ($0.04/kkg).
7. Electricity = $0.014/kw-hr.
8. Water = $0.032/1,000 1 ($0.012/1,000 gal).
9. Wastewater = $0.018/1,000 1 ($0.062/1,000 gal), $0.022/kg BOD ($0.01/
Ib BOD), $0.044/kg SS ($0.02/lb SS).
94
-------
The low capital investment needed for a water blancher is the reason for
its low cost of operation. The higher costs of steam, water and wastewater
treatment for the water blancher as compared to the Vibratory Spiral Blan-
cher Cooler are not sufficient to compensate for the lower costs associated
with capital investment. A nine-fold increase in the cost of steam or a
16-fold increase in wastewater cost would be required to give the Vibratory
Spiral Blancher-Cooler a lower operating cost than the water blancher. The
Hot-Gas Blancher1s high cost is attributable mostly to high capital related
costs, while that of the Hydrostatic Steam Blancher is because of steam
costs.
No attempt was made in these calculations to account for the loss of vegeta-
ble weight when air cooling is used. If frozen vegetables are valued at
$440/kkg ($0.20/lb), a 2% loss of yield would add the equivalent of $8.80/kkg
to the cost of blanching and cooling. Since frozen vegetables are marketed
by weight, such a large penalty for air cooling cannot be economically justi-
fied. A change in the way frozen vegetables are marketed would be required
to take full advantage of the wastewater reduction possible with air cooling.
One should also note that at a price of $440/kkg for frozen vegetables, the
entire cost of blanching and cooling is less than 0.2% of the cost of produc-
tion. The small impact of blanching and cooling on the total cost of
production gives the processor little economic incentive for capital invest-
ment in new blanchers.
Table 9 gives a comparison of the cost of the Vibratory Spiral Blancher
(without cooler) and the water blancher (without flume) as they would be
used for canning vegetables. All other conditions of the cost calculation
remain the same. The water blancher still gives the lowest cost of opera-
tion, but here a five-fold increase in the combined cost of fuel, water and
wastewater cost would make the cost of operating the Vibratory Spiral Blan-
cher about the same as that of the water blancher (A five-fold increase
would be equivalent to a 17% annual increase for ten years). By removing
the necessity of cooling, the fixed capital investment of the Vibratory
Spiral equipment is reduced by 53%, while for the water blancher the removal
of the flume reduces the investment by only 28%.
CONCLUSION
Although water blanching has the highest hydraulic waste load, the low capi-
tal investment of water blanching makes its cost significantly less than any
other blanching technique.
The steam efficiency of the Vibratory Spiral Blancher is higher than other
blanchers and, when combined with a Vibratory Spiral Cooler, its wastewater
generation is much lower than other blanching techniques except for Hot-Gas
Blanching.
Large increases in the cost of energy or wastewater treatment would make
the cost of blanching and cooling with the Vibratory Spiral system compara-
ble to that of a water blancher.
95
-------
Although the Hydrostatic Steam Blancher has a much higher steam efficiency
than conventional steam blanchers, its efficiency is still much less than
that of a Vibratory Spiral Blancher or a water blancher.
Since frozen vegetables are sold by weight, yield loss from air cooling
gives a large economic penalty.
TABLE 9. COST OF BLANCHING WITHOUT COOLING FOR VIBRATORY SPIRAL BLANCHER
AND WATER BLANCHER
Vibratory Spiral Water
Blancher Blancher
Operating Costs ($/kkg)
1. Operating labor 0.63 0.63
2. Supervisor, fringe 0.41 0.41
benefits, etc.
3. Maintenance 0.33 0.08
4. Depreciation 0.65 0.16
5. Insurance, Taxes, 0.52 0.13
etc.
6. Steam
7. Electricity
8. Water
9. Wastewater
$3725 $274T
Fixed Capital Investment ($_)
Purchase Cost 51,000 11,400
Delivery 2,550 570
Installation 10,200,-, 2,280
Floor Space 2,030U' 1,730
Indirect Costs 16,440 4,000
$82,220 $19,980
(1) 7.5m2
96
-------
REFERENCES
1. National Canners Association. "Liquid Wastes from Canning and Freezing
Fruits and Vegetables," Office of Research and Monitoring, Environmental
Protection Agency, Washington, D. C. (1971).
2. Rails, J. W. and Mercer, W. A. "Low Water Volume Enzyme Deactivation
of Vegetables Before Preservation," Office of Research and Monitoring,
U.S. Environmental Protection Agency, Washington, D. C. (1973),
3. Lund, D. B. "Wastewater Abatement in Canning Vegetables by IQB Blan-
ching," Office of Research and Development, U. S. Environmental
Protection Agency, Washington, D. C. (1974).
4. Rails, J. W. and Mercer, W. A. "Continuous In-Plant Hot-Gas Blanching
of Vegetables," National Environmental Research Center, Office of
Research and Development, U. S. Environmental Protection Agency,
Corvallis, Oregon (1974).
5. Bomben, J. L., Dietrich, W. C., Hudson, J. S., Hamilton, H. K. and
Farkas, D. F. Yields and solids loss in steam blanching, cooling and
freezing vegetables. J. Food Sci. 40:660 (1975).
6. Lazar, M. E. and Rasmussen, C. R. Dehydration plant operations. In
Food Dehydration, Vol. 2. p. 132. Avi Publishing Co., Westport, Conn.
(1964).
7. Bomben, J. L. Unpublished data, USDA Western Regional Research Center,
Berkeley, Calif.
8. Perkins, G. Private Communication. Artichoke Industries, Castroville,
Calif. (1974).
9. Bomben, J. L., Dietrich, W. C., Hudson, J. S., Durkee, E. L., Rand, R.,
Farquhar, J. W. and Farkas, D. J. Evaluation of Vibratory Blancher-
Cooler for snap beans and lima beans. Proceedings Seventh National
Symposium on Food Processing Wastes, Atlanta, Georgia. Industrial
Research Laboratory, Office of Research and Development, U.S. Environ-
mental Protection Agency, Cincinnati (1976).
10. Ray, A. Steam blancher uses 50% less energy. Food Processing 36(1):64
(1975).
11. Layhee, P. Engineered F F line yields 5 big production benefits. Food
Engineering 47(2):61 (1975).
12. Peters, M. S. and Timmerhaus, K. D. "Plant Design and Economics for
Chemical Engineers," 2nd ed. McGraw-Hill, New York (1968).
97
-------
DISSOLVED AIR FLOTATION TREATMENT
OF SEAFOOD PROCESSING WASTES —
AN ASSESSMENT
by
D.B. Ertz*, J.S. Atwell**, and E.H. Forsht***
INTRODUCTION
Although dissolved air flotation (DAF) has performed well for many years in
treating wastewaters of varying characteristics, it has not been as success-
ful to date for seafood processing wastes. DAF with chemical addition,
generally considered as primary treatment, represents one of the more soph-
isticated technologies employed to control wastewaters generated by the
United States seafood processing industry. Interest in this physical-
chemical process increased shortly after the adoption of PL 92-500 into law
which required a reduction of pollutants for all industrial dischargers.
Prior to this time, seafood processors which are generally located adjacent
to tidal waters discharged process wastewaters with little or no treatment.
Pursuant to the provisions of PL 92-500, the 1977 effluent limitations for
the tuna processing segment were based on DAF as the Best Practicable Control
Technology Currently Available (BPCTCA). In developing the 1983 require-
ments, other segments of the industry including the processing of crab,
shrimp, salmon, bottom fish, sardine, and herring were assessed with DAF as
the Best Available Technology Economically Achievable (BATEA).
As required by PL 92-500, EPA is currently reassessing and updating the
promulgated regulations. The Edward C. Jordan Co., Inc., through EPA Con-
tract No. 68-01-3287, is developing the technical information basic to re-
assessing the effluent limitations and performance standards for the Canned
and Preserved Seafood Processing Industry. This paper traces the develop-
ment of DAF as a viable treatment alternative for seafood processing wastes
and presents full-scale operating data from several treatment facilities.
It also evaluates current capabilities of the physical-chemical process in
light of established design criteria. Solids handling and disposal alter-
natives are discussed along with the progress of recent investigations
concerning DAF and its applicability to other segments of the seafood in-
dustry.
*Environmental Engineer, Edward C. Jordan Co., Inc., Portland, Maine
**Project Engineer, Edward C. Jordan Co., Inc., Portland, Maine
***Project Officer, U.S. Environmental Protection Agency, Effluent Guidelines
Division
98
-------
BACKGROUND
The Development Documents (1)(2) which support the promulgated seafood
processing effluent guidelines review the applicability and implementation
of DAF within the industry. At the time of publication, four significant
pilot plant studies (3)(4)(5)(6) had been conducted to determine the ef-
fectiveness of this physical-chemical treatment technology. In addition,
two full-scale systems were operating; the Canadian demonstration facility
at British Columbia Packers, Limited (Steveston, British Columbia) and a DAF
unit treating tuna processing wastes at Terminal Island, California. A
third DAF unit was installed at a sardine plant in Maine; however, limited
information was developed due to mechanical problems.
The Canadian studies (7) conducted during the 1971-2 seasons reported re-
moval efficiencies achieved for various seafood processing wastewaters as
listed in Table 1.
TABLE 1. DAF REMOVAL EFFICIENCIES - VARIOUS SPECIES
Percent Removal
Species
Salmon
Herring
Ground fish
Stickwater
Chemical
Additives
Alum and
Polymer
COD
84
72
77
""""
TSS
92
74
86
95
Oil and
Grease
90
85
-
95
The testing performed on the EIMCO flotator installed at Terminal Island for
treating tuna processing effluent was short-term and represented only three
trial runs under varying operating conditions. The results of this limited
testing program are shown in Table 2.
99
-------
TABLE 2. DAF REMOVAL EFFICIENCIES
TERMINAL ISLAND - 1972
Condition A
Chemical Additive Parameter
Sodium Aluminate COD
Polymer TSS
Influent
(mg/1)1
2,850
1,170
Percent
Removal
37
56
„ Alum
Condition B
Polymer
1. Thaw water not included
2. Based on two runs
3. Based on one run
COD
TSS
5,100
667
58
65
The pilot plant study (3) for tuna processing wastes preceded the instal-
lation of the full-scale flotator at Terminal Island. Two similar inves-
tigations were conducted by the National Marine Fisheries Service which
included the treatment of menhaden bailwater (4) and the wastewater from an
Alaskan shrimp processor (5). The Gulf Coast shrimp canning study (6) was
undertaken by EPA Region VI and a local engineering firm. Presented in
Table 3 is a summary of the pollutant reductions achieved in all four in-
vestigations. Research efforts concerning the use of DAF for treating
seafood processing wastewater realized the importance of chemical additives
such as alum and polymers to achieve significant pollutant removals and this
concept has been maintained through the design and operation of full-scale
systems. In some studies, optimizing pH was a recognized factor for achiev-
ing effective operation.
DISSOLVED AIR FLOTATION TREATMENT OF TUNA WASTEWATERS
General
Five tuna processing facilities, three at Terminal Island, California, and
two in American Samoa, have been operating DAF treatment systems for more
than one year. Facilities in San Diego and Ponce, Puerto Rico, have recent-
ly initiated DAF treatment of process wastewaters. An additional system in
Puerto Rico is currently in the construction stage.
With the exception of a few demonstration systems operating in other seg-
ments of the industry, flotation cells treating tuna processing wastes have
been employed as an indicator of the capabilities of dissolved air flotation.
NPDES permit monitoring data and additional information collected by the
100
-------
TABLE 3, SUMMARY OF PILOT PLANT DAF PERFORMANCE
Wastewatef bource
Tuna
Tuna
Menhaden Bailwater
Alaskan Shrimp
Gulf Shrimp
Gulf Shrimp
Chemical Additive No. of Samples
Lime (pH 10-10.5) 1
Polymers
Cationic
Anionic
Lime 1
Ferric Chloride
Alum or Acid (pH 5-5.3) 5
Polymer
Alum 22
Polymer
Acid (pH 5) 5
Alum
Polymer
Acid (pH 5) 2
Alum
Polymer
Parameter
BOD,.
TSS
O&G
BOD,.
S
TSS
O&G
COD
TSS
O&G
COD
TSS
BODS
COD
TSS
COD
TSS
O&G
% Reduction
65
66
66
22
77
81
80
87
Near 100
73
77
70
64
83
51
68
85
-------
processors have historically provided the basis for technological eval-
uation. However, much of this information was accumulated using sample
collection and analytical techniques which were inconsistent with EPA ap-
proved procedures and Standard Methods (8).
To evaluate in-place DAF systems, a field sampling effort was conducted at
the Terminal Island facilities. All sample handling and analytical pro-
cedures were accomplished in accordance with EPA approved methods.
Description of Terminal Island Facilities
Figure 1 shows a schematic diagram of the wastewater treatment system at the
tuna cannery sampled during May 1976, T.I. No. 1. The cannery includes
facilities which process fish meal from tuna scrap and non-tuna petfood in
addition to canning tuna for human consumption and red meat tuna for pet-
food.
From the various processing locations, wastewater is pumped to screens
(0.030-inch openings) where coarse solids are removed. Equalization of the
waste streams is accomplished in the 210,000-gallon surge tank. By throt-
tling the discharge end of a constant speed pump, the equalized wastewater
is introduced into the flocculation tank at a controlled flow rate, usually
250 to 1,000 gpm. Approximately 50 mg/1 of sodium aluminate is added and
mixed in the tank to encourage the formation of floes.
Following the flocculation process, the flow is then split between two
similar flotators (designated as units "A" and "B") which are operated using
a pressurized recycle of 20 to 25 percent. Anionic polymer is added to each
cell at a 2 mg/1 dosage to enhance the removal of suspended solids and oil
and grease.
Even though the two flotation cells are identical in structure, the piping
and polymer feed arrangements are dissimilar. As shown in Figure 1, the
manner in which the pressurized recycle enters the waste stream varies
between the two DAF units. For unit "A", the untreated wastewater is fed
directly into the cell with the recycle injected perpendicular to the waste
stream prior to entering the cell. For unit "B", the pressurized recycle
enters the cell directly with the untreated wastewater entering the pres-
surized recycle stream at right angles. In addition, the polymer is fed
directly into the center feed well of the unit "B", whereas polymer for unit
"A" is injected at a point just before the waste stream enters the cell.
During low flow periods, the chemical feeds are terminated and fresh water
is added to each flotator to inhibit the development of anaerobic conditions.
Incoming wastewater is stored and aerated in the surge tank. The treated
effluent is discharged to Los Angeles Harbor while the sludge (float) is
removed by a skimming mechanism and collected in a holding tank. A cent-
rifuge is available for dewatering the accumulated solids prior to transport
to a landfill site by a licensed hauler.
102
-------
s
LEGEND
— WASTEWATER
•- PRESSURIZED RECYCLE
— SLUD6E
SCREENED
WASTEWATER
FLOCCULATION
TANK
210,000
6ALLON
SURGE
TANK
Na2AI2O4
RETENTION TANK
LOS
*• ANGELES
HARBOR
TO LANDFILL
CENTRIFUGE
FIGURE 1. T.I. NO. 1 DAF TREATMENT SYSTEM SCHEMATIC FLOW DIAGRAM.
-------
At T.I. No. 2, the water pollution control facility, depicted in Figure 2,
includes screens (0.020-inch openings), a 300,000-gallon surge tank, and a
DAF unit. The wastewater treated at this facility (400-2,000 gpm) is
generated during the processing of red meat tuna and non-tuna petfood in
addition to canned tuna for human consumption. Process water from tuna
scrap reduction is discharged to a separate, but similar treatment system.
Total flow pressurization is employed with the back pressure in the reten-
tion tank maintained between 42 and 67 psi in accordance with the flow rate.
The desired flow rate is controlled by throttling the pressure control valve
and pump discharge valve, and by varying the number of constant speed pumps.
Prior to entering the flotation cell, alum is added to the pressurized waste
stream at a feed rate between 40 and 60 mg/1. Approximately 1.0 mg/1 of
anionic polyelectrolyte is introduced into the center of the cell. Fol-
lowing treatment, the effluent is discharged to the harbor while the float
is collected in a storage vessel for loading into a tank truck. The sludge
which contains chemical additives is not dewatered with the available cent-
rifuge. Disposal of the material is accomplished at an approved landfill by
a licensed hauler.
Terminal Island Sampling Program
Throughout the sampling program, the emphasis was placed on determining the
effectiveness of the DAF units operating at both tuna canning facilities.
At T.I. No. 1, automatic samplers were employed to obtain hourly aliquots of
the DAF influent and effluent on a 24-hour basis. Flow proportioning of the
influent samples was not possible; however, it should be noted that the
surge tank equalized the flows somewhat and aliquots were not collected
during periods when fresh water was added to the flotators. For effluent
samples, flow proportioning was accomplished over the entire 24-hour period.
During the sampling effort at T.I. No. 2, aliquots were obtained manually
for the DAF influent and effluent. A composite was formulated for each
sampling point through proportioning the hourly aliquots based on the flow
rate through the flotation cell.
*
Composite samples collected at both wastewater treatment facilities were
used in the determination of all analytical parameters with the exception of
oil and grease. In compliance with EPA approved methods, four grab samples
for oil and grease analysis were randomly obtained. The results of each
determination were arithmetically averaged to evaluate the effectiveness of
the DAF unit for removing oil and grease.
The sampling program for the Terminal Island tuna canneries consisted of two
10-day sessions. At T.I. No. 1, the effort was conducted during May 1976
while T.I. No. 2 was sampled the following month. Since T.I. No. 1 employs
two similar flotation cells with diverse piping and polymer feed arrange-
ments, the effluent for each unit was sampled individually in an effort to
determine variations in their performance. The influent, originating from
the flocculation tank, is identical for each cell. Over the 10-day sampling
104
-------
LEGEND
WASTE WATER
SLUDGE
RECYCLE
SCREENED
CP—
WASTEWATER
3OO,OOO GALLON
SURGE TANK
AIR
INJECTION
|
t-CP—' ,.
FIGURE 2. T.I. NO. 2 DAF TREATMENT SYSTEM SCHEMATIC FLOW DIAGRAM.
-------
period, no significant differences were observed in the performance of the
units "A" and "B". Verification is provided in Table 4 where the concen-
trations and percent removals are shown for the various parameters of in-
terest.
TABLE 4. INFLUENT AND EFFLUENT CONCENTRATIONS WITH RESPECTIVE
PERCENT REMOVALS FOR DAF UNITS "A" AND "B"
T.I. NO. 1
Mean
Influent
BOD
Total Suspended Solids
Oil and Grease
Unit "A" Effluent
BOD
Total Suspended Solids
Oil and Grease
Unit "B" Effluent
BOD
Total Suspended Solids
Oil and Grease
mg/1 3
2,563
1,263
475
1,262
248
81.1
1,374
282
60.4
o Removal
45.3
76.8
80.7
40.5
72.8
86.4
Range
mg/1
1,770-5,803
576-2,910
267-571
768-1,675
184-320
10.9-267
550-1,795
108-384
11.2-173
% Removal
11.0-79.3
59.7-90.0
19.5-98.0
4.6-76.6
33.3-91.4
67.1-98.0
The overall removal rates for the treatment system were obtained by
averaging the values obtained for each DAF unit. The summarized removal
efficiencies for T.I. No. 1 are compared to the projected removals for the
tuna industry and BPCTCA as shown in Table 5.
TABLE 5- REMOVAL EFFICIENCIES - T.I. NO. 1
Parameter
BOD5
Total Suspended Solids
Oil and Grease
Mean
42.9%
74.8%
83.5%
Range
7.8-77.9%
46.5-89.6%
43.3-98.0%
Projected
for BPCTCA
40%
70%
85%
As can be seen from Table 5, the removal efficiencies for BOD,, and total
suspended solids observed for the entire DAF system exceeded the projected
106
-------
removals. However, the 10-day average for oil and grease was slightly below
the anticipated 85 percent. The operating pH in the flotation cell ranged
from 6.4 to 8.7.
During June 1976, a sampling effort was undertaken to evaluate the effec-
tiveness of the T.I. No. 2 water pollution control facility which consisted
of one flotation cell. The influent and effluent concentrations quantified
over a period of 10 days are summarized in Table 6.
TABLE 6. DAF INFLUENT AND EFFLUENT CONCENTRATIONS - T.I. NO. 2
Mean (mg/1) Range (mg/1)
Influent
BOD 883 563 -1,149
Total Suspended Solids 367 306 - 418
Oil and Grease 113 50 - 206
Effluent
BOD 669 339 - 936
Total Suspended Solids 190 132 - 308
Oil and Grease 40.4 4.5- 123
Employing the daily concentrations determined over the duration of the
sampling effort, BOD,., total suspended solids and oil and grease removals
were calculated. The effectiveness of the DAF unit in decreasing the mass
of pollutants discharged to receiving waters is exhibited in Table 7.
TABLE 7. REMOVAL EFFICIENCIES - T.I. NO. 2
Parameter
BOD
Total Suspended Solids
Oil and Grease
Mean
24.3%
48.2%
64.3%
Range
12.0-47.0%
18.5-62.5%
0 -96.8%
Projected
for BPCTCA
40%
70%
85%
For each of the three parameters listed in the table above, the observed
effectiveness of the wastewater treatment system was significantly less than
that projected for the 1977 tuna limitations and the efficiencies documented
at T.I. No. 1. The operating pH was slightly more variable, ranging from
6.3 to 9.3.
107
-------
Since both DAF systems operate in the neutral pH range or slightly above,
this cannot be considered as an overriding factor in*T.I. No. 1 achieving
higher removal rates. It was also observed that the treatment facilities
were operated with almost equal proficiency. Therefore, the major consid-
eration was the significant difference between the influeift concentrations
measured. For each parameter considered, the influent concentrations quan-
tified at T.I. No. 1 were at least three times greater than those observed
during the sampling period at T.I. No. 2. Greater pollutant concentrations
can be partially attributed to the lower flow ratio, in terms of gallons per
ton of tuna processed, maintained by the T.I. No. 1 processing facility. A
thaw system which recirculates the water during a specific cycle while in-
jecting steam and adding sulfuric acid has made a substantial contribution
in reducing the volume of wastewater requiring treatment. The resulting
flow ratio was determined to be an average of 47 percent of that documented
at T.I. No. 2.
During the sampling program conducted at the Terminal Island tuna canneries,
influent samples were analyzed for soluble BOD,.. A standard method is not
available for determining this parameter; however, it is general practice to
use the filtrate from the total suspended solids analysis which has passed
through 0.45 micron filter paper. The relationship between the soluble
portion of the influent BOD^ as measured and total BOD_ removal is shown in
Figure 3. For this data, the line of best fit was determined through the
method of least squares. A fairly good correlation (r= -0.80) indicates
that as the influent soluble BOD,, portion increases, the percent total BOD,.
removal decreases. Therefore, minimizing the soluble portion of the influent
through pH optimization to the isoelectric point, becomes critical in
maintaining consistent BOD,, removals.
At the present time, the sludge removed from the flotation cells operating
at T.I. No. 1 and T.I. No. 2 is landfilled. Samples of this material were
collected on a daily basis and subjected to chemical analysis. The results
of analysis as it relates to production and flow are shown in Table 8.
TABLE 8. DAF SLUDGE PRODUCTION - T.I. NO. 1 AND T.I. NO. 2
T.I. No. 1 T.I. No. 2
Volume Ratio (gal/1,000 gal
wastewater treated) 19.4 5.5
Average Percent Solids 9.2 15.6
Production Rate - Dry Weight Basis
(lb/1,000 Ib of tuna processed) 12.1 12.4
108
-------
100-
ui
cc
in
o
O
CO
z
yj
O
oc
Ul
Q.
80-
60-
40-
20-
LEGEND
T T.I. No.1
• T. I. No. 2
.20
.40
.60
.80
1.00
SOLUBLE BOD,
TOTAL BODC
OF DAF INFLUENT
FIGURE 3. PERCENT B0t>5 REMOVAL AS A FUNCTION OF THE
SOLUBLE PORTION OF THE INFLUENT BOD5 .
1Q9
-------
Description of American Samoa Facilities
The tuna canneries in American Samoa operate wastewater treatment facilities
which are very similar to the system installed at T.I. No. 2. However,
the characteristics of the wastewater generated by A.S. No. 1 and A.S. No. 2
are significantly different. Neither cannery in American Samoa operates a
solubles plant, thereby requiring the precook water and presswater to be
discharged to the treatment system which results in higher raw waste loads.
In addition, petfood production in American Samoa involves red meat tuna
only.
Relative to the A.S. No. 1 and A.S. No. 2 wastewater treatment systems, the
incoming process flow is screened and subsequently pumped to a surge tank.
Following equalization, the total flow is pressurized and chemically con-
ditioned with alum and polymer prior to entering the flotation cell. The
treated effluent is discharged to receiving waters while the DAF sludge is
collected and hauled to a land disposal site. ,Solids dewatering is not
accomplished at either facility. During periods of low incoming flows, the
treatment system at A.S. No. 2 is capable of recycling the effluent to the
surge tank as opposed to A.S. No. 1 which is not equipped to operate in this
mode.
American Samoa Sampling Program
In August 1976. the EPA National Enforcement Investigation Center (NEIC)
conducted a compliance monitoring survey of the two American Samoa canneries
(9)(10). The DAF influent and effluent were sampled on a 24-hour basis or
during the period of discharge to determine treatment efficiencies at each
facility. Effluent aliquots were collected hourly and composited on a
flow-weighted basis for BOD,, and total suspended solids analyses. Samples
for influent total suspended solids analysis were composited on an equal-
volume basis. During each day, grab samples of the influent and effluent
were obtained to determine oil and grease removals. All samples were
collected and analyzed as specified by the NEIC chain of custody and an-
alytical quality control procedures.
The NEIC sampling program was conducted during the period of August 18-26.
During that time, samples were collected for six 24-hour periods at A.S. No.
1 while the treatment efficiency at the A.S. No. 2 was evaluated based on
eight days of data. Due to analytical limitations, influent BOD,, concen-
trations were not determined at either facility.
The influent and effluent data collected during the NEIC compliance mon-
itoring survey is summarized in Table 9. It should be noted, however, that
the oil and grease analysis of the DAF influent was performed on four of the
six sampling days.
-------
TABLE 9. DAF INFLUENT AND EFFLUENT CONCENTRATIONS - A.S. NO. 1
Mean (rog/1) Range (mg/1)
Influent
BOD
Total Suspended Solids 4,880 2,700-13,100
Oil and Grease 1,500 1,080-2,770*
Effluent
BOD 2,480 1,500-3,900
Total Suspended Solids 220 70-620
Oil and Grease 120 26-340
*0il and grease concentration of 2,770 mg/1 is based on one grab sample
and not the normal averaging of 3-4 grab samples.
The ability of the treatment system to remove constituents from tuna pro-
cessing wastewater is measured in terms of removal efficiencies. Using the
respective concentrations for total suspended solids and oil and grease, the
appropriate removal rates were determined as shown in Table 10.
TABLE 10. REMOVAL EFFICIENCIES - A.S. NO. 1
Proj ected
Parameter Mean Range for BPCTCA
BOD
Total Suspended Solids 95% 94-98% 70%
Oil and Grease 88% 64-997= 85%
The removal efficiencies for both parameters substantially exceeded the
removals projected in developing the BPCTCA guidelines for the canned tuna
industry. The operating pH of the flotator ranged from 4.2-6.5, approaching
the isoelectric point for fish protein (4.5-5.0).
Employing a similar method for A.S. No. 2, monitoring data collected for a
total of eight days can be summarized. In Table 11, the characteristics of
the wastewater treated by the DAF unit and the resulting effluent are pre-
sented.
Ill
-------
TABLE 11. DAF INFLUENT AND EFFLUENT CONCENTRATIONS - A.S. NO. 2
Mean (mg/1) Range (mg/1)
Influent
BOD
Total Suspended Solids 860 520-1,100
Oil and Grease 540 87-730
Effluent
BOD 970 820-1,120
Total Suspended Solids 290 60-510
Oil and Grease 230 18-280
To evaluate the effectiveness of-the single flotation cell over the sampling
period, a comparison can be made between the removals achieved at A.S. No. 2
and the 1977 levels. This is accomplished in the data summary presented in
Table 12.
TABLE 12. REMOVAL EFFICIENCIES - A.S. NO. 2
Projected
Parameter Mean Range for BPCTCA
BOD - - 40%
Total Suspended Solids 66% 23-93% 70%
Oil and Grease 57% 33-97% 85%
Although BOD,, removals were not monitored, it appears that the A.S. No. 2
treatment system is operating well below the desired levels. The operating
pH which is critical to the performance of the physical-chemical process was
maintained within a range of 5.9-7.4, considerably higher than the iso-
electric point of fish protein.
The DAF system with the addition of alum and polymer operating at A.S.
No. 1, achieved significantly greater removal efficiencies than the A.S. No.
2 facility in addition to exceeding the projected removals for 1977 or best
practicable treatment. In evaluating the performance of the two facilities,
the higher removal rates observed at A.S. No. 1 can be attributed, in part,
to a lower operating pH range within the flotator. In addition, the greater
influent concentrations can provide a basis for better performance while
actual operation of the two systems also demands consideration. As docu-
mented by NEIC, operating personnel at A.S. No. 1 were attentive and ap-
peared to have a better understanding of the treatment process. The flow
112
-------
rate was maintained relatively constant throughout the day. An evaluation
of the design of the two wastewater treatment systems by NEIC concluded that
the A.S. No. 2 DAF unit is overdesigned for the flow undergoing treatment.
Similar to the T.I. No. 1 facility, influent concentrations for total sus-
pended solids and oil and grease measured at A.S. No. 1 exceeded those
observed at A.S. No. 2 by more than 300 percent. Effluent BOD values were
also significantly greater at A.S. No. 1. Recirculated thaw water with
sulfuric acid addition produced lower flow ratios (approximately 50 percent
of A.S. No. 2), increased influent waste loads, and depressed the pH of the
wastewater subjected to treatment.
RECENT INVESTIGATIONS
Treatment technologies more sophisticated than screening are not required
for the remaining segments of the seafood industry until the implementation
of 1983 limitations. Therefore, investigations of DAF involving wastewater
generated during the processing of other commodities to date have not been
extensive.
On a pilot scale, the National Marine Fisheries Service has evaluated the
application of air flotation to treat the effluent from a cannery in the
Puget Sound area which processes salmon and tuna. The treatment system
included screening (20-mesh) followed by a wastewater concentrator with a
165-mesh or 400-mesh screen. During the concentration process, air is
entrained into the waste stream which is subsequently released within a
flotation cell placed in series with the concentrator. This scheme is
essentially a dispersed air flotation system with alum and lime added up-
stream of the concentrator and polymer assisting in the flotation process.
The pH in the cell was maintained in the 8.0-10.0 range as opposed to the
isoelectric point of fish protein.
Additional information is currently being developed in Louisiana through an
EPA demonstration study. A full-scale DAF unit has been employed for the
treatment of shrimp canning wastes as well as the wastewater produced from
the oyster canning operation at the same processing facility. Further
investigations will be conducted during the 1977 shrimp canning season.
SOLIDS HANDLING AND DISPOSAL ALTERNATIVES
During the initial assessment of waste management within the seafood indus-
try, byproduct recovery was emphasized. In-plant controls to prevent pro-
duct related materials from entering the waste stream were thought to be an
integral part of the total waste management program. Incorporating solids
which are produced from the implementation of wastewater treatment tech-
nologies, including screens and DAF, into a saleable product was given con-
sideration as a viable alternative for ultimate disposal.
To minimize handling costs, sludge generated from relatively large scale
wastewater treatment systems should be concentrated prior to further pro-
cessing or final disposal. Flotation sludge produced during the treatment
113
-------
of wastewater containing fats and oils of animal origin can be difficult to
dewater. Currently, one facility which dewaters the DAF float generated
from treating seafood processing wastes has been identified. In this par-
ticular case, a centrifuge has proven successful in substantially reducing
the moisture content of the sludge accumulated from a DAF system operating
within the tuna industry.
To determine current practices of industries producing sludge with charac-
teristics similar to that experienced by tuna processors, a survey of meat
packers, poultry processers and Tenderers was conducted. Emphasis was
placed on identifying dewatering methods other than decanting and processors
rendering sludge which contains chemical coagulants such as alum and poly-
mers.
A total of 40 processors employing DAF to treat their wastewaters were
contacted to discuss methods of float handling and disposal. The results of
the survey are summarized in Table 13.
TABLE 13. DAF FLOAT HANDLING AND DISPOSAL FOR RELATED INDUSTRIES
Total Number Surveyed
Render
Chemical Additives +
Render
Chemical Additives +
Dewater + Render
Land Disposal
Other
Red Meat
20
12
2
5
3
Poultry
10
7
0
2
1
Renderer
10
10
0
0
0
Only two red meat processors reduced the moisture content of the sludge
through mechanical means, which was identified as centrifugation. Gravity
settling with subsequent decanting was found to be the most common dewater-
ing method with 15 processors employing this alternative. In some instances,
the sludge was heated to encourage liquid-solids separation; however, this
is generally employed in conjunction with in-house byproduct recovery
operations such as rendering. Slightly more than half of the facilities
contacted did not concentrate the float to yield a greater solids content
prior to further processing or ultimate disposal.
114
-------
Approximately two-thirds of the facilities operating DAF units rendered the
float in-house or hauled it to an off-site rendering operation. As one
would expect, all Tenderers reintroduced their residuals into their oper-
ations; while nine of ten employed chemicals to aid removals. Seven of the
eleven meat packers and poultry processors using coagulants subsequently
subjected the recovered solids to a rendering operation.
The results of this survey are by no means conclusive relative to the abil-
ity of the seafood industry to recover DAF sludge and incorporate it into
a useful byproduct. It does appear that rendering sludge containing chem-
ical coagulants is feasible, reinforced by the current practices of other
food related industries. However, further investigation is required to
determine its applicability to the seafood processing industry. At the
present time, the major considerations include the effect of chemicals on
the quality of the rendered product as well as the economics of operating
this type of byproduct facility.
SUMMARY
In general, the seafood processing industry can be considered relatively
unsophisticated with regard to treating its process wastewaters. Much of
the industry provides either screening or no treatment before discharging to
inland or tidal waters. Prior to July 1, 1977, measures must be implemented
to provide coarse solids removal with the equivalent of 20-mesh screens.
However, tuna processors which are generally large volume producers with a
somewhat consistent supply of raw material have been mandated to meet more
restrictive limitations based on DAF. The larger tuna canners located in
Terminal Island and American Samoa and most recently those in San Diego and
Puerto Rico have become the forerunners with respect to dissolved air flo-
tation treatment on an industrial level for the entire seafood industry.
The capability of any treatment system is first dependent on the design and
secondly, actual operation. Based on the operating data presented herein,
the facilities monitored meet the generally accepted design criteria and in
some cases fall on the conservative side. When evaluating DAF in light of
industrial waste treatment, the criteria of concern are overflow rate,
solids loading, air to solids ratio, hydraulic retention time and the abil-
ity to maintain the appropriate pH level. During previous investigations,
it has been shown that the design criteria outlined in Table 14 will provide
acceptable results. The pH for effective chemical coagulation of a pro-
teinaceous waste is optimal at the isoelectric point which for fish pro-
cessing wastes falls into the 4.5-5.0 range. In addition, the treatment
system should be designed to optimize flocculation of suspended solids with
the correct chemical dosages determined for the specific wastewater to be
treated. Incorporating flexibility and ease of operation into the final
design is also a necessity.
115
-------
TABLE 14. DAF DESIGN CRITERIA FOR SEAFOOD PROCESSING WASTES
Overflow Rate 2 gpm/sq ft
Solids Loading 1 Ib/hour/sq ft
Air to Solids Ratio 0.01-0.04
Hydraulic Retention Time 1 hour
With a well-designed system, the capabilities of the water pollution control
facility relies heavily on the operator and his working knowledge of the
treatment process. In other words, operation is the key to successful
wastewater treatment whether it is an industrial or municipal facility.
With regard to'industrial wastewater treatment, in-plant measures to reduce
water consumption and concentrate individual waste streams gains importance.
These concepts are borne out when considering the American Samoa DAF system
which achieved the best removal efficiencies. A lower flow ratio for an
adequately designed system, in conjunction with an attentive operator and a
pH close to the desired range, produced excellent removals for suspended
solids and oil and grease.
It should be realized that data collected at the four tuna canneries was
summarized from information accumulated over a period which ranged from 6 to
10 days at each facility. In reviewing the monitoring data, two of the
canneries equalled or exceeded the removal efficiencies for total suspended
solids and oil and grease which were projected in developing the 1977 lim-
itations. However, the level of consistent BOD,, removal for tuna processing
wastewater has not been documented conclusively. The processing of non-tuna
petfood at Terminal Island, where extensive BOD,, data was collected, creates
some uncertainty. The soluble portion of BOD,, remains a critical factor
wij:h increased removal rates being achievable through good operation and by
maintaining the optimal pH range. Further optimization will be required to
achieve the more restrictive 1983 limitations.
As operating experience increases and additional information develops at
these and other facilities within the tuna industry, the capabilities of
DAF will be more fully realized. This information will become the basis for
subsequent comparisons relative to the removals necessary to meet the 1977
limitations as well as those for BATEA in 1983. Operating data for tuna
wastewater treatment will be further supported by information developed
through subsequent pilot plant investigations and ongoing full-scale demon-
stration studies.
For the most part, the handling and disposal of residuals produced by DAF is
considered to be a difficult problem by the seafood processing industry.
Information has been presented outlining the approach other industries have
taken. While some meat processors are dewatering float through mechanical
devices, a greater number incorporate this material which contains chemical
116
-------
additives into byproducts. Of prime concern is the value of the processed
material and the ability to meet acceptable specifications for certain
products. One tuna processor has found the means to successfully dewater
residuals by centrifugation; however, byproduct recovery from this material
has not been implemented on a full-scale basis. The economics of solids
disposal requires further investigation with viable alternatives developed,
especially for those segments of the seafood industry which experience more
seasonability and higher variability with raw material supply than the
tuna processors.
REFERENCES
1. Development Document for Effluent Limitations and Guidelines and Stan-
dards of Performance for the Catfish, Crab, Shrimp, and Tuna Segment of
the Canned and Preserved Seafood Processing Industry, Point Source
Category, U.S. Environmental Protection Agency. Washington, DC 20460,
June 1974.
2. Development Document for Effluent Limitations Guidelines and Standards
of Performance for Fish Meal, Salmon, Bottom Fish, Clam, Oyster, Sar-
dine, Scallop, Herring and Abalone Segment of the Canned and Preserved
Seafood Processing Industry, Point Source Category, U.S. Environmental
Protection Agency, Washington, DC 20460, September 1975.
3. Jacobs Engineering Company, "Pollution Abatement Study for the Tuna
Research Foundation, Inc.," 1971.
4. Baker, D.W. and C.J. Carlson, "Dissolved Air Flotation Treatment of
Menhaden Bailwater," Proceedings of the 17th Annual Atlantic Fisheries
Technology Conference (AFTC), Annapolis, Maryland, 1972.
5. Peterson, P.L., Treatment of Shellfish Processing Wastewater by Dis-
solved Air Flotation, Unpublished Report, Seattle: National Marine
Fisheries Services, U.S.D.C. 1973.
6. Mauldin, A.F. and A.J. Szabo, Shrimp Canning Waste Treatment Study,
Project S 800904, Office of Research and Monitoring, U.S. Environmental
Protection Agency, Washington, DC 20460, 1974.
7. Claggett, F.G., Clarification of Fish Processing Effluents by Chemical
Treatment and Air Flotation, Technical Report No. 343, Fisheries
Research Board of Canada, 1972.
8. Standard Methods for the Examination of Water and Wastewater, 14th
Edition, American Public Health Association, American Water Works
Association, and Water Pollution Control Federation, Washington, DC,
1975.
117
-------
9. "Compliance Monitoring and Wastewater Treatment Evaluation, Van Camp
Sea Food Company, Pago Pago, American Samoa," EPA-330/2-76-035, Office
of Enforcement, U.S. Environmental Protection Agency, December 1976.
10. "Compliance Monitoring and Wastewater Treatment Evaluation, Star-Kist
Samoa, Inc., Pago Pago, American Samoa," EPA-330/2-76-036, Office of
Enforcement, U.S. Environmental Protection Agency, December 1976.
118
-------
COMMERCIAL FEASIBILITY OF RECOVERING TOMATO PEELING RESIDUALS
by
W. G. Schultz*, H. J. Neumann*, J. E. Schade*, J. P. Morgan*,
P. F. Hanni*, A. M. Katsuyama**, and H. J. Maagdenberg**.
ABSTRACT
In the United States, tomatoes are peeled for canning by first immersing
in a caustic bath to loosen the skin; then, the peel is removed either
mechanically with rubber discs or with water sprays. When the peel is
removed mechanically, the peel solids are not diluted and therefore are
similar to the pulp of whole tomatoes. Since this removed peel is at least
12% of the unpeeled weight and the peel is about 96% pulp, this peel pulp is
a potential source of food material. It also is attractive economically
because there is a possible pulp recovery value of $230/hr from a typical 40-
t/hr peeling operation; processing this material would cost about $188/hr
the first year and $43/hr thereafter, leaving a $187/hr net return in the
second and subsequent years. A two-year project was undertaken and funded
jointly by USDA-WRRC, NCA, EPA, and the California tomato processors. In
1975 peel from regular cannery operations was processed through a 20-gpm
(5t/hr) continuous-flow line. This processing consisted of acidifying
the peel to pH 4.2 with food-grade hydrochloric acid, then separating the
pulp from the skin with a paddle finisher. Recovered peel pulp was found
to be of food quality, but contained high peeling-aid residues (150-450 ppm).
Practically all tomato peeling operations use a peeling aid in the caustic
bath to facilitate uniform peeling, particularly on the shoulder of the
tomato. Peeling aids in current use are approved for peeling but not as
additives to the final product. In 1976, a 1-t/hr pilot peeling line was
set up at a cannery to study modifications in the peeling process. The
purpose of the modification was to pretreat the tomatoes by immersion in
a 150 F aqueous bath (pH 3.6) containing about 0.15% food-grade octanoic
acid. Recovered* pulp could meet USDA Quality Grade A, and the octanoic
acid levels were low, about 30 ppm. Discussions are being held with FDA
on several aspects because the proposed use of this recovered peel pulp
is in combination with tomato pulp from regular sources for canned products,
such as tomato sauce, puree, catsup, paste, and fill juice for peeled
tomatoes. The compositions of these products are governed by the FDA
Standards of Identity.
*Western Regional Research Center, Agricultural Research Service,
U.S. Department of Agriculture, Berkeley, California 94710
**National Canners Association, 1950 Sixth Street, Berkeley,
California 94710
119
-------
INTRODUCTION
In the United States, tomatoes are normally peeled by loosening the skin
with a hot-caustic bath and removing the peel (skin with adhering pulp)
either mechanically with rubber discs or with water sprays. The use of
water sprays has declined because of the large amount of water needed
(500-1500 gal/t) and the subsequent problem of waste disposal of the
dilute solution. Removal of the peel mechanically with rubber discs
reduces the water consumption to a negligible amount so that the peel
has about the same solids content as fresh tomatoes. This material is
currently discarded as solid waste. Peel removed that is recoverable
constitutes at least 12% of the original tomato weight. Since this peel
is about 96% pulp, it is a potential source of food material. This pulp
is also attractive economically because there is a potential net pulp
worth of $187/hr for a typical 40 t/hr peeling operation based on a raw
material value of $50/t. There are about 1.3 million tons of tomatoes
peeled each year in the United States, resulting in at least 150,000 t/yr
of recoverable pulp. Currently that pulp is discarded as peel at an
expense of $2.50 to $5.00/t, or as much as $750,000/yr for the peeling
industry. Despite the economic incentive, there were several technical
obstacles such as insecticide residues, the lye and surfactants from the
caustic-peeling applicator, acidification of the alkaline peel, recovered-
pulp quality, product labeling, etc. With these potentials and obstacles
in mind, a two-year project, beginning in 1975, was undertaken jointly by
the USDA Western Regional Research Center, National Canners Association
Western Research Laboratory, and the tomato processing industry. The ini-
tial plans were described in April 1975 (1) and were based on trends in
commercial practice and prior information on pulp recovery potential (2)
(3) (4) (5). Experimentation was implemented during the 1975 tomato
processing season. The 1976 work was to develop methods to answer the
problems discovered in 1975.
METHODS
1975 Experimentation—Peel Processing & Pulp Characterization
Peel was experimentally processed on a daily basis during 1975 as received
from conventional caustic peeling at the Tillie Lewis Foods, Plant W,
Antioch, California. This cannery processed VF-145 tomatoes through washing
and sorting, then about 40 t/hr were diverted to their peeling operation.
The diverted tomatoes were next immersed in a caustic bath; this was a
typical industrial situation using 10-12% (w/w) sodium hydroxide with up
to 0.2% sodium 2-ethylhexyl sulfate at 200-210°F and a nominal half-minute
immersion. From this bath the tomatoes went into two types of mechanical
peel removers, a flat-bed disc type followed by peel-tag removal rolls (FMC
PR-20 Tomato Peel Remover, 1 machine) and rotary-cylinder, rubber disc
types (Magnuson Model C Peel Scrubbers, 4 machines). All of these peel
removers fed into a common pump which sent the peel to the experimental
peel-pulp recovery area. This peel, from the normal cannery peeling, was
received about 5 minutes after the clean tomatoes first entered the caustic
applicators.
120
-------
Figure 1 shows the 1975 experimental equipment layout, and Figure 2 is a
diagram of the pertinent operations for both the conventional and the
experimental processing. The primary variables were: (1) the extractor
screen size and paddle clearance, (2) place of acidification, (3) hot-break
temperature, (4) lag time between the hot break and canning, (5) evaporator
temperature and degree of pulp-solids concentration, and (6) the heat-
processing time of cans. These variables were evaluated in terms of:
(1) pulp yield, (2) product quality, and (3) tomato Standards of Identity.
Peel was received continuously at 10-30 gpm, and acidified with food-grade
hydrochloric acid either immediately in the Peel Tank or later in the Hot-
Break Vessel; this acidification was continuously controlled by an automatic
pH recording controller. The Peel Tank volume was 48 gal. and the Hot-Break
Vessel was 250 gal. Peel flowed continuously into each of these vessels
and constant volumes were held by overflow weirs.
From the Peel Tank, the peel flowed into the Extractor which was used to
separate the pulp from the skin, seeds, and fibers; seeds and fibers are
present from tomatoes that disintegrate during peel removal. This Extrac-
tor was a standard FMC Model 50 Pulper with a 0.030-in. screen and set with
a 0.5-in. screen-paddle clearance.
A Hot-Break Vessel was provided to inactivate enzymes that might be pre-
sent, to reduce subsequent microbiological growth if the incoming pulp
was to be held at the projected incoming pulp temperature of 120 F for an
extended period. It also provided thermal-exposure testing since caustic
exposed tomato pulp is more susceptible to color and flavor changes.
Next the pulp flowed into the Pulp Tank for a final check and recording
of pH. A material (mass) balance was made for each trial by weighing
the Extractor waste and measuring the volume of the recovered pulp.
Recovery was determined with 400-1,000 gal. batches; the weight of the
recovered pulp and Extractor waste was equal to that of the incoming
peel. Recovered peel pulp was concentrated in 1,000 gal. batches at
160-200°F to concentrations of 10-20% TS (total solids) in the cannery
single-stage vacuum evaporator.
Both fresh and canned samples were made up for subsequent analyses. All
canned samples were hand filled into size 211 x 400 unenameled cans,
sealed with a double seamer, heat processed in boiling water for 40 minutes,
and cooled to 100°F in 75°F water. These canned samples were analyzed and
judged on a 100% recovered-pulp basis without blending with other tomato
materials.
1976 Experimentation—Pilot Scale Modified Caustic Peeling
The 1976 experimentation was primarily directed towards solving the
peeling-aid residue problem which was identified during the 1975 pulp
recovery experimentation. This development was carried out during the
spring laboratory-peeling tests and the summer cannery pilot-peeling tests.
In the spring, laboratory peeling was conducted on tomatoes using a wide
121
-------
ro
ro
PEEL FROM
CANNERY
OPERATION
DISCARD
PEEL TANK, ACIDIFICATION
TO
VACUUM
EVAPORATOR
Figure 1. 1975 Equipment,
-------
Conventional Processes
Experimental Process
1
Field
Puree
*
Wash
1
Sort
1
Crush
1
Hot Break
I
Pulp
IT*. Tl
Finish
1
Concentrate
1.
Canning
1
Run Tomatoes
1 Peeling
*
Wash
1
Sort
1. .
Caustic Dip
!
Peel Removal
1
Inspect
jmace
V
Canning
1
Acid
i
_ . .
1
Extractor -
Pulp i
Hot Break
1
|
1
— »-Skin
1
1
Heat
Process
J
Cool
Figure 2. 1975 Flow Diagram.
123
-------
range (about 70) of compounds including many surfactants (surface active
agents) to determine the chemical structure that aided peeling. The wetting
action (interfacial tension) of the potential peeling aids was checked,
and this surface-wetting action was found to provide little assistance
in selecting peeling aids because much of their real benefit seems to be
in chemical activity. Potential peeling aids were applied in two ways:
(1) directly in the caustic bath in the traditional manner, and (2) as a
pretreatment prior to immersing the tomatoes in the caustic bath; these
were compared to peeling with a plain caustic bath. The purpose of the
pretreatment was to apply only enough peeling aid to permeate the skin as
needed to aid peeling or to allow the caustic to act more effectively in
the applicator. It was also assumed that the optimum temperature and time
for applying the peeling aid might differ from those existing in the caustic
applicator. The best conditions found in the laboratory peeling tests were
incorporated in a 1-t/hr pilot*-peeling line at Hunt-Wesson Foods, Plant A,
Hayward, California during the 1976 tomato processing season. This line
operated with the regular cannery tomatoes, usually Variety UC- 134. Washing,
sorting, and pretreatment was carried out solely on the pilot equipment.
Typical operating conditions and equipment are shown in Figure 3. Peeling
was performed continuously, typically in 45 minute runs. Since there was
insufficient peel flow from this pilot line to continuously acidify the
peel, acidification was performed batchwise at about 4-5 min. intervals.
Extraction of skin and pulp was done at the end of each run.
This cannery received tomatoes usually in bulk 20-t loads (truck and trailer)
as is typical for California canners. The tomatoes were removed from the
trucks by the cannery personnel through a water wash-out and carried by a
flume into a sump; from the sump they were elevated out and spray washed,
passed over a screen to remove gross trash and tomatoes less than 1.2-in.
in diameter and then flumed on to further rinsing and hand sorting. From
this flume, prior to further cannery washing and sorting, part of the toma-
toes were diverted to the pilot-peeling line. These tomatoes were immersed
in water, elevated out, and passed over a 1-ft x 10-ft rubber disc flat-bed
scrubber having water sprays; this was the final washing. The tomatoes
were then passed over a sorting belt for hand sorting; the degree of hand
sorting was varied so as to compare mold counts in the recovered pulp.
The pretreatment immersion was in a 17-in. x 10-ft trough having a paddle-
type conveyor which carried the tomatoes through in positive displacement
fashion. The heated solution was recirculated from entry to exit at about
20-gpm, and was controlled and varied from 75°F to 200°F, depending on the
experiment. Immersion time could be varied from 15 seconds to three
minutes. From this pretreatment, the tomatoes were removed on an open-
mesh elevator for a variable draining time of 10 seconds to 2 minutes.
After draining, the tomatoes dropped into the Caustic Applicator for
10-sec. to 2-min. immersion in 11% (w/w) sodium hydroxide at 210°F. An
applicator such as the commercial FMC Hi-Ton Tomato Peeler has a drain
period of about 50% of the immersion time, which not only removes excess
caustic solution but provides a further period for the caustic to act on
the tomato. The pilot applicator did not have a similar drain period
so this was simulated by a variable-speed, open-mesh belt normally held to
0.18-min. residence time. Tomatoes then passed over rotating slitting blades
124
-------
WASH
HOLD
SLIT
ACID
PRETREATMENT
PEEL REMOVAL
T
peelings
CAUSTIC
Material
Balance
peeled tomatoes (pdt) 85%
FINISHER
skin (waste) O.I
peel pulp (product) 14.4%
100.0%
Nominal Conditions
Tomatoes: Variety 134, field run,
Washing: 1st and 2nd by water immersion; 3rd by flat-bed
rubber discs win spray rinse.
Sorting: hand.
Pretreatment: 0.5 min., 150°F, 0.15% w/w octanoic acid.
0.5 min. drain on elevator.
Caustic Applicator: 0.5 min., 200°F, 11% w/w sodium hydroxide.
Drain & Hold: 0.18 min.
Slitter: rotating knives.
Peel Removal: Flat-bed, rubber discs.
Acidification: (whole) peel with hydrochloric acid.
Peel-pulp & Skin Separation: paddle pulper with 0.030-in.
screen and 0.5-in. clearance.
Figure 3. 1976 Pilot Peeling.
125
-------
and onto a 12 -in. x 10-ft set of flat-bed rubber-disc peel removers which
were operated without water sprays so as not to dilute the peel. This dry
removal of peel is increasingly being practiced commercially. Peel dropped
onto a full-length pan, flowed down to a 10-gal. pot, where the peel was
acidified with hydrochloric acid to pH 4.2 ±0.2. This acidified peel was
separated into skin and pulp fractions with a Langsenkamp Indiana Laboratory
Pulper equipped with a 0.030-in. screen. This recovered pulp was canned
in 211 x 400 enameled containers, processed for 45 minutes in boiling water,
and cooled to about 100°F in 75°F water. For a material balance on each
trial, both the recovered pulp and peeled tomatoes were weighed, typically
1,000 to 2,000 Ibs/trial.
Analyses
The analytical methods used in 1975 and 1976 are listed in the Bibliography,
Analytical Procedures. In addition, all canned recovered pulps were graded
by the USDA Agricultural Marketing Service using the same U.S. Grade Stand-
ards for Canned Tomato Pulp as for regular commercial products. The fatty
acids were analyzed by a solvent extraction, esterification, and GLC detec-
tion method which is currently undergoing further refinements.
RESULTS & DISCUSSION
While the overall 1975 results (Table 1) showed there is food potential in
recovering pulp from caustic tomato peelings, there was one major detrac-
tant—the peeling aid residue at 150-450 ppm. Of the currently used commer-
cial peeling aids (sodium 2-ethylhexyl sulfate, sodium mono- and di-methyl
naphthalene sulfonates, or fatty-acid mixtures which contain predominantely
odd-numbered carbons), none seemed suitable for clearance as food additives.
Even if an additive-grade peeling aid was currently available, the 150-450
ppm residue is large enough to raise questions as to whether it would need
to be declared on the product label as an intentional additive. These
questions on the peeling aid determined the direction of the 1976 work, the
second and final scheduled year of the project. While it is possible to
peel tomatoes without a peeling aid, it is generally acknowledged that higher
caustic concentration and temperature are required; these lead to higher
peel loss, and as a result the peeled tomato quality suffers since the
vascular veins become more pronounced. If a peeling aid is used in the
caustic, the residue level possibly could be reduced to less than found in
1975, but it was assumed to be impractical to reduce the residue to zero.
Therefore, means were developed in 1976 to minimize peeling-aid residues
by using potential food-additives as peeling aids.
In normal cannery operations, the juice and pulp supplies are interconnected
so that these materials can be shunted among the different sources and utili-
zation points to satisfy changing production requirements. Therefore, in
actual cannery practice recovered peel pulp would be combined with juice
and macerate from other sources before processing into standard products
such as tomato sauce, catsup, fill juice for whole-canned tomatoes, or other
salted products. During acidification and possibly at the rubber-disc peel
removal point, a small amount of water will be incorporated in the peel. This
126
-------
TABLE 1. SUMMARY OF TYPICAL 1975-1976 RECOVERED PEEL PULPS
Recovery of peel (w/w)%
NTSS, (natural tomato soluble
solids), %w/w
Salt, (sodium chloride), g/100 gra
Total Solids, (salt free), % w/w
Vitamin A, (beta-carotene),
I.U./100 gm
Vitamin C, (ascorbic acid),
mg/100 gm
Color Grade, (as puree)
Flavor Grade, (as puree)
Peeling Aid residue, pulp.
ppm (c)
Insecticide residue, pulp
(f) toxaphene ppm
Insecticide residue, skin,
toxaphene ppm
Typical Single-Strength Tomato Pulps
1975
96.7%
5.3
3.-
5.6
665
nil
A-C
C
300
0.4
5-60
1976
95%
5.2
1.1
5.9
(b)
nil
A
A-C
30
trace
34
Conventional
Processing
(a)
n/a
5.4%
0.08
5.71
516
10.7
A
A
0 (d)
trace
7 (e)
trace is positive amount less than 0.08 ppm.
n/a not applicable (no current commercial recovery)
(a) industry 4-yr averages except Salt which is 2-yr average.
(b) not analyzed.
(c) sodium 2-ethylhexyl sulfate in 1975, octanoic acid in 1976.
(d) no peeling aid used in conventional juicing/pulping.
(e) pomace from juicing, seeds with proportionality less skin than in
peel-pulp recovery.
(f) tolerance is 7.0 ppm in canned and fresh tomatoes.
127
-------
water must be removed, either through subsequent concentration, such as for
tomato sauce, or during the evaporation that occurs from holding tanks.
This water can be removed after combining with other juices and pulps.
There seems to be little justification for keeping recovered pulp isolated
for use in a special product since economically there is an insufficient
quantity of recoverable pulp to establish a separate processing line.
Tomato Washing and Sorting
The 1975 cannery peel had low mold, insect fragment, and bacterial counts
that were well within regulatory tolerances. This showed that the cannery
(Tillie Lewis Foods, Plant W) had an excellent washing and sorting system.
In 1976, tomatoes were received before initial commercial processing;
experimental sorting was varied and the results showed that hand sorting,
or the equivalent, is essential to removing mold prior to peeling. Sorting
was varied from zero to 20% of the peeled tomato weight; the higher sorting
was necessary when the California State Grade Certificate showed 3% mold.
While the Grade Certificate is an indication of mold, the tomatoes were
graded about 24 hours prior so the actual mold count may be higher at the
time of processing. An alternative to hand sorting is the use of high-
pressure water sprays (70-120 psig) as practiced for a number of years by
most canners to remove broken and moldy tomatoes.
Peel Acidification
Early and rapid acidification is necessary because the tomato is more stable
at pH 4.2 than at pH 11-12; at the high pH, both color and flavor deteri-
orate. A rapid decrease, in less than 10 seconds, from pH 11 to 4.2 is
necessary because off odor was sometimes observed when the acidification
paused at pH 6-9 for several minutes. Acidification was performed either
before or after the skin was separated from the pulp. Since recovery was
96% or more, there was little reason to be concerned with acid economy by
acidifying after the skin removal. Therefore, the Peel Tank (Figure 1) is
the preferred location for acidification because it provides a quick pH
response and is an easy place to control pH.
Pulp Extraction
With the Extractor (Figure 1), the best pulp recovery and skin extraction
was obtained with a 0.030-in. screen and 0.5-in. paddle-screen clearance.
Larger screen sizes allowed too many skin particles and seed fragments from
broken tomatoes to remain with the pulp. The 0.5-in. clearance was necessary
because closer clearances, as in normal cannery practice, would grind the
skin and incorporate it into the pulp. Caustic action has already loosened
the tomato cells so a wider clearance is necessary. Skin and seed fragments
are undesirable because they are graded as defects in products; also, the
waxy skin layer carries the insecticide. Typically, the recovered pulp
contained 0.4 ppm toxaphene and the skin waste had 5-60 ppm.
128
-------
Peeling Aids
During the Spring 1976 laboratory peeling study, the C and C saturated,
monocarboxylic fatty acids were the most effective peeling aids. The 150°F
pretreatment was as effective, or more so, as using the peeling aid directly
in the caustic applicator. For easy-peeling tomatoes, such as those suitable
for steam peeling,, it was possible to peel without caustic by using a 1-min.
immersion in a 150°F aqueous bath containing 0.2% w/w octanoic acid. While
it may be convenient to apply the peeling aid with caustic, there is no
inherent reason why a peeling aid requires the same application temperature,
pH, and immersion time as the caustic. Peeling aids are commonly referred
to as "wetting agents" but the most effective ones may do more than reduce
the interfacial tension between the tomato surface and the caustic. Some
wetting agents, such as sodium oleate or sodium lauryl sulfate, will show
high-wetting improvement, but they will have little effect on peeling,
whether applied as a pretreatment or directly in the caustic bath. Others,
such as sodium 2-ethylhexyl sulfate and sodium mono- and di-methyl naphthalene
sulfonates, perform better when applied directly in the caustic bath than
when used as a pretreatment. The most effective peeling aids appear to react
chemically and (or) to disrupt the cell structure and allow enzyme action.
This is illustrated by peeling tomatoes with only an acidic (about pH 3.6),
aqueous solution of octanoic acid at 150°F. While octanoic acid performed
best among the candidates, the C, to C.._ saturated monocarboxylic acids
showed the most promise. The time was limited and an extensive pursuit of
the ideal peeling aid was not feasible. Octanoic acid occurs naturally
in coconut oil, is readily available commercially in a food grade, and
is priced similar to the currently used peeling aids. A food-grade peeling
aid should be biologically metabolized in predictable fashion by both
humans and animals or microorganisms associated with man. Octanoic acid
fits these requirements.
Carboxylic Acid Peeling
The carboxylic acid peeling was done at about pH 3.6 with a 150°F aqueous
solution containing 0.2% octanoic acid with a one to three minute immersion.
This completely peeled the tomato varieties Tropic, Walter, Roma-VF, and
VF-145-21-4 (this last one contains a uniform-ripening gene). For the VF-
145-7879, 134, 198, and 13L, which are typical California processing toma-
toes, the skin was loosened and peeling aided, but a subsequent caustic
application was needed. With octanoic acid peeling, the peeling loss
averaged about 5% as compared to 12% for caustic using commercial peeling
aids. The difference was visually dramatic because caustic peel was red
due to the adhering pulp, whereas the octanoic acid peel was a translucent,
pale yellow because no pulp adhered.
Peeling Aid Pretreatment
Pretreatment temperature was varied from 75°F to 210°F. Initially, 150°F
was chosen so as to be below enzyme-inactivation temperature. Experimenta-
tion showed that below about 140°F the peeling aid pretreatment was less
effective or required long immersion, such as up to 10 minutes. Above
about 170°F, even with a short dip, the tomatoes became increasingly soft
and the peel loss increased. Overall, the 150°F temperature was best with
129
-------
the range of 140 to 160°F being practical. Even a 150°F plain (100%) water
pretreatment usually showed some peeling improvement, but a peeling aid is
definitely better.
Modified-Caustic Pilot Peeling
The 1976 pilot peeling at the cannery was mostly with the tomato Variety UC-
134, which usually was the only tomato available and is a more difficult
tomato to peel than the more prevalent VF-145-7879. When the VF-145, 198,
and 13L were used, they responded similarly to that of the 134 with respect
to peeling-aid pretreatment and residue in the recovered pulps.
When comparing peeling methods with a pilot line, it is possible to
simulate commercial conditions but nearly impossible to duplicate scale
because tomatoes change with each truck load and even within the load. Also,
loading pilot equipment and duplicating caustic solutions can be quite
different. For example, the usual cannery caustic applicator takes about
one week to reach equilibrium between caustic and disintegrated tomato
solids. However, the experimentation necessitated changing the caustic
solution daily. The rubber-disc peeler is quite sensitive to the degree
of loading because peel removal depends not only on the discs contacting
the tomato but also on the inter-tomato rubbing and contact. The peeler
was operated at about 1-t/hr so as to use nearly the same tomatoes from a
three-hour period for one set of experimental-peeling variables. From a
3-hr supply period, three peeling comparisons were normally made: (1)
plain caustic bath, (2) caustic bath containing peeling aid, and (3) peeling-
aid pretreatment followed by a plain caustic immersion. Times, temperatures,
and loading were constant during these three variations.
Effluents & Wastes
One of the prime considerations when initiating the project was not only to
reduce liquid and solid effluents in terms of caustic, BOD, and COD, but to
avoid creating new ones; this was successfully managed. There was no conti-
nuous liquid discharge except from washing the tomatoes, and this is present
commercial practice. There was a carryover from the Pretreatment to the
Caustic Applicator, and from the Applicator to the disc peel remover. These
carryovers were food-grade materials, not inedible peeling aids nor efflu-
ents. Since 96% of the peel is recovered as pulp, the normal peel effluent
was drastically reduced. The skin, seeds, and fibers separated by the
Extractor are normal processing wastes. Since these wastes have been re-
acidified, they are more acceptable than with the current caustic peel for
disposal on agricultural land or into a municipal waste treatment plant.
The Pretreatment bath liquid was not operated for extended periods and
through BOD and COD measurements were made, these likely do not represent
what might be experienced under commercial conditions. Some canners
currently operate their caustic applicators the full 3-mo. processing
season without changing solutions, others change the caustic once a week.
The pretreatment aqueous solution of octanoic acid is biodegradable
whereas some of the current peeling aids are not. After peel removal,
all canners rinse or flume the peeled tomatoes, and this practice would
be continued with this 1976 modified-caustic peeling. Therefore, this
130
-------
modified caustic peeling and peel recovery would decrease current peel
discharge by 96%, and the discharge-would have an improved pH character.
Product Mold, Insect Fragments, & Bacterial Counts
In 1975 the peel received from the Tillie Lewis peeling operation was almost
devoid of mesophilic and thermophilic bacteria. There was a slight increase
during the peel-pulp recovery, but the counts were very small in comparison
to those found in commercial operations at similar processing steps.
Therefore, it can be assumed that the caustic-peeling operation has a strong
bactericidal effect on the peel. Also in 1975, the mold and insect-fragment
counts on the canned recovered pulp were uniformly low and reflect a thorough
washing and sorting system at the Tillie Lewis cannery. Therefore in 1976,
the experimental tomato washing was held constant, but the hand sorting was
varied from zero to 20% of the peeled-tomato weight. The 1976 mold counts
showed that hand sorting, or the equivalent, would be essential to maintain
the mold level within tolerance. Truck loads with a California State Grade
Certificate showing 1% or less mold required minimal sorting on loads above
3% mold, up to 20% of the tomato weight might need to be removed to be
within tolerance on the recovered pulp. This 20% included both tomatoes
showing mold and broken tomatoes which would disintegrate in the bath and
not yield properly peeled whole tomatoes. Since there is normally about
a 24-hr lapse between the time of picking and grading and actual processing,
the mold count will increase. All this points to the necessity for an
efficient sorting of tomatoes prior to peeling if the recovered peel pulp
is to be usable. The results indicate that good manufacturing practices
will be necessary to control the mold and insect fragments.
Product Color, Flavor, & Vitamins
Product color and flavor differed between 1975 and 1976, and this illustrates
how the peeling operation affects the recovered pulp. In 1975 the Caustic
Applicator liquid overflow went into the peel rather than to waste disposal;
it was a local situation which could not be altered at the time, but the
results show that recovered pulp could be food grade even under adverse
conditions of excessive (3-fold) caustic. The color and flavor (Table 1)
were degraded by caustic as were some amino acids. Amino acids were used as
indicators of processing severity. The argininerhistidine ratios were 0.4
in 1975, 1.1 in 1976, and 1.3 for conventional tomato pulp. Arginine is
more susceptible to caustic degradation than histidine. Flavor also
suffered when the pulp was exposed to excess caustic as seen in Table I.
Ascorbic acid is so easily decreased in the presence of hot caustic that
it was not present in pulps from either year. Beta-carotene is quite stable,
and since the outer surface of the tomato is richer than the whole tomato
in this provitamin, the recovered pulp had a higher carotene value than
conventional tomato pulp. In general, the 1976 recovered pulps exhibited
improved color and flavor and reduced peel-aid residues compared to the
1975 pulps.
131
-------
Consistency, Viscosity, and Pectin
In this pulp, there was a desirable above average number of tomato whole
cells which give the pulp a heavier, thicker appearance compared to
conventional tomato pulps. In general, this recovered pulp shows the
consistency and texture of a good cold-break processed pulp. Serum
viscosity was less than those of conventional pulps which indicates
degradation of the pectin as would be expected from the caustic exposure.
Evidently the consistency of this peel pulp is due to undissolved solids
(cells, fibers, etc.). Insoluble solids are the primary contributors
to tomato-product consistency, but pectin is important because it serves
to hold the insoluble solids in suspension and reduce the tendency for
syneresis. This recovered peel-pulp would be appropriate for pizza, sauce
or soups, which normally may use a cold-break pulp because it is desirable
to thicken with starch. For other sauces or pulps where a hot-break
material is normally used, combining recovered pulp with conventional
pulp in a ratio of 1:3 will result in a material with the consistency
and character of a hot-break pulp.
U.S. Standards of Identity
Since the proposed process modifications and materials may or may not be
totally covered by the existing U.S. Food & Drug Regulations, a letter was
sent to the FDA Bureau of Foods requesting their judgment on these main
concerns: (1) utilization of pulp derived from caustic peeling, (2) acidi-
fication of the caustic peel with food-grade hydrochloric acid, (3) use of
food-grade octanoic acid as a peeling aid with residue present in the
recovered pulp, and (4) the labeling requirements when recovered pulp is
utilized. Whether a commercial installation can be made under the existing
regulations or whether further technical and regulatory considerations are
needed will depend on the response to this letter. Pertinent regulations
are 21 CFR 53.10-.40, tomato products; 21 CFR 121.1070, food-grade fatty
acids; and 121.1091, chemicals used in lye peeling.
Economics
The economics for a cannery peeling operation of 40 t/hr are summarized in
Table 2. Such a peeling operation might use two lye applicators, such as
the FMC Hi-Ton, and either two FMC PR-20 Peel Removers or four Magnuson
Model C Peel Scrubbers. The projected costs and savings are based on a 12%
recoverable peel loss, 60 days operation per year, 16 hrs of peeling each
day, and a $50/t pulp value. Capital and operating expenses include only
those directly associated with pretreatment and pulp recovery, not the
balance of the peeling process that is presumed to already exist. The
capital cost could easily be greater or smaller, depending on whether
existing equipment and utilities are readily adaptable or entirely new
equipment would be needed. One example is that the pretreatment could be
applied with an existing flume or it might require a special, positive-flow
pretreatment such as that used in the pilot installation. Undoubtedly
132
-------
TABLE 2. PROJECTED ECONOMICS OF A COMMERCIAL INSTALLATION
Based on: 40 t/hr tomatoes x 12% peel
=4.8 t/hr peel 96% recovery
= 4.608 t/hr pulp recovered.
60 dy/yr x 16 hr/dy = 960 hr/yr
Recovered Peel-Pulp Value
Based on $50/t gross value of pulp.
Fixed Cost, installed Capital Equipment.
Depreciated in 1 yr, 960 hr.
Pretreatment (circulation, temp, control) -
Chemical Supplies (tanks, pumps, piping) -
Extraction (pulper, piping)
Acidification (pH control, agitator) -
General (utilities, piping, etc.)
Contingency
15,000
16,000
14,000
15,000
35,000
20,000
(installed total) -115,000
Variable Costs
Direct Labor (operator, cleanup, QC, supv.) -
Startup labor, 1st year
Indirect labor (mechanic, clerk) -
Superintendence -
Utilities, steam, 7,230 Ib/hr
water, 10 gpm -
electricity, 78 kw
Chemicals, fatty acid, 1.33 Ib/hr
hydrochloric acid, 150 Ib/hr
Maintenance Supplies, 5%/yr of capital -
Miscellaneous (operating & cleanup supplies)-
12.48
24.96
3.13
0.94
10.84
0.07
1.57
1.33
6.00
5.92
1.00
Value
Costs
NET RETURN ON PEEL PROCESSING
Saving on Caustic Peel Disposal ($2.50/t)
OVERALL RETURN ON PEEL PROCESSING
1st Year
Operation
$/hr
+ 230.40
- 119.79
- 40.57
0.94
12.48
7.33
5.92
1.00
+ $230.40
- 188.03
+ $ 42.37
+ 12.00
+ $ 54.37
or
$52,195.20
1st year
2nd Year
and
Thereafter
$/hr
+ 230.40
0.00
- 15.61
0.94
- 12.48
7.33
5.92
1.00
+ $230.40
- 43.28
+ $187.12
+ 12.00
+ $199.12
or
$191,155.20
per year
thereafter
133
-------
different canners will use pulp values different than the $50/t which was
based on the approximate value of delivered fresh tomatoes before proces-
sing.
SUMMARY
The objective of this evaluation and development was to determine the
"Commercial Feasibility of Recovering Tomato Processing Residuals for Food
Use". The 1975 work was an evaluation and characterization of peel from a
typical commercial peeling operation. For 1976, the caustic peeling
process was modified to reduce the unacceptable peeling-aid levels found
during 1975. The commercial peel-pulp recovery operation envisioned
would have a tomato pretreatment with a food-grade fatty acid as used
in 1976 and the acidification and pulp extraction features used in 1975.
The recovered 1976 pulp was Grade A, and with a capital investment which
can be depreciated ciated in less than one year, the pulp recovery has
a net positive value of possibly $187/hr of operation after the first
year. As of now, pulp recovery from tomato peel appears to be both
technically and economically feasible.
ACKNOWLEDGEMENTS
The authors appreciate the support given to this project by the tomato pro-
cessing industry; not only did it ease the burden, but tasks and judgments
were often shortened and made possible in a few months which otherwise might
have required several years. In particular, we wish to thank Tillie Lewis
Foods and Hunt-Wesson Foods at whose canneries the field work was performed.
This allowed a practical approach which could not be duplicated in a labora-
tory or pilot plant. The U.S. Environmental Protection Agency supported in
part this experimentation, and this allowed the work and development to be
carried out in the short period of two years.
REFERENCES
GENERAL
1. Schultz, W. G., Graham, R. P., and Hart, M. R. Pulp recovery from
tomato peel residue. Proceedings of 6th National Symposium on Food
Processing Wastes, Madison, Wisconsin, April 5-11 (1975).
2. Hart, M. R., Graham, R. P., Williams, G. S., and Hanni, P. F. Lye
peeling of tomatoes using rotating rubber discs. Food Technology
28:38 (1974).
3. Schultz, W. G., Graham, R. P., Rockwell, W. C., Bomben, J. L.,
Miers, J. C., and Wagner, J. R. Field processing of tomatoes, Part 1,
process and design. J. Food Sci. 36:397 (1971).
134
-------
4. Miers, J. C., Wagner, J. R., Nutting, M-D., Schultz, W. G., Becker, R.,
Neumann, H. J., Dietrich, W. C., and Sanshuck, D. W. Field processing
of tomatoes, Part 2, quality and composition. J. Food Sci. 36:400
(1971).
5. Ostertag, R. and Robe, K. Waterless peel removal. Food Processing
36(1):60 (1975).
6. Schultz, W. G. et al. Commercial feasibility of recovering tomato pro-
cessing residuals for food use. Report for 1975, Interagency Agreement
EPA-IAG-D5-0795. Western Regional Research Center, ARS, USDA, 800
Buchanan St., Albany, CA 94710.
U.S. Government Code of Federal Regulations.
U. S. Government Printing Office, Washington, D.C. 20402.
21 CFR 52.5081-.5091. Canned tomato pulp grades.
21 CFR 53.10, .20, .30, .40. Standards of identity, tomato catsup, pulp,
and paste, and canned tomatoes.
21 CFR 121.1070. Fatty acids.
21 CFR 121.1091. Chemicals used in washing or to assist in the lye peeling
of fruits and vegetables.
Analytical Procedures
Consistency
Natural Tomato Soluble
Solids (NTSS)
Mold and Insect Fragments
Peeling Aid,
Fatty Acids
Peeling Aids,
(Na 2-ethylhexyl
sulfate, sodium
mono- and di-methyl
naphthalene sulfonates)
National Canners Assoc. Laboratory Manual
for Food Canners and Processors, Vol. 2.
Avi Publ. Co., Westport, Conn. 294-6 (1968).
ibid, p 560.
ibid, pp 300-316, 324-325.
Modification of: Macpherson, J. K. and
Buckee, G. K. Estimation of free fatty
acids (C to C ) in wort and beer.
J. Inst. Brew., ^0(6): 540 (1974).
Intercontinental Chemical Company. "Photo-
metric Determination of Con-0 Peel-Eze 1497."
Sacramento, Calif.
135
-------
Pesticide (Toxaphene)
Quality Factors (color,
defects, flavor, odor)
Salt (Total Chloride)
Total Solids
(Percent Moisture)
Viscosity
Vitamin A (Carotene)
Food and Drug Administration. Pesticide
Analytical Manual, Vol. 1. U.S. Dept. of
HEW, Washington, B.C. (1971) §212.13a(2)
U.S. Dept. of Agriculture. United States
Standards for Grades of Canned Tomato
Puree (Tomato Pulp).
National Canners Assn. loc. cit. pp 291-
292.
Horowitz, William (ed.). Official Methods
of Analysis of the Association of Official
Analytical Chemists. AOAC, Washington, D.C.
(llth ed., 1970) p 559.
ibid, p 369.
ibid, pp 769-771.
Reference to a company and/or product named by the Department is only for
purposes of information and does not imply approval or recommendation of the
product to the exclusior of others which may also be suitable.
136
-------
WASTE REDUCTION BY PROCESS MODIFICATION IN SWEET CORN PROCESSING
by
G. H. Robertson*, M. E. Lazar*, J. M. Krochta* and D. F. Farkas*
INTRODUCTION
The processing of sweet corn results in the generation of large quantities
of liquid and solid waste. Liquid wastes contain soluble sugars and starches
and are responsible for average wastewater loadings of 27 Ibs of BOD (1)
per ton of corn in husk. Nearly 2.6 million Ibs of sweet corn (in husk)
are processed (2) hence, 70 million pounds of BOD are produced annually.
Solid wastes are composed of husk, cob, and edible corn which are not
removed from the cob during cutting and constitute up to 2/3 of the weight
of the starting raw material.
The predominant form of preserved sweet corn is the "whole" or cut style,
which constitutes 64 to 70% of all sweet corn and is produced by forcing
husked ears of sweet corn against revolving or stationary sets of knives.
Furthermore, most of the liquid wastes indicated above are attributed to
processing this style of corn and occur during the contact of the cut-corn
surface with water. Transfer of starch and sugars from the kernel to the
blanch, wash, and flume fluids occurs by diffusion and by bulk mixing
between the aqueous medium and the semi-solid endosperm.
The USDA Western Regional Research Center has conducted studies to ascer-
tain whether sweet-corn kernels can be removed from the cob as intact units
(unit kernels) which lack a cut surface, and whether this modification
would provide sufficient waste reduction and yield improvement incentives
for further development of new processing methods. Manual experiments
using market corn were conducted in 1975 (3). These indicated 70% decre-
ments for washing and blanching effluents and 20% increments for recovery
of kernels on a per kernel basis.
The work described here was conducted to continue effluent and yield com-
parisons for intact kernels and cut kernels produced from commercial varie-
ties of sweet corn harvested at their processing maturities, and to evaluate
three experimental processing methods relative to their potential for a full-
scale application.
EXPERIMENTAL METHODS
Sweet corn for these studies was grown during 1976 on irrigated land in
Gilroy, CA, by the Del Monte Corporation. The varieties Golden Jubilee and
Stylepak were grown as representative of commercial processing sweet corn.
The glumeless variety Golden Happiness was also grown.
*Western Regional Research Center, U.S.D.A., ARS, Albany, California
137
-------
When the corn attained processing maturity (monitored by moisture content of
cut kernels), processing was begun (Table 1). Sweet corn was manually
harvested, transported by truck to Albany, CA, composited to eliminate
picker bias, divided into sublets of 35 to 70 Ibs, and husked in a commercial
husking machine (Food Machinery Corporation). The ears were then trimmed
to exclude insect, bird, and tnicrobial damage at the tip end. The sublets
of trimmed corn were then delivered to the appropriate kernel generation
stations.
TABLE 1. PROCESS SEQUENCE FOR SWEET-CORN EXPERIMENTS
Step Number
Action
Weight
Kernel
Sample
Effluent
Sample
1.
2.
3.
4,
5.
6.
8.
9.
10.
Harvest
Transport
Composite X
Husk X
Trim X
Generate Kernels X
(Four methods)
Clean X
(Two methods)
Steam blanch X
Air Cool X
Freeze X
Cut kernels were produced using a commercial rotating cutting machine
(Food Machinery Corporation). The machine was set to give approximately
the depth of cut obtained in commercial practice. Following production
of kernels the juices adhering to the inside of the cutter were washed
with a fixed volume of water and collected. This effluent was considered
as part of the wash effluent since in a continuous process it would normally
be included with the wash effluent.
Intact kernels were generated in three ways. The first alternative was to
use the conventional cutting machine by setting the depth of cut at its
maximum or deepest cut. Cutter washing was conducted as above. The second
alternative was to produce kernels by the "hole-saw" technique. Here, ears
were processed in a fixed diameter cutting or sawing machine sized to the
138
-------
ear diameter to produce a mixture including intact kernels, kernels
with attached cob fragments and saw dust (3). These kernels were later
screened in a batch reel separator to eliminate loose fines. The proto-
type hole-sawing machine was provided with a rotating sawing element into
which pre-sized ears were manually fed. A third alternative was to produce
intact kernels by rubbing, pressing, or plucking kernels row-by-row from
longitudinal half-sections of ears produced by mechanically splitting.
In the prototype for this process manually supported ear sections were
placed against the surface of a continuous, moving, textured silicone
rubber belt. An air stream heated to 350 to 400 F was directed against
the belt surface to evaporate juices from prior contacts so as to maintain
contact friction.
Normal-depth cut kernels, deeply cut kernels, hole-sawed and screened ker-
nels, and pressed kernels were then washed in water using a pilot scale,
shaker washer (A.B. McLauchlan Co., Inc.). The water used in washing was
collected and its volume recorded. In some cases a brine flotation in
5% NaCl was applied to eliminate excessive cob fragments. Cleaned ker-
nels then were blanched for 3.0 minutes on trays in a continuous steam
blancher operating at 210 F. Biancher effluent was collected during and for 12
minutes after blanching, its volume measured and recorded. Finally, the
blanched kernels were air cooled in forced air, and then frozen on trays in a
cross-flow, air-blast freezer operating at -20 F and at 700 fpm.
Material balances and sample analyses. The mass of each sublot of corn was
measured before husking and after each step indicated in Table 1. Weighed
kernel samples were drawn at the points indicated in Table 1. A 70-90 gm
sample was drawn for moisture analysis and a 130 to 200 gm sample was drawn
for analysis of kernel characteristics. Material balances were corrected
for sample weights, when appropriate. Liquid samples were drawn from the
cutter wash effluent, the kernel washing effluent, and the blancher conden-
sate. Liquid samples were analyzed for total organic carbon (TOC), chemical
oxygen demand (COD), total solids (TS), soluble solids (SS), and biological
oxygen demand (BOD).
Characterizations were made by visually segregating kernels into groups
according to whether they were cut kernels, intact kernels, intact kernels
with adhering cob tissues, or completely smashed kernels. Cob fragments
and miscellaneous debris were also identified here. Each group was weighed and
expressed as a percentage of the total sample weight. A classification of
each group into smaller subgroups according to the depth of cut and the type
of cob fragments, adhering to unit kernels will be reported later.
RESULTS
Effluent produced during the processing of sweet corn kernels from each of
the four kernel generating methods are summarized in Tables 2. and 3. These
data and those which follow represent 2 Stylepak, 2 Jubilee, and 1 Golden
Happiness harvest. The data are shown plus or minus one standard deviation.
Moreover, the basis of comparison is an equal mass of frozen product.
139
-------
TABLE 2. LIQUID EFFLUENT OBSERVED DURING SWEET-CORN WASHING
Kernel Style
Intact (Pressed)
Cut (Normal)
Cut (Deep)
Intact (Hole Saw)
COD
(pph)
0.62 +_ 0.51
3.53 + 0.91
2.12 + 0.43
0.87 + 0.16
TOG
(pph)
0.27 +_ 0.18
1.33 +_ 0.38
0.79 +_ 0.17
0.31 +_ 0.06
SS
(pph)
0.11 +_ 0.07
0.73 +_ 0.13
0.36 _+ 0.07
0.17 +_ 0.06
TS
(pph)
0.79 +_ 0.46
2.96 +_ 0.77
1.83 _+ 0.40
0.88 +_ 0.03
Basis: 100 Mass Units of Frozen Product.
TABLE 3. LIQUID EFFLUENT OBSERVED DURING SWEET-CORN BLANCHING
Kernel Style COD TOG SS TS
(pph) (pph) (pph) (pph)
x
Intact (Pressed) 0.059 +_ 0.035 0.022 +_ 0.013 0.0035 +_ 0.0017 0.073_+0.037
Cut (Normal) 0.148 + 0.040 0.056 + 0.015 0.0154 +_ 0.0030 0.095+0.0726
Cut (Deep) 0.155 +_ 0.040 0.060 +_ 0.016 0.0137 +_ 0.0059 0.173+0.0529
Intact (Hole Saw) 0.078 +_ 0.041 0.029 _+ 0.015 0.0043 +_ 0.0012 0.099+^0.040
Basis: 100 Mass Units of Frozen Corn
The characteristics of the kernels produced by each method are summarized in
Table 4. This gross characterization into groups of cut kernels, intact
kernels, and intact kernels with adhering cob tissue indicated the overall
suitability of the material as well as the requirement for extensive cleaning
or separation.
The material balances for process runs in which all four methods were exa-
mined are shown in Table 5. All of the data except those in the last
line which are labeled "Frozen (gross yield)" are yields which include
cob fragments as well as kernels with adhering cob tissue. The last line of
entries labeled "Frozen-useable" is corrected by applying the percentage
140
-------
TABLE 4. MEAN KERNEL WEIGHT PERCENT DISTRIBUTIONS FOR FRESHLY PREPARED MIXTURES
Kernel
Mixture
Cut (Normal)
Cut (Deep)
Intact (Hole Saw)
Intact (Press)
Cut
Kernels
80
41
22
0
Intact Intact Kernels
Kernels with Attached Cob
12
37
52
95
0
13
23
3
Smashed
Kernels
3
4
1
1
Cob
Fragments
6
4
2
1
screened
-------
TABLE 5. MEAN EXPERIMENTAL YIELD DURING PROCESSING OF SWEET-CORN
(71.8 + 1.3% MOISTURE)
Action
Mass percent recovery
Husked
Trimmed
Prepared
Kernels
Washed
Blanched
Air Cooled
Frozen
66.1 + 1.9
64.1 + 1.3
Cut Cut Intact Intact
(Normal) (Deep) (Hole-saw) (Press)
38.1 + 3.1 46.3 +_ 3.0 46.4 + 3.3 40.9 +_ 2.3
36.3^3.8 47.0^3.5 49.9+4.6 43.2^2.6
35.1 + 3.8 44.2 +_ 2.9 46.8 +_ 5.3 42.8 4- 2.5
33.1^3.6 41.6_+2.9 44.0^5.3 40.4+2.1
32.0 + 3.8 39.6 + 2.8 42.8 + 5.2 38.8 + 1.9
(Gross Yield)
Frozen
(Useable Yield)
31.0 + 3.8
34.9 + 4.1
34.0 + 4.0 37.7 + 2.4
142
-------
of useable corn determined from individual classification analyses similiar
to those described above. This line entry is assumed to approximate the
product which would be achieved after an ideal flotation separation.
DISCUSSION
The purpose of this investigation was to assess the merits of the alter-
ative kernel producing methods in terms of effluent generated and yield
obtained. Analysis of the data presented above indicated that the most
likely candidate for further development was the intact kernel produced by
the pressing alternative. The factors entering this decision are described
below.
Effluent data based on crude yields (i.e. yields for kernel mixtures of
suitable corn with defective corn as produced in testing without a satisfac-
tory flotation separation) indicated that the pressing process would result
in the least generation of COD, TOC, SS, or TS of any of the four alterna-
tives. Favorable reductions were also obtained for the hole-sawed intact
kernels. However, the effluent from deeply cut kernels was more like that
from normally cut kernels, the small difference being attributed to the
inclusion of more kernel mass in the deep cut sample and to the inclusion
of more intact kernels.
Organic loading in blancher effluents were much lower than in washer effluents
since the blanching operation represented the second contact with water
(See Introduction) and since steam blanching is inherently a low effluent
method. However, the relative order of effectiveness in reducing waste
was the same as observed for washing.
The reasons for the different organic loadings for each of the different
kernel producing methods can be deduced from the kernel distributions shown
in Table 4. Clearly, the loading is proportional to the proportion of the
sample containing cut kernels. Perhaps a lower loading than was realized
would have been expected from the unit-kernel, but since some kernels are
broken before and during the separation by pressing, and since the kernel
base is not a con.^letely sealed unit at this stage of its maturity, this
expectation was not realized.
The quality of each mixture, ie. the relative amounts of corn which must be
separated and/or upgraded, can also be seen in the classification data. For
instance, the pressed intact sample contains small accounts of kernels with
adhering cob tissue (about 3%) whereas the cut (normal) contains about 3%
of smashed kernels. In a conventional line both of these, or a portion of
them would be separable by a flotation. Much larger amounts of both of these
defects occur in the deep-cut and hole-sawed product making a separation
and/or reclamation imperative.
The mean yield history for each style kernel is shown stepwise in Table 5;
Generally, there is a loss of yield for each kernel mixture due to leaching
and dehydration during the process sequence. However, this loss is greater
for the cut kenels than for either of the intact kernel mixtures and reflects
the greater susceptibility of the cut surface to losses.
143
-------
A comparison of freshly prepared cut (normal) and intact (pressed) kernel
yields shows that the intact yields were higher by a factor of 3.2% for
Stylepak, 7.8% for Golden Jubilee, and 17.8% for Golden Happiness corres-
ponding to cutting recoveries of 40.8, 37.1 and 34.4% respectively.
The value of the yield increase factor depends on the value of the cutting
yield, ie., the higher the cutting yield, the lower the percentage
increase factor. However, large percentage increase factors do appear
to be realistic since normally cutting yields fall with the range of 28 to
40% (4) depending on variety and maturity. Haber (5) reported an average
recovery of 33.7% for 9 commercial varieties over an 8-year study
period. Moreover, higher yield factors might have been obtained for intact
kernels if all of the available kernels had been recovered. Tightly
packed kernels near the butt end, ruptured kernels, and immature kernels
at the tapered end were not recovered effectively.
Judging from the results shown in Table 5 it would appear that the greatest
yield could be attained by either flh'e hole-saw or the deep-cut options.
However, when these yields are corrected by applying experimentally
determined percentages of useable kernels, the balance shifts to favor
the pressed kernels. In an attempt to take advantage of the hole-saw
yield potential we proposed thaf. the off-grade kernels of either deep-cut
or hole-saw mixtures might be upgraded to suitable kernels by mechanical
means. Many different approaches including abrasive milling of fresh,
blanched and frozen hole-sawed kernels were applied, but were not successful
since failure of the sensitive kernel pericarp usually occurred before
failure at or near the kernel abscission layer. Orientation techniques
were investigated which would be applied along with a cutting or pinching
action to release the cob fragment. The large number of kernels and
the widely different shapes and sizes which would have to be processed
in this fashion precluded development along these lines.
Furthermore, when useable yield factors were applied to the effluent
data of Table 2 to assess the predicted effluent in terms of useable
kernels, differences obtained between the different styles are greater
and the balance shifts farther in favor of the pressing option. For
instance, the COD values for cut (normal), cut (deep), intact (hole-
saw), and intact (pressed) kernels became respectively 3.64, 2.42, 1.10,
and 0.64.
Real separations via density flotation are not as sharp as the ideal
separations utilized in the arguments described above. Preliminary
flotation data performed on freshly prepared and washed intact (hole-saw)
kernels indicated that the percentage of off-grade kernels can be reduced
to acceptable levels, but that this is achieved at the expense of loss of
yield of edible kernels. For instance, to reduce the proportion of off-
grade kernels produced by hole-sawing from the 20% level to the 3% level
as found in freshly prepared intact (pressed) kernels would require immer-
sion in an 11.0% NaCl solution (or its density equivalent) and would result
in the loss of nearly 30% of the useable sweet corn. Because of this poor
separation factor, there is additional "strong incentive to utilize the
preparation sequence which produces the cleanest sample to begin with,
i.e., the pressing procedure.
144
-------
The yield and effluent inducements for substitution of the unit or intact
sweet corn kernel are augmented by consideration of the organoleptic
acceptability of intact kernels. For instance, untrained panels which were
served frozen and cooked samples of intact and cut kernels prepared from
the same harvest, consistently expressed a preference for the intact kernel
of 0.5 to 0.8 scoring units on a 1 to 9 hedonic preference ranking. This
preference appears to be due to improved flavor and mouth feel of the intact
kernel. Indeed, the product flavor resembles that of corn-on-the-cob.
CONCLUSIONS
Based on the experiments described above, it was concluded that:
1. Intact kernel mixtures provided for substantial effluent reductions
over conventionally cut kernel mixtures.
2. Intact kernel mixtures produced by pressing provided for substan-
tially greater yields than normally obtained during cutting.
3. The intact kernels produced by pressing provided for the highest
yield with lowest effluent of the four experimental methods.
and
4. Intact kernels could have greater acceptability than cut kernels.
ACKNOWLEDGEMENT
The authors express their appreciation to Harold Thompson of the
Environmental Protection Agency, and John Farquhar of the American
Frozen Food Institute and their sponsoring agencies whose financial
support was utilized in this project. In addition, we acknowledge
the assistance of Bill Hagan (Del Monte) for sweet corn growing and
harvesting, Joe Ohler (Libby, McNeill, and Libby) for arranging the
loan of FMC cutter and husker, Christina Merlo (NCA) for analysis of
waste samples, and Joyce Hudson (WRRC) for corn sample analyses.
REFERENCES
1. U.S. Environmental Protection Agency. Pollution Abatement in the
Fruit and Vegetable Industry. Vol. 3. Wastewater Treatment in
the Food Processing Industry. Office of Technology Transfer (1965).
2. Anon. The Almanac of the Canning, Freezing, Preserving Industries.
Edward Judge & Sons, Westminster, MD (1973).
3. Robertson, G. H., Lazar, M. E., Galinat, W. C., Farkas, D. F., and
Krochta, J. M. Unit operations for generation of intact or unit
kernels of sweet-corn. Manuscript in preparation for J Fd Sci.
145
-------
4. Smith, G. M. "Sweet-Corn." Chapter XI in Sprague, G. F. Corn and
Corn Improvement. Academic Press, Inc. New York (1955).
5. Haber, E. S. Variations in yield and cutting percentage of sweet-
corn hybids. Am Soc for Hort Sci 53 302-304 (1949).
146
-------
APPLICATION OF FINE SCREENS
IN THE
TREATMENT OF FOOD PROCESSING WASTEWATER
by
R. Neal*, R. Chaney*, and A. Bubp*
INTRODUCTION
Fine screens are used primarily as a pretreatment device in the treatment
of food processing wastewaters. That is, they perform an operation on
wastewater to make it more suitable for introduction into a more advanced
treatment system. In most food processing screen applications, waste-
waters are discharged to publicly owned municipal treatment systems where
sewer use charges are assessed. Applications in privately owned treatment
systems are usually found at larger plants or in remote locations.
The major reasons for pretreatment include protection of treatment pro-
cesses to insure maximum efficiency, protection of the receiving system
from damage, satisfaction of legal requirements imposed by sewer use ordi-
nances, and reduction of treatment costs due to user surcharges.
Fine screens are presently used in numerous food processing plants to
recover suspended solids from plant effluents and process streams. Fine
screen devices include static screens, rotating drum screens, and vibrating
screens. This paper concentrates on the diverse application and benefits of
the static screen in the treatment of various food processing wastewaters.
A review of applications in the vegetable and fruit processing, meat packing,
poultry processing, seafood processing, dairy, and beverage industries is
included. Static screen acceptance, as illustrated by this paper, is the
result of proven effectiveness, simplicity, and economy.
The static screen, illustrated in Figure 1 (a), is a very simple device. The
wastewater or slurry to be treated is introduced through a headbox to
dampen turbulence and provide even distribution over the weir and screen.
As the wastewater overflows the weir, it accelerates toward the screen. A
hinged baffle reduces turbulence and optimizes flow distribution. Most of
the wastewater is removed on the first slope (25° from vertical) of the three
slope screen as illustrated in Figure 1 (b). Remaining wastewater is
removed on the second (35°) slope and the collected solids begin to roll on
the surface due to their residual kinetic energy. On the final (45°) slope,
*C-E Bauer, Division of Combustion Engineering, Inc. , Springfield, Ohio
147
-------
the solids drain briefly before being displaced by oncoming material.
Screen openings vary from 0. 010 - 0. 080 inch depending on the specific
application. In practice, openings of 0. 020 - 0. 040 inch are most common.
Gravity feed
of liquids/solids I
Self cleaning,
non clogging stainless
steel screen for
continuous dewatering
Headbox
Alternate
feed inlet
Removed or
recovered
solids
Solids
Kb)
Figure 1. Schematic of static screen operation.
VEGETABLE AND FRUIT PROCESSING
The vegetable and fruit processing industry uses fine screens to reduce sus-
pended solids resulting from washing, peeling, slicing, rinsing, and packing.
Essentially all processing plants, regardless of the produce handled, have
two potential screening applications in common. These are wash water
recycle systems and final effluent pretreatment.
The first step in vegetable and fruit processing is washing to remove dirt,
stems, leaves and other residue brought in from the field. Accumulation of
waste material in wash waters requires continual makeup and discharge to
maintain acceptable water quality for effective washing. Fine screens are
used to remove solids so wash water can be recycled. Screening and
recycle results in significant savings through reductions in water consump-
tion and wastewater treatment requirements or surcharges.
148
-------
Processing plant wastewaters, containing skins, seeds, trimmings, culls,
stems, leaves, and dirt, have very high BOD5 and suspended solids
concentrations. Pretreatment is usually required to minimize municipal
sewer surcharges or treatment plant operating problems. Static screens
have been widely accepted for pretreatment of total plant effluents. The
static screen has replaced vibrating and rotating fine screens in many
installations due to lower operation and maintenance requirements.
Typical performance data reported by plant personnel from various instal-
lations are presented in the following table.
TABLE 1. TYPICAL STATIC SCREEN PERFORMANCES
Produce
Potatoes
Mushrooms
Tomatoes
Pickles
Suspended
Raw
(mg/1)
5200
570
Solids
%R
35.0
40.0
84.0
51. 5
B O
Raw
(mg/1)
4860
940
1726
D5
%R
9.7
23. 0
63.0
39.7
Screen
Opening
(inches)
0. 060
0. 060
0. 020
0. 080
Screenings recovered in vegetable processing applications are usually
removed by pet food manufacturers for use in their products. Farmers
also use the waste solids as a cattle feed additive or plow it into fields for
the nutrient value. Solids disposal seldom incurs operating costs to the
processor.
Citrus processing plants use fine screens for pulp dewatering, pectin ex-
traction, press effluent dewatering, and flume water clarification. The
recovered solids are frequently dried and used as a cattle feed additive.
Static screens have been used in wineries to remove 45 - 55% of the avail-
able juice from crushed grapes prior to the pressing operation. By
removing the free juice, a press can be fed thicker grape pulp and brought
to operating pressure quicker. Thus, the total capacity of the pressing
operation is increased considerably.
149
-------
MEAT PACKING
It is common in meat packing plants to provide separate sewers for stock-
pen wastes, manure wastes, grease wastes, clear waters (chilling, con-
densing, and cooling), and sanitary wastes. Newer plants often segregate
further to permit removal of waste materials before dilution with other
wastes. Much meat packing waste is originally solid (meat and fat) or
sludge (manure). It is important to remove the solid material before it is
subjected to turbulence which breaks it down to soluble BOD and colloidal
solids that are more difficult and expensive to treat. As a result, screens
have found wide use in meat packing plants for the treatment of segregated
high concentration waste streams.
Screens are used to remove solids from stockpen wastes, recover paunch
manure, recover hog stomach contents, remove solids from stick water,
separate solids from expellar grease, dewater catch basin sludge, and
recover solids from total plant wastewater.
Prior to slaughter, the animals are held in stockpens. Runways and pens
are hosed down periodically producing a concentrated waste composed of
manure, dirt, straw, corn, and hair. This concentrated wastewater can
be effectively treated with static screens. In a typical application, a 72
inch screen with 0. 040 in. screen openings would handle flows of 600 -
1100 gpm typically producing 12 tons per day of 40% dry weight solids.
Paunch manure (cattle stomach contents) and hog stomach contents contain
fluids, corn, straw, and hair. This material is removed from the
stomach for the recovery of tripe. The material is segregated from the
total wastewater stream due to the extremely heavy loadings and operating
problems it would create for a treatment facility. The wet method of
paunch recovery consists of cutting the paunch open in a water flow.
Static screens with 0. 040 in. openings provide excellent recovery of solids
from the paunch slurry. Flows of 600 gpm can be handled with a 72 inch
unit producing 25% dry weight solids. Recovered solids, a usable feed-
stock by-product, are usually picked up by local livestock producers for
use as a feed additive.
In hog processing, hogs are scalded at 150 - 190° F and dehaired in a
beater-scraper type machine. Hair and scurf, a dandruff-like flake,
build up in the water along with foam produced by gelatin cooked from the
skins. By screening and recycling, the life of scald water can be extended
reducing the total volume of wastewater and conserving heat. A typical
hair recovery application would use a 72 inch screen with 0. 020 in.
openings to handle flows of 400 - 500 gpm.
150
-------
Screens are used to remove solids from stick water in the rendering
process. Stick water is a combination of product and condensation water
produced in steam rendering. The stick water layer from the rendering
vessel is screened prior to evaporation to produce a high protein feed
additive. Solids in stick water are coarse and fibrous in inedible render-
ing, and soft and stringy in edible rendering. Grease normally does not
accumulate on the screen because the stick water is hot (130 - 160° F) when
processed.
An expellar or screw press is used to remove additional grease from ren-
dered solids. This recovered grease contains solids which are normally
removed by gravity clarification before the grease is filtered. Static
screens with 0. 020 in. openings have been used in this application elimi-
nating the settling operation. Due to the fine screen openings (0. 020 in. )
and high grease viscosity, flow rates are low (5-10 gpm for an 18 inch
screen). Even with the low flows, screens are an attractive alternative
for this solids removal application.
Static screens have been used to dewater sludge from catch basins. Flows
of 200 gpm can be handled with a 48 in. x 0. 040 in. screen. Solids
recovered from catch basin sludge are usually collected for landfill disposal.
Screens are used as a primary treatment device for the total wastewater
from packing plants. In many cases solids collected from the total waste-
water stream can be rendered. A typical application using a 72 in. x
0. 040 in. screen would handle a hydraulic loading of 500 - 700 gpm.
Suspended solids removals of 60% or 25,000 Ibs/day of solids (40% dry
weight) can be expected.
DAIRY
The modern dairy uses a considerable amount of water to wash down milk-
ing stalls, pits, lanes, and holding pens. Fine screening this highly con-
centrated manure laden flush water eliminates pipe plugging and secondary
treatment process operating problems. Typical dairy flush water applica-
tions use a 0. 040 in. screen opening to achieve 40 - 45% suspended solids
and 30 - 35% COD removals. Screenings are generally 15% dry weight
solids with 95% volatile material.
POULTRY GROWING AND PROCESSING
Wastewater from poultry growing farms is composed of manure, feathers,
grain, wood chips, and grit from washing the growing barns. Screens are
effective for the treatment of these solids laden wastewaters to remove
151
-------
approximately 50% of the suspended solids prior to biological treatment
processes.
Poultry processing plants generally segregate wastewaters into offal and
feather streams. The offal wastewater normally passes through an offal
separator (coarse drum screen) to remove large renderable material.
The feather laden wastewater passes through a drum or stationary screen
where the feathers, a saleable by-product, are removed. Following these
coarse process screening operations, the two waste streams which still
contain significant quantities of small suspended matter are combined into
a total plant effluent. The Hydrasieve is generally used on the total efflu-
ent to remove additional renderable solids and reduce the load on the plant? s
secondary treatment system.
Poultry processing plant applications normally use a 72 in. screen with
0. 020 in. openings to handle up to 500 gpm. Suspended solids removals of
50% are typical.
Texas A fk M University retently conducted evaluation tests using a static
screen to pretreat chill tank effluents. Test results on this low flow, high
load application indicated 40 - 50% suspended solids and 30% BOD5
reductions.
SEAFOOD PROCESSING
Various seafoods, even though physically quite different, are processed
through a series of similar steps. Fresh catch arriving at the processing
plant is sometimes prewashed to remove loose dirt, silt, scales, and
other debris. The seafood is then conveyed through peelers, sealers, or
shellers to remove protective coverings. The roughly cleaned seafood
then passes through a series of washing operations where waste materials
are removed from the meat. Inspectors then remove culls and remaining
bits of waste solids prior to packaging. Wastewaters from the various pro-
cesses flow to a collection sump from which they are pumped to a static
screen with 0. 020 - 0. 030 in. openings. The screen effectively removes
heavy solids (scales, shells, feelers, viscera, culls, etc.) which produce
high sewer surcharges if discharged to municipal systems or sea gull
problems if ocean discharged.
Due to the high protein and nutrient content of waste seafood solids, they
are usually sold to pet food manufacturers who pick up the waste solids at
the processing plant. Hydrasieves have been used as described in all
types of shell and finfish processing plants.
152
-------
BREWERY, DISTILLERY, AND WINERY
Screens are used in the brewing industry to recover spent grains, dewater
spent hops, and remove labels, glass, caps, cork, etc. from caustic
bottle washing solutions.
Wineries use screens to remove stems from de-steamming operations
and free juice from crushed grapes prior to pressing.
SUMMARY
Static screen application guidelines are summarized onthe following page.
As illustrated here, the fine static screen has gained wide acceptance for
solids removal in the food processing industry due to its effectiveness,
simplicity, and economy. It has replaced other fine screening devices in
numerous applications where minimum maintenance is required.
Extremely rapid return on investment has been experienced by many food
processing users through reduced sewer surcharges or by-product
recovery.
153
-------
TABLE 2. STATIC SCREEN APPLICATION GUIDELINES
Industry /Application
Vegetable and Fruit
Processing 0.
Meat Packing
. Pen Washing
. Paunch Manure Recovery
. Scald Tank Recycle
. Stick Water
. Expellar Grease
. Catch Basin Sludge
. Total Plant Wastewater
Dairy
Poultry
. Growing Facilities
. Processing Plant
Seafood Processing 0.
Brewery
. Grain-Hops Recovery
. Caustic Wash Water
Recycle
Winery •
Screen
Opening
(in.)
020-0. 080
0. 040
0. 040
0. 020
0. 020
0. 020
0.040
0.040
0. 040
0.040
0. 020
020-0. 030
0.030
0. 040
Susp. Solids
Removal
(%)
35 - 85
30 - 35
.3,0 - 35
30 - 35
30 - 35
30 - 35
30 -.35
30 - 35
40 - 45
50
50
30 - 50
50
75 - 80
%
Dry Wt.
Solids
•
10 - 15
40
25
25
5-10
10
20
40
15
15
. 15
10 - 15
20
5
Flow Rate
•(72" Unit)
(gpm)
400 - 1000
600 - 1100
600
400 - 500
300- 400
20 - 40
300
500 - 700
450
400 - 50Q
400 - 500
600
500
300
. Juice Removal 0. 020 25 8 400
154
-------
PRELIMINARY EVALUATION
OF ANAEROBIC SLUDGE DIGESTION FOR
THE TUNA PROCESSING INDUSTRY
by
A. Kissam*, H. Barnett**, F. Stone**, P. Hunter**
INTRODUCTION
The tuna processing industry of southern California is somewhat unique among
the segments of the seafood industry in that considerable quantities of bio-
degradable waste are routinely available to support biological treatment
processes. The plants are congregated and produce similar waste products
for disposal. Although one plant may close down for periods ranging from
days to several weeks, other nearby plants typically continue to produce
considerable amounts of waste solids.
Due to the high organic solids .and lipid content of the processing wastes es-
caping screening, the most suitable means of waste component capture is
dissolved air flotation (DAF). The flotation process is assisted by chemical
addition (i.e., lime, alum, polyelectrolytes, ferric chloride, sodium alumi-
nate) to promote particle contact and coagulation. The coagulated organics
are buoyed to the surface by the dissolved air and removed by surface skim-
ming paddles. The skimmed waste is called sludge.
Disposal of sludge has created operational and economic difficulties for the
tuna industry. Presently, the high water content tuna sludge is trucked to
land disposal at the rate of 9,000 GPD by one major processor located at
Terminal Island, California (1). Although land disposal of unstabilized
sludge is presently practiced, restrictions concerning the disposal of such
material is envisioned. The cessation of land disposal would require barging
or incineration operations, both of which may be prohibitively expensive.
Sludge digestion is a common practice in municipal wastewater treatment
systems, with anaerobic digestion being the predominant process. Anaerobic
digestion reduces the sludge volume, improves dewaterability, and provides
a useful energy supplement in the generated methane gas. The process is
commonly termed as a two-stage biological process. In the first stage, the
complex organics of the wastes are converted to volatile acids. The second
stage involves the conversion of this volatile acid substrate to methane
and carbon dioxide gas. The bacteria responsible for the stages of treatment
are two distinct groups which must be in balance for the process to proceed.
*Lieutenant, N.O.A.A. Commissioned Corps
Assigned to Washington Sea Grant, Seattle
**Utilization Research Division, Northwest and Alaska Fisheries Center, Nat-
ional Marine Fisheries Service , N.O.A.A.
T55
-------
Anaerobic digestion has been considered as a sludge treatment application for
the tuna industry, but past in-plant practices have negated serious consid-
eration (2). Saltwater is used to partially thaw the catch aboard the vessel
and is transferred into the plant waste sump during unloading. The catch is
further thawed, prior to processing, in brine tanks located inside the pro-
cessing facility. The saltwater used in these operations becomes highly
polluted and requires treatment by the processor prior to discharge. This
saltwater intrusion into an otherwise fresh water process creates periods of
high salt content in the generated sludge that could cause toxicity in anaer-
obic organisms (2, 3). The net result on an anaerobic sludge treatment
process would be either organism acclimation to the salt concentration with
a reduced treatment efficiency or digester failure in the case of slug loads
on a digester. Proposed alternatives are:
(1) Provide fresh water thawing, with recycle capability, inside the plant.
This is presently practiced in many plants for reasons exclusive of
sludge salt concentration.
(2) Return holding water to the vessel upon completion of catch offloading.
The polluted water could then be pumped overboard at sea. With the
institution of fresh water thawing inside the plant, the salt discharged
from vessel holds may not be a significant source.
If such alternatives were instituted, more reliable wastewater treatment
would result, and the tuna sludge disposal costs may be reduced by incorpora-
tion of the anaerobic digestion. Anchovie and mackeral sludge from other
lines would have to be segregated and disposed of without digestion due to
an unavoidable salt influx during processing.
In light of the potential for in-plant changes and the persistent sludge dis-
posal problems, an anaerobic digestion demonstration project was undertaken
by the Utilization Research Division, Northwest and Alaska Fisheries Center,
National Marine Fisheries Service (NMFS), Seattle, Washington.
TABLE 1. TEST SLUDGE ANALYSIS*
PH
Alkalinity
Ammonia NItro.
Protein
Lipids
Total Solids
Vol. Solids
Total COD
Soluble COD
Spec . Grav .
NaCl
Average
6.4
962
30.8
1.9
3.5
8.2
83
139.4
9.6
1.02
0.07
SD
—
27.5
2.3
.2
.7
.5
1.9
3.3
0.4
—
.01
n
—
3
6
6
8
8
8
4
5
—
5
Units
—
mg/1 as CaC03 to pH=4.6
mg/1
% of Sludge
% of Sludge
% of Sludge
% of TS
gm/1 COD
gm/1 COD
—
% of Sludee
*Digester feed diluted 1:1 with distilled water
156
-------
Test sludge was obtained from a pilot DAF unit operated by NMFS at Whitney-
Fidalgo Seafoods, Anacortes, Washington. The sludge was essentially salt-
free, as the tuna were offloaded partially frozen and air thawed at the
processing plant. Wastewater treatment consisted of screening, followed by
DAF using 200 ppm alum, 100 ppm lime, and 5 ppm anionic polymer. The test
sludge should adequately represent a sludge which would result from the
previously discussed in-plant changes at the major tuna processors. The test
sludge analysis is presented in Table 1.
Toxicity
Toxicity of any substance must be discussed in terms of concentration. The
effect of any substance on the metabolic rate of an organism is a function
of the substance concentration. As evidenced in Figure 1, the increase of a
substance concentration produces organism growth which may be described as
stimulatory, constant, and finally, toxic.
-------
(NH4+), potassium (K+), calcium (Ca4-f), and magnesium (Mg++). In addition,
the antagonist for the primary cation present in sea water (Na+) was identi-
fied as being potassium. The concentration of antagonist required to count-
eract a toxin was very small, as evidenced in Table 2.
TABLE 2. EFFECT OF POTASSIUM ANTAGONIST ON SODIUM TOXIN
No Antagonist Range of Peak Peak Antagonism
Toxin Reaction Rate Antagonist Concentration Reaction Rate
Concentration % Control M % Control
.3
.4
54
30
0.002
0.005
- 0.06
- 0.03
72
56
Because of the limitations of a control reaction comparison, the effects of
potassium and sodium combinations were studied at various solids retention
times (6). By incorporation of various solids retention times, it is possible
to study the kinetic effects of toxicity and antagonism.
The effect of sodium on the organisms was manifest as a lowering of the cell
yield constant and an increase in the organism decay rate for increases in
sodium concentration. This is significant, because the ability of the organ-
isms to process substrate was not affected while the organism population was
lower than would normally occur. With the addition of 0.03 M potassium to a
process retarded by 0.35 M sodium, the yield and decay rates returned to near
normal values. Therefore, achieved was a confirmation of the antagonistic
ability of potassium to sodium toxicity.
The significance of the kinetic approach to toxicity is that process design
variations dependent upon irregular cation concentrations can be systematic-
ally approached. After obtaining the ionic character of a waste to be
treated anaerobically, the designer can more accurately predict effluent
quality and antagonist concentration required to achieve adequate treatment.
This is accomplished by adjusting the kinetic constants for the ionic con-
centrations present in the waste.
For this study, insufficient data concerning the ionic character and kinetics
of tuna sludge digestion are available. This limits speculation as to the
ultimate value of the kinetic constants for high-salinity tuna sludge diges-
tion. The effect of high-salinity sanitary waste discharges from vessles to
municipal biological treatment systems has been appraised for possible antag-
onistic combinations of cations in sea water (7). It was theorized that
sufficient antagonist cations are present in sea water to sustain the anae-
robic process in a digester at 12,000 ppm NaCl. Other investigators (8)
surveyed municipal treatment systems in Florida which operate anaerobic di-
gesters with constant NaCl inputs of 7,000 to 10,000 ppm. The units operate
satisfactorily, but continuous levels above 13,000 ppm were considered to be
excessive for stable operation.
158
-------
Star-Kist Foods (1), located at Terminal Island, California, reports NaCl
concentrations ranging from 6,000 to 14,600 ppm in the final sludge from
their DAF unit, with average NaCl concentrations being about 13,000 ppm.
Although the possible accumulated NaCl concentration in a digester cannot
be estimated, it is reasonable to expect a toxic condition due to variations
in salt loading and a high average salt concentration.
The tuna sludge used in this study was obtained from a brine-free process
and had a measured NaCl of 700 ppm. This sludge should adequately repres-
ent a sludge which would result from the in-plant thawing operation changes
previously discussed.
Test Procedure
The study was conducted using three 2,000 ml flasks suspended from a wrist
action shaker into a controlled temperature (97°F) water bath. Generated
gas was collected over an acid/brine solution contained in calibrated 4,000
ml flasks.
Precautions were taken to assure a fresh test sludge by refrigerating the
sludge following collection. The sludge was transported to the laboratory
and divided into 400 ml packets prior to freezing at 0°F. Sludge packets
were thawed as required and refrigerated when not in use.
The digesters were seeded with 750 ml of digesting sludge obtained from a
municipal digester. Sludge feeding was commenced on the following day at
solids concentration less than 1%, with a gradual increase to the test
solids concentration of 4.1% (raw sludge concentration = 8.2% solids). The
raw sludge was diluted 1:1 with distilled water and warmed to 97°F prior to
feeding. The units were fed once per day using volumes consistent with the
desired digester detention time. Air contamination was avoided by using a
50 ml plastic syringe for all feeding operations. However, the feed sludge
may have been oxygen saturated due to vigorous mixing during syringe filling
operations.
Three digesters T*ere successfully operated at detention times of 8, 12,
and 15 days. Prior to data collection, all digesters were operated a
minimum of 3 detention periods to assure greater than 95% seed sludge
removal.
159
-------
Laboratory Analysis
All analyses were conducted in accordance with Standard Methods (9) with the
following exceptions:
Raw sludge volatile solids analysis conducted as prescribed by
A.O.A.C. (10). The method involved pre-drying of the sample
using heat lamps prior to ashing in the muffle furnace. This
was required to avoid flare-up in the muffle furnace due to
the high lipid content.
Protein analysis was conducted on a Kjeldahl apparatus using a
wet sample. Ammonia nitrogen content was subtracted from the
TKN prior to protein concentration calculation.
Total alkalinity of the sludge was determined by titration with
0.02N H2S04 to pH = 4.6 using electrode pH probe and magnetic
stirrer.
Gas samples were collected over acid/brine solution in glass
transfer bottles. Analysis for CC>2 was by Orsat; analysis
for CH^ by gas chromatograph. Methane percentage was a calcu-
lated average using peak height on a strip chart recorder.
The instrument was calibrated prior to each day's use.
NaCl by Quantab #1176 chloride titrator after glass fiber
paper filtration.
Results
The results are presented in Tables 3 and 4, listing observed removal and
operating parameter values for each digester. Of particular significance is
TABLE 3. SPECIFIC REMOVALS
Digester Detention-Days
1 I! ±5
Total Solids 41% 54% 47%
Volatile Solids 49 65 57
Total COD 47 59 64-70*
Protein 26 47 47
Lipids 77 77 83
*COD-methane balance indicates the raw sludge packet fed to the 15-day diges-
ter was of considerably higher COD concentration, 85 gm/1 versus 69.7 gni/1
average.
160
-------
TABLE 4. DIGESTER OPERATING PARAMETERS
pH
Total Alk.
Total Solids
Vol. Solids
Total COD
Soluble COD
Protein
Lipids
Vol. Acids
Atnmonia-Nitro .
Gas Prod.
Gas - % CH4
Gas - % C02
VS Loading
CH4/VS
8
6.8
2909
2.3
70.6
36.9
6.9
0.7
0.4
431
1000
1279
70
26
0.27
4.5
SD
58
0.3
6.3
4.3
0.9
0.1
0.1
47
90
90
3.1
1.4
n
5
11
11
6
3
3
4
3
3
8
5
5
12
6.8
3450
2.1
63.1
28.8
7.2
0.5
0.4
400
890
1287
76
23
0.18
7.3
SD
202
0.2
5.1
1.5
1.2
0.1
.1
40
42
67
2.3
1.2
n
6
8
9
6
3
3
4
4
3
15
5
5
15
6.9
3580
2.3
67.9
25.0
6.9
0.5
0.3
355
1095
1306
80
22
0.14
9.9
SD
141
0.2
11.7
2.1
0.2
0.1
0.2
21
19
39
1.8
0.4
n
8
10
9
6
3
3
9
5
4
13
4
4
UNITS
mg/1 as CaC03
%
% of TS
gm/1
gm/1 COD
% of sludge
% of sludge
mg/1 as acetic
mg/1
ml/day @ STP
%
%
#vs/Fx2. day
FT3CH4
#VS added
-------
the high methane content of the generated gases, which exceed the accepted
average production (67% CH^) achieved by municipal digesters. Rudolfs (11)
identified a high correlation between grease content and combustible gas
production.
As may be observed, the percentage removals for the 15-day digester are some-
what lower than one would expect. This may be explained by the fact that the
15-day digester data were collected after completion of the 8- and 12-day
units. Although every effort was made to assure uniform distribution of the
test sludge during initial 400 ml packet proportioning, the sludge packet fed
to the 15-day digester evidently was of a different content than previous
packets. The analysis of the feed sludge was conducted using samples from
different freezer packets, but the feed sludge was not continually tested
throughout the experiment. Although the removal percentage relationships
between the digesters are not consistent, the purposes of this experiment
have been achieved through successful digester operation.
Daily methane production, effluent COD, and the stoichiometric ratio of 350
ml methane per gram COD stabilized may be utilized to estimate the influent
COD concentration for a digester. The 8- and 12-day digesters balance within
8% of the analytically determined influent concentration of 69.7 gm/1 COD.
The 15-day digester, however, requires an assumed influent COD of 85 gm/1
in order to balance. This value has been applied to Table 3 as indicated.
The non-degradable COD of the feed sludge was approximately 24 gm/1. This
assumes that the degradable COD remaining at higher retention times was com-
parable to the COD levels experienced by O'Rourke (12). The work by O'Rourke
demonstrated that the effluent degradable COD from a digester is predictable
and independent of the influent strength. Therefore, for a complex waste,
the COD percentage removal by a digester depends on the temperature and
solids retention time.
The settling properties of the sludge were greatly improved as a result of
digestion. Although fines are evident in the supernatant during a short
settling period, laboratory centrifugation of the digester effluent produced
a visually clear effluent. This improved dewaterability is considered to be
a result of releasing bound water from the organic solids and the high remo-
val of lipids from the sludge. However, it should be pointed out that the
sludge was frozen prior to digestion. Resuspension of the raw sludge after
thawing produced a sludge with poor settling properties. Centrifugation of
the raw sludge produced a multi-layered sample with considerable solid reten-
tion in the lipid layer.
•
APPLICATION EXAMPLE
The costs for certain facets of discharge control may be minimized through
application of recognized techniques. It is the purpose of this paper to
discuss anaerobic sludge digestion in terms of sludge volume reduction, com-
bustible gas generation rates, and sludge disposal frequency.
The industry is encouraged to conduct an investigation on a scale far greater
162
-------
than undertaken by this study. The construction and operation of a pilot
scale system is advisable for two reasons:
(1) Obtain specific design data;
(2) Train and familiarize personnel with the process. This may prove to be
as important as the collection of the pilot plant operating data.
The application example presented in this section is sized for one plant
using conservative design values. In actuality, the most economical method
of sludge disposal would be a communally operated sludge processing facility.
The processors are located virtually next door to one another and have common
sludge disposal costs, which identifies cooperation as a major factor in
reducing overall sludge disposal costs.
Based on the results of this study, and Metcalf and Eddy (13), it is reason-
able to assume for a single complete mix heated digester:
8 FT3 CH4 per Ib. VS added
55% total solids reduction
9000 GPD raw sludge (1)
Other sludge characteristics the same as the test sludge.
Tank capacity to meet the requirements of an actual installation should be
determined from pilot plant data and, hopefully, the application of a mathe-
matical model. The digestion of industrial waste is a very specific process
and digester loading parameters may vary from successful municipal loadings.
In the interest of conservatism, a high sludge detention time will be used to
size the digester.
Metcalf and Eddy, Inc. (13) report that loading rates of 0.10 to 0.40 Ib of
volatile solids per cubic foot per day and hydraulic detention periods of 10
to 20 days are practicable for high-rate digesters. The main purpose of
digestion of the tuna sludge is improved dewaterability and volume reduction.
In order to achieve these goals with a minimum of operational difficulty, a
30-day detention period with a loading of 0.14 Ib of volatile solids per
cubic foot per day will be proposed. This loading is conservative and well
within the above suggested loading range. A schematic of the example system
is presented in Figure 1. The Port of Los Angeles reports that non-water-
front sites, near the processors, could be leased. The land lease cost is
based on a 100 x 100 parcel and is listed in Table 5.
163
-------
jj
(&}
Gas
9000 GPD
3.6% Solids
Complete Mix
digester
36,000 FT3
20 GPM firm
Centrifuges
t
Solids
to disposal
1,300 GPD
25% Solids
POIT cT,,Am> f »uuu \—7 ^ to n A F
9000 GPD -*| I 3.6% Solids >\/ > £ J-^«
8% Solids
Figure 1. Schematic diagram of example installation
The digested sludge observed during the course of this study exhibited good
settleability and, therefore, should centrifuge to approximately 25% solids
after digestion (14).
Projected Performance
A total dry solids loading of 6,100 pounds per day is calculated (sp. grav.=
1.02). After digestion with a 55% total solids reduction, the total dry
solids remaining is 2,750 pounds per day.
If we assume that this effluent (containing 2,750 pounds dry solids) is cen-
trifuged to 25% solids concentration and a specific gravity of 1.04 is
attained, the reduction in disposal volume and weight may be calculated.
These calculations result in approximately 86% reduction for both disposal
weight and volume.
The combustible gas production achievable from such a system is insignificant
in comparison to the natural gas volumes used by a processor for cooking,
canning, and can fabrication operations. However, the generated gas would
be too valuable to flare off and must be considered as a marketable item.
Assuming the gas is 78% methane and has a net heating value of 700 BTU per
cubic foot of digester gas, the equivalent value in terms of natural gas
($0.14/therm) may be calculated. After consideration of the digester heat-
ing requirements (15), the residual gas value of $15,500 per year is
obtained.
Present Value Comparison
In the following section, the example system will be cost estimated and
subjected to a present value analysis versus the present disposal costs. An
important feature of this comparison is that the present disposal method is
severely affected by fuel inflation, since the costs incurred are primarily
for trucking. Inflation for trucking costs can be expected to exceed the
national overall inflation rate.
164
-------
The comparison was conducted for various before-tax cost of capital, national
inflation, and fuel inflation rates. An 7% investment tax credit was applied
against the construction costs and the capital equipment was depreciated
using the straight line method over 20 years. The tax savings due to depre-
ciation (50% allowed) was applied against the yearly operation and mainten-
ance cost. The equivalent value of the excess sludge gas in terms of natural
gas was computed by converting from BTU to therms and multiplying by the
local billing factor. This equivalent value is considered to be a positive
cash flow value, since plant energy requirements can be supplemented. Finan-
cing costs are not included in the cash flow because they are inherent in
the present value calculation.
The cash flow result of this analysis is presented in Table 5, which shows a
total annual reduction in expenditures of approximately $78,000. Information
on present annual tuna sludge disposal costs was provided by Star-Kist
Foods (1).
TABLE 5. DISPOSAL OPTION COST SUMMARY
SEPTEMBER, 1976 DOLLARS
PRESENT EXAMPLE
Total Construction $500,000
$ 5,000 Yearly O&M $41,000
100,000 Yearly Disposal 12,000
Yearly Land Lease (Port of
Los Angeles) 2,300
Yearly Gas Savings (-) 15,500
Yearly Depreciation Tax
Savings (-) 12,375
$105,000 COST PER ANNUM $27,425
TOTAL CONSTRUCTION COST $500,000
It should be pointed out that the construction, maintenance, and operation
costs were developed from graphs published by the Environmental Protection
Agency (EPA) (16). The EPA graphs were produced for use in cost estimating
municipal plants, not small industrial applications. A complete engineering
cost study for a proposed installation may provide more accurate and current
cost figures on which to base management decisions. The costs obtained from
the EPA were updated to 1976 costs by data provided in the Engineering News-
Record (17).
Equations and values used in the present value comparison are:
165
-------
Present Method
PV1 =
t i
l+c/c
+ B
where
= present value of disposal cost and 0/M cost over 20 years
= yearly disposal cost; $100,000
= yearly 0/M cost; $5,000
= assumed fuel inflation rate; 5 to 25% by 5% increments
= assumed national inflation rate; 4 to 20% by 2% increments
c/c = assumed before-tax cost of capital 2 to 50% by 2%
increments
t = terms; 20 years
A
B
if
Example System
PV2 = M +
20
t=l
(N-T)
(P-Q)
t -i
where
P?2 = present value of construction, disposal, tax savings,
excess gas, and 0/M costs.
M = construction cost; $500,000
N = yearly 0/M and land lease costs; $43,300
P = yearly disposal cost; $12,000
Q = yearly excess gas value; $15,500
T = yearly depreciation tax savings; $12,375
if, IL, c/c, t as before
These equations and values were run on a computer with the printout showing
PV2~PVi as a function of progressing rates of national inflation, fuel infla-
tion, and before-tax cost of capital. The cross-over point, where PV2"-PVi
equals zero, occurs approximately at capital rates of 20, 27, 34, and 40%
for fuel inflation rates of 5 to 20% by 5% increments, respectively. These
cross-over points hold for all values of national inflation analyzed, which
implies that fuel cost and initial construction cost are the prime values.
Similarly, the point at which the 20-year savings equals the initial con-
struction cost ($500,000) can be determined. This point occurs at capital
rates of 9, 15, 21, and 27% corresponding to the 5 to 20% rates of fuel
inflation analyzed.
A firm would have to achieve a before-tax return on an investment in excess
166
-------
of the above capital rates in order to justify alternative uses for the
money. If the analysis parameters are excluded and consideration is given
only to the total construction cost and annual cost savings presented in
Table 5, the minimum internal rate of return on the investment is 14.5% for
a term of 20 years.
CONCLUSIONS
Admittedly, the installation and operation of a digestion system is a major
undertaking. However, unlike other treatment systems, anaerobic digestion
offers long-term operational savings. This savings would be realized pri-
marily because of energy unavailability and energy-dependent industries
(trucking) cost increases. This will be particularly true, according to
Business Week (18), in southern California during the next five years. In
California, natural gas is the source of 55% of energy. Curtailments of
natural gas supply to industry results in shifts to alternative energy
sources (petroleum-based) and overtaxing of the alternative supply.
The resulting shift from natural gas to alternative fuels will certainly
strain the production/distribution capacity and the consumer's pocketbook.
Therefore, sludge disposal costs may rise disproportionately to normal
inflationary costs. This would tend to highlight the savings in disposal
volume due to solids reduction and improved dewaterability possible through
anaerobic digestion.
Such trends would only serve to accelerate the cost justification for instal-
lation of anaerobic digestion facilities in the tuna processing industry.
Due to the close proximity of the plants, a commumal effort would serve to
reduce the operating costs per plant.
REFERENCES
1. Evich, V.J., Engineering Manager, Star-Kist Tuna, T.I., CA. Personal
communication, 1976.
2. Development Document; Catfish, Crab, Shrimp, and Tuna; Point Source
Category EPA-440/l-74-020-a (1974).
3. Chun, M.J., £lt al., "A Characterization of Tuna Packing Waste." Water
and Sewage Works, 117:10, 3-10, 1970.
4. Kugelman, Chin, "Toxicity, Synergism, and Antagonism in Anaerobic Waste
Treatment Processes," Advances in Chemistry Series 105, 1971.
5. Kugelman, I.J., McCarty, P.L., "Cation Toxicity and Stimulation in
Anaerobic Waste Treatment. I. Slug Feed Studies." J. Water Pollut.
Contr. Fed. (1965) 3_7, 97 and "...II. Daily Feed Studies," Proc. Ind.
Waste Conf., 19th, Purdue University, 1965.
167
-------
6. Chin, Kugelman, Molof, "The Effect of Monovalent Cations on the Methane
Phase in Continuous Digestion Systems," as cited by reference (4).
7. Lellelid, "Appraisal of Watercraft Waste Impacts on Municipal Waste-
water Systems, " Masters thesis, University of Washington, 1977.
8. Reynolds, Smith, Hills, "Engineering Report on Shore Disposal of Ship
Generated Sewage at Activities in the Eastern Area," Naval Facilities
Engineering Command, NTIS: AD 747 998, June, 1969.
9. Standard Methods for Examination of Water and Wastewater, American
Public Health Association, Wash. D.C. (1975).
10. Assn. of Official Analytical Chemists, Washington, D.C. (1973).
11. Rudolfs, W., "Decomposition of Grease During Digestion, Its Effects on
Gas Production and Fuel Value of Sludges," Sewage Works J. (1944).
12. O'Rourke, J.T., "Kinetics of Anaerobic Treatment at Reduced Tempera-
tures," Doctorial thesis, Stanford University, 1968.
13. Metcalf and Eddy, "Waste Water Engineering Collection, Treatment,
Disposal," McGraw-Hill Company, 1972.
14. E.P.A., "Processing Design Manual for Sludge Treatment and Disposal,"
E.P.A. 625/1-74-006 (1974).
15. Water Pollut. Contr. Fed. Manual of Practices No. 8 (1972).
16. E.P.A., "Estimating Costs and Manpower Requirements for Conventional
Wastewater Treatment Facilities," E.P.A. 17090 DAN (1971).
17. Engineering News-Record, "Costs Scoreboard," Vol. 197, No. 9; Vol. 187,
No. 1.
18. Business Week, "The Natural Gas Shortage Gets Worse. . .," No. 2451,
9-27-76.
ACKNOWLEDGEMENTS
The authors wish to acknowledge the assistance of Dr. James Bray, economist
for Washington Sea Grant, and Dr. John Ferguson, Professor of Civil Engineer-
ing, University of Washington. Any errors and/or omissions in this article
should not be attributed to anyone other than the authors.
168
-------
REUSE OF BRINES IN COMMERCIAL CUCUMBER FERMENTATIONS
by
R.F. McFeeters*, M.P. Palnitkar**, M. Velting**, N. Fehringer*** and W. Coon*
INTRODUCTION
Spent brine remaining after cucumber fermentations is a difficult waste dis-
posal problem. Usual fermentation practice will result in production of about
40% by volume spent brine in a fermentation tank. As a rule this brine will
contain 9-15% NaCl, the BOD will be 10,000-15,000 ppm and the pH will be 3.2-
3.5. These characteristics make biological treatment of the waste relatively
difficult. In locations with limits on total dissolved solids it may be
impossible to discharge the salt produced by normal tankyafd operations.
Recent efforts have been directed toward development of recycling procedures
for brines as a means of minimizing the waste generated by a tankyard. The
current project was designed to evaluate commercial application of pasteuriz-
ation and chemical treatments for spent brine.
Potential for Waste Reduction by Recycling Brine
Figure 1 shows theoretical calculations of the pounds of NaCl which will be
discharged per bushel (50 Ib) of cucumbers in a normal fermentation. Curve
"a" shows the salt losses without brine recycling as a function of the salt
concentration in the fermentation tank. Curve "b" is the salt loss from
cucumbers which are desalted in fresh water to bring the salt concentration
down to acceptable levels. This loss will occur whether or not recycling is
practiced. The difference between curves "a" and "b" at any given salt con-
centration is the potential saving of salt which can be realized as a result
of recycling.
It must be recognized that salt losses beyond the theoretical loss will occur
in most tankyards. First, the tank facilities are usually of wood construc-
tion. A significant fraction of these tanks will leak. The amount of leak-
age can vary from slight to very substantial depending upon the state of
repair. Secondly, the tank is normally open in order to utilize sunlight to
inhibit yeast and mold growth on the surface of tanks. During periods of rain
the tank will overflow with the loss of some brine. The salt loss from tanks
was calculated by subtracting the amount of salt in a tank based upon salt
concentration and the volume of brine and cucumbers in the tank from the total
amount of salt added as recycled brine, dry salt and make up brine. This
evaluation of salt balance showed a loss of about 1 Ib of salt per bushel of
cucumbers brined.
*Department of Food Science and Human Nutrition, Michigan State University,
East Lansing, Michigan 48824.
**Vlasic Foods, Inc., W. Bloomfield, Michigan 48033.
***Detroit District, Food and Drug Administration, Detroit, Michigan.
169
-------
CO
0£
LU
CQ
o
LU
X
CO
z>
pa
Q
LLJ
O
ce
o
15,0
CO
CO
Q
o
Q-
10,0
v«*
S 5,0
20 30 40 50 60
SALT CONCENTRATION (°SALOMETER)
70
Figure 1. Theoretical salt discharge from cucumber fermentations.
Calculations based upon a 65:35 fresh cucumber:brine ratio,
Curve a—salt discharge without recycling. Curve b—salt
discharge with brine recycling.
170
-------
Geisman and Henne (1,2) developed a chemical procedure for brine treatment.
Brine pH was raised above pH 10 and held for 24-48 hr. A sludge layer was
removed and the pH of the brine was adjusted to near 7 with HC1. After
treatment, the brine could be reused as a cover brine in subsequent fermenta-
tions with no apparent adverse effects on the fermented cucumbers. Further
investigations of this recycling technique (3) confirmed these results, but
also indicated that pasteurization of the brine followed by a pH adjusted to
4.5-5.0 would serve as an alternative treatment method.
The objectives of this project were to compare the chemical and pasteurization
procedures in commercial brining operations and to evaluate the adequacy of
the treatment procedures.
Brine Treatment Procedures
The primary purpose of brine treatment is to assure the processor that
recycled brine, which is normally held for a period of months in open brining
tanks, will be free of pectinase enzyme activity which could cause product
softening. A preliminary recommendation was to pasteurize brine at 175°F for
30 sec. Based upon data on inactivation of a heat stable commercial pectinase
from Aspergillus niger this treatment appeared to be marginally adequate (3).
The inactivation of enzyme at high pH had received limited investigation. It
was decided to raise the pH of brine to 11.0 with NaOH and hold it at this pH
for 24-48 hr.
Studies of the thermal inactivation of pectinases from eight fungal species
reported to be commonly present on cucumbers and flowers (4) showed that
pectinase from Penicillium janthinellum was the most heat stable among those
investigated (5). Detailed studies of the effect of pH and salt concentration
on the stability of this enzyme led to the conclusion that pasteurization of
brine at 175°F for 30 sec is sufficient to assure pectinase free brine over
the range of salt and pH conditions which are likely to be encountered in
commercial practice (5). It was recommended that the pH of pasteurized brine
be adjusted to near 4.5 with NaOH just prior to use.
Denaturation studies of P. janthinellum pectinase at high pH showed complex
denaturation kinetics. This result contrasted with the results of the thermal
inactivation studies between pH 3.0 and 4.7 in which first order kinetics were
observed. To estimate the time required for chemical treatment in spent brine
denaturation experiments were done at a series of pH values and the time
required for 90% or 99% inactivation of the enzymatic activity was determined
(Table 1). To assure destruction of at least 99% of the initial enzyme
activity it is necessary that the brine be held at 72°F or higher, that the
pH is 11.0 or higher and that the time of treatment is at least 36 hr. After
treatment at high pH, the brine is adjusted to a pH near 4.5 by addition of
vinegar (acetic acid).
The pectinase activity in recycled brines before and after treatment and the
activity in brine samples taken from fermentation tanks at the time cucumbers
were removed were measured according to the procedure of Bell et al. (6). No
significant pectinase activity was found in these samples. As a result it is
not possible to evaluate the effect of brine treatment procedures on pectinase
171
-------
activity "naturally" present in the brines. This was not unexpected since
enzymatic softening is relatively rare in Michigan and the fermented cucumbers
obtained in these experiments were of normal firmness.
TABLE 1. INACTIVATION OF PECTINASE FROM PENICILLIUM JANTHINELLUM
AT HIGH PH IN SPENT BRINE. TEMPERATURE 22°C (71.6°F).
SALT CONCENTRATION 8.0%
Time for Time for
. P ° . 90% inactivation 99% inactivation
incubation /, \ /•, ^
(hr) (hr)
10.6 50% inactivation in 70 hr
11.0 21 . 34
11.2 5.0 37
11.3 6.5 27
11.6 2.5 13 (est)
On the scale that chemical treatment was practiced no large equipment was
required. Brine samples were titrated and the amount of vinegar or NaOH
required for pH adjustment of a tank were calculated. Vinegar (30% acetic
acid) was added as a liquid. Sodium hydroxide was added as pellets by a
person dressed in a protective rubber suit with mask and gloves.
Pasteurization of brine was done with a portable APV titanium alloy heat
exchanger which was developed for this application. This was a regenerative
heat exchanger heated by propane gas. The exchanger was set to heat brine at
190°F with a holding time of 30 sec. Under these conditions brine was treated
at a rate of 50 gal/min.
Design of the Commercial Evaluation of Brine Recycling
Figure 2 shows the path of brine through three fermentation cycles. The
project began in 1975 with the segregation of 1st cycle spent brine from the
general tankyard brines. After pasteurization or chemical treatment it was
used for 2nd cycle fermentations in 1975- The 2nd cycle brine was saved and
treated in the spring of 1976 and used for 3rd cycle fermentations.
Control fermentations in which cucumbers were covered with fresh salt brine
were run in both 1975 and 1976.
Samples of brines and cucumbers were taken at appropriate points and analyzed
to determine the effects of using recycled brine.
172
-------
Fresh Salt Brine
I
1974 1st Cycle Fermentation
I
1st Cycle Spent Brine
Pasteurization
Treatment
I
Chemical
Treatment
1975 2nd Cycle
Fermentation
1
i
1975 2nd Cycle
Fermentation
1975 Control
Fermentation
t
2nd Cycle
Spent Brine
1
2nd Cycle
Spent Brine
I
I
1st Cycle
Spent Brine
Pasteurization
Treatment
Chemical
Treatment
1976 3rd Cycle
Fermentation
1976 3rd Cycle
Fermentation
1976 Control
Fermentation
i
3rd Cycle
Spent Brine
3rd Cycle
Spent Brine
1st Cycle
Spent Brine
Figure 2. General design of the commercial evaluation of brine recycling,
173
-------
Changes in Recycled Brines
Table 2 shows the effect of pasteurization treatment on first cycle (1975) and
second cycle (1976) spent brines. The rise in pH and the decrease in titrat-
able acidity are the expected results of NaOH addition. Salt and BOD would
not be expected to change significantly as a result of pasteurization treat-
ment. There is an increase in BOD from the first to the second cycle.
TABLE 2. ANALYSIS OF SPENT BRINES BEFORE AND AFTER PASTEURIZATION
TREATMENT AND PH ADJUSTMENT
1st cycle brine
Acidity (%)
pH
Salt (%)
BOD (ppm)
initial
0.41
3.56
12.5
10,000
treated
0.17
4.57
12.6
9,900
2nd cycle brine
initial
0.62
3.52
11.4
12,500
treated
0.21
4.54
11.8
14,000
Table 3 shows the corresponding data on the effect of chemical treatment.
Similar changes in pH and titratable acidity were observed. Since vinegar is
added to drop the pH, there is an increase in BOD of about 3,000 ppm after
treatment. Like the pasteurized brines there is some increase in BOD between
the first and second cycle.
TABLE 3. ANALYSIS OF SPENT BRINES BEFORE AND AFTER CHEMICAL TREATMENT
Acidity (%)
pH
Salt (%)
BOD (ppm)
1st cycle
initial
0.47
3.50
12.8
10,100
brine
treated
0.23
4.38
12.8
13,000
2nd cycle
initial
0.63
3.66
11.1
14,500
brine
treated
0.33
4.83
10.8
17,400
Chemical Use with Brine Treatment
Table 4 shows the chemical use with pasteurization or chemical treatment of
the brine. The pasteurization process requires propane for the heat exchanger
174
-------
and NaOH for pH adjustment from the original brine pH to pH 4.5. Nearly a
doubling of the NaOH required for pH adjustment of pasteurized brine occurred
between 1975 and 1976. Data obtained independently on experimental tanks, a
complete tankyard operation and analysis of spent brine titration curves (3)
all indicated a requirement of about 14 Ib NaOH/1000 gal of brine. Therefore,
the 1975 result is believed to be incorrect. Chemical treatment results in
use of greater amounts of NaOH to raise the pH above 11 plus acetic acid to
drop the pH back to 4.5. There was an increase in the use of both NaOH and
vinegar in 1976 compared to 1975.
TABLE 4. CHEMICAL USE FOR BRINE TREATMENTS PER 1000 GALLONS
OF TREATED BRINE
Material
NaOH (Ib)
Acetic acid (Ib)
Propane (gal)
Pasteurization
1975
7.5
0
4.8
Treatment
1976
14
0
Chemical
1975
44
30
0
Treatment
1976
52
36
0
In the treatment operations about a 5% loss occurs in the pasteurization pro-
cess as a result of discarding a few inches of sludge in the bottom of brine
tanks which contain too much dirt, seeds and broken cucumbers to risk plugging
the screen on the pasteurizer. For chemical treatment a 17% loss of brine was
found. This occurs because a precipitate is formed which settles as a sludge
layer. Care was taken to pump off the clear layer without disturbing the
sludge layer. However, it is likely that this loss can be reduced to perhaps
10% in commercial operation without great difficulty.
An economic analysis of brine treatment is very much dependent upon the costs
associated with a particular locality. Tables 5-7 show an analysis of the
particular situation encountered in this project. Table 5 shows the savings
realized from not discarding spent brine. The variation in the cost of BOD
treatment in different localities would be the most significant variable in
the overall savings. There are almost exactly 100 Ib BOD/1000 gal of spent
brine. Therefore, a change of one cent in the cost of BOD treatment trans-
lates in a $1.00/1000 gal change in treatment costs.
Table 6 shows the evaluation of costs for chemical treatment. The labor cost
was difficult to evaluate based upon the small scale of the operation. Since
the time required is intermittent rather than continuous as is the case with
the pasteurization operation, the labor cost for chemical treatment was esti-
mated to be 50% of the labor cost for pasteurization. The major costs are
NaOH and vinegar. There was a net cost for chemical treatment in both 1975
and 1976.
175
-------
TABLE 5. SAVINGS FROM RECYCLING 1000 GALLONS OF SPENT BRINE
Dollars Saved
NaCl, 1.08 Ib/gal @ $20/ton 10.76
BOD, 12,000 ppm at $0.08/lb 8.00
Water, $0.47/1000 gal .45
Total Savings 19.21
TABLE 6. ECONOMIC EVALUATION OF CHEMICAL TREATMENT OF SPENT BRINES
1975 Cost 1976 Cost
(dollars/1000 gal) (dollars/1000 gal)
Labor 0.84 1.32
NaOH, 1975 19.5
-------
the estimate were correct, the cost of the pasteurizer would be $2.23/1000
gal. Based upon the actual use and a 5 yr depreciation period a substantial
net savings was obtained in 1975. In 1976, with the pasteurizer cost/1000
gal substantially increased, the net savings was much smaller.
TABLE 7. ECONOMIC EVALUATION OF PASTEURIZATION TREATMENT
OF SPENT BRINES
1975 Cost 1976 Cost
(dollars/1000 gal) (dollars/1000 gal)
Labor @ $4.00/hr
Propane
Pumping
NaOH, 1975 19.5c/lb
1.67
1.23
0.71
2.76
2.64
1.29
0.71
2.41
1976, 17C/lb
Pasteurizer 5 yr depreciation 8915/yr 6.37 10.51
Total Cost 12.74 17.56
Total Savings 19.21 19.21
Net Savings 6.47 1.65
Characteristics of the Fermentations
Tables 8 and 9 show summaries of the number of tanks operated in the project,
the amount of cucumbers and the amount of brine used. The control tanks were
covered with 6.6% NaCl brines at the beginning of fermentation. This follows
recommendations of Etchells (7) for cucumber fermentation. Recycled brines
were used at the salt concentration present after treatment, approximately
11-12%. The use of a high salt cover brine is a necessary aspect of recycling
since a 2-fold dilution to 6.6% would make it necessary to discard a large
fraction of the spent brine because excess brine volume would be generated.
Cucumbers were held at 6.6% NaCl during fermentation. The additional salt
required to maintain this salt concentration as the cucumbers equilibrated
with the brine was added as dry salt on the top of the tank headboards.
177
-------
TABLE 8. SUMMARY DATA ON FERMENTATION TANKS, 1975
Control
Heat
Treatment
Chemical
Treatment
No. of tanks
Bushels of cucumbers
Gallons of brine
Gallons of brine/bushel
Packout ratio
10
7660
34321
4.48
57:43
11
8668
31418
3.62
62:38
10
7388
30881
4.18
49:41
TABLE 9. SUMMARY DATA ON FERMENTATION TANKS, 1976
Control
Heat
Treatment
Chemical
Treatment
No. of tanks
Bushels of cucumbers
Gallons of brine
Gallons of brine/bushel
Packout ratio
8
5727
19832
3.46
63:37
8
5604
20780
3.71
62:38
8
5727
19960
3.49
63:37
TABLE 10. FINAL PH, MAXIMUM TITRATABLE ACIDITY AND TIME REQUIRED
TO COMPLETE FERMENTATIONS WITH RECYCLED BRINE, 1975
Control
Pasteurization
Treatment
Chemical
Treatment
Final pH 3.23
Maximum acidity (%) 0.80
Completion of fermentation (days) 17.1
Reducing sugar (%) .035
3.33
0.92*
16.0
.042
3.44*
0.97*
15.1
.032
*Significantly different from control at p 10.01 level.
178
-------
TABLE 11. FINAL PH, MAXIMUM TITRATABLE ACIDITY AND TIME REQUIRED
TO COMPLETE FERMENTATIONS WITH RECYCLED BRINE, 1976
„ . , Pasteurization Chemical
Control „
Treatment Treatment
Final pH 3.55 3.65** 3.72*
Maximum acidity (%) 0.79 0.87* 0.98*
Completion of fermentation (days) 19.8 15.0* 16.6*
*Significantly different from control at plo.01 level.
**Significantly different from control at p 10.05 level.
Tables 10 and 11 show summaries of the characteristics of fermentations in
recycled brines. There are several statistically significant differences
between control fermentations and recycled brines. First, the final pH of the
fermentations are higher with recycled brines. The chemically treated brine
was approximately 0.2 pH units higher than the control in both 1975 and 1976.
The pasteurized brine is 0.1 pH unit higher. This difference was statistic-
ally significant in 1976, but not in 1975. The average final pH values were
about 0.3 units higher in 1975 than 1976. The reason for this difference is
not known. The titratable acidity is higher in the recycled brines. This is
not unexpected since there is significant titratable acidity in the treated
brines (Tables 2 and 3). The average time of fermentation was significantly
shorter in 1976, but not in 1975. This is a result of a longer time for the
control fermentations in 1976.
Despite some differences which are statistically significant, the changes
observed would not be significant from the standpoint of actual tankyard oper-
ations. These data and the observations of tankyard operations using recycled
brines show that recycling can be done with no changes in the tankyard opera-
tions during the fermentation period.
The data in tables 12 and 13 show the salt stock quality in 1975 and 1976.
Good stock is that with no visible defects. Commercially acceptable stock
includes the good stock plus cucumbers with slight defects which do not sig-
nificantly decrease the economic value of the product. Commercially unaccept-
able stock is that with major bloater or honeycomb defects such that the
cucumbers are suitable only for relish products.
In 1975, the quality of the salt stock was equivalent to that of the control.
In 1976, the percentage of commercially unacceptable stock was somewhat higher
in recycled brines. However, the differences were not statistically signifi-
cant. There was a large decline in good stock in 1976. This was attributed
to a higher percentage of defects in the fresh fruit.
179
-------
TABLE 12. SALT STOCK QUALITY OF CUCUMBERS FERMENTED
IN RECYCLED BRINES, 1975
Good stock (%)
Commercially acceptable (%)
Commercially unacceptable (%)
Control
72.8
85.3
14.7
Pasteurization
Treatment
72.8
86.5
13.5
Chemical
Treatment
73.0
86.5
13.5
I
No significant differences among treatments.
TABLE 13. SALT STOCK QUALITY OF CUCUMBERS FERMENTED
IN RECYCLED BRINES, 1976
Good stock (%)
Commercially acceptable (%)
Commercially unacceptable (%)
Control
52.1
84.9
15.1
Pasteurization
Treatment
46.0
80.2
19.9
Chemical
Treatment
54.0
82.6
17.4
No significant differences among treatments.
These tanks were not gas purged. Based upon the data of Costilow et al. (8),
it would be expected that the level commercially unacceptable fruit could be
significantly reduced by power application of a side-arm purging system.
The texture of the salt stock was not significantly different from the con-
trols. Commercial production and sale of cucumbers fermented in recycled
brines indicates no detectable differences in the flavor of products produced
from these cucumbers.
Table 14 shows the results of analysis for 12 elements in cucumber salt stock,
desalted cucumbers and spent brine from the 1975 commercial recycling experi-
ment. At the 0.05 level of significance there were few differences between
recycled and control spent brines, salt stock or desalted cucumbers. Among
the heavy metals only Pb showed any significant differences. The control salt
stock averaged about 0.2 ppm higher than the salt stock from pasteurized brine.
This was significant at the 0.05 level. After desalting, the control cucum-
bers were lower than the cucumbers from recycled brines. However, the differ-
ences were not significant at the 0.05 level.
180
-------
00
TABLE 14. EFFECT OF BRINE RECYCLING ON THE DISTRIBUTION OF MINERAL LEVELS IN
COMMERCIAL SALT STOCK, DESALTED CUCUMBERS AND SPENT BRINES. CON-
CENTRATIONS ARE EXPRESSED IN PPM ON A FRESH WEIGHT BASIS. 1975
Samples
Salt Stock Cucumbers
Control
Pasteurization
Treatment
Chemical
Treatment
Desalted Cucumbers
Control
Pasteurization
Treatment
Chemical
Treatment
Spent Brines
Control
Pasteurization
Treatment
Chemical
Treatment
Pb
.93
.69
.83
.48
.82
.76
.18
.18
.18
Cd
.057
.032
.061
.039
.081
.064
.008
.006
.005
Hg
.006
.009
.019
.015
.019
.009
0
0
.001
P
130
145
136
56
58
53
128
134
127
Ca
924
920
992
372
327
331
1130
1180
1120
Mg
134
136
113
59
56
53
137
142
120
Mn
1.7
1.8
2.4
.06
.29
.06
3.0
3.1
2.3
Fe
9.5
13
11
6.1
7.8
6.0
8.9
10
11
Cu
1.1
1.3
1.6
.65
.72
.63
2.9
1.8
1.3
B
1.4
1.5
1.6
.59
.59
.58
1.3
1.4
1.3
Zn
3.3
3.4
4.1
.26
1.2
.18
4.3
4.7
3.9
Al
7.2
11.5
8.4
7.0
7.1
4.9
7.3
12.6
9.9
-------
Tables 15 and 16 show 1975 data for the distribution of carbaryl and endo-
sulfan in cucumbers and brines when cucumbers were sprayed in the field with
twice the prescribed levels of these materials. These cucumbers were fer-
mented under laboratory conditions.
Three cycles of fermentation were completed with control cucumbers covered
with fresh salt brine. The brines for recycling were given the appropriate
treatment after each fermentation. Analysis of cucumbers and brines prior to
treatment and brines after treatment were done. The results do not show
buildup of endosulfan in the second cycle. There is an increase of 1.1 ppm
of carbaryl in the cucumbers fermented in pasteurized brine. This is probably
a result of a 0.6 ppm residue in the first cycle brine after treatment.
Very low initial levels of carbaryl and endosulfan were present in the fresh
cucumbers for the third cycle as a result of rain shortly after spraying.
The presence of carbaryl in brine after the second cycle pasteurization treat-
ment resulted in 0.5 ppm of carbaryl in the third cycle cucumbers.
The endosulfan is an example of a substance which remains primarily in the
cucumbers. As a result, if it enters fermentation vats on cucumbers, it
should be removed with the salt stock and not cause significant buildup in the
recycled brine. The pattern of parathion distribution is similar to endo-
sulfan. A significant proportion of the carbaryl diffuses into the brine.
Therefore, if cucumbers are put into tanks with excessive levels, it may be
possible to have significant contamination of the next batch of cucumbers.
The high pH chemical treatment of brine caused breakdown of carbaryl, endo-
sulfan and parathion.
DISCUSSION
The use of fermentation brines for 3 cycles of fermentation shows no differ-
ences between fermentations in recycled brines and those carried out in con-
trol brines. The differences observed in pH, titratable acidity and time of
fermentation would not alter the way in which fermentations are carried out.
The salt stock produced by recycling is equal to that obtained in control
fermentations. There have been no differences between the cucumbers obtained
from pasteurized vs chemically treated brine which indicate any preference for
either treatment method.
The pasteurization treatment will allow about 95% recovery of brine. Chemical
treatment will result in a somewhat higher loss of brine, but a recovery of
90% is feasible. The chemical treatment did degrade the pesticide residues of
those materials tested in this project, while the pasteurization treatment did
not. Other than these two factors it appears that selection of either of the
brine recycling procedures test can be made on the basis of economic factors.
Economic analysis of the two treatment procedures show a small net savings for
the pasteurization treatment. This saving could be quite substantial if the
pasteurizer is used to capacity and if the useful life of the pasteurizer
proves to be equal to advance predictions. The chemical treatment shows a net
cost. This major factor in reducing this cost would be to use less expensive
chemicals or to decrease the amount of chemicals utilized. Chemical treatment
does not require any large capital investments.
182
-------
00
C*>
TABLE 15. CARBARYL LEVELS IN TREATED CUCUMBERS AND FERMENTATION BRINES.
RESULTS ARE EXPRESSED AS PPM ON A FRESH WEIGHT BASIS. 1975
Cycle
1
2
3
1
2
3
1
2
Sample
Cucumbers
Cucumbers
Cucumbers
Untreated Brine
Untreated Brine
Untreated Brine
Treated Brine
Treated Brine
Cucumbers C°ntro1 Af.ter
Fermentation
3.0 2.7
5.5 2.5
0.1 Tr
1.0
1,2
Tr
Pasteurization
Treatment
2.3
3.6
0.5
0.6
1.4
.14
0.6
1.5
Chemical
Treatment
2.4
2.5
Tr
0,8
1.3
Tr
0
0
-------
00
TABLE 16. ENDOSULFAN LEVELS IN TREATED CUCUMBERS AND FERMENTATION BRINES.
RESULTS ARE EXPRESSED AS PPM ON A FRESH WEIGHT BASIS. 1975
Cycle
1
2
3
1
2
3
1
2
Sampl e
Cucumbers
Cucumbers
Cucumbers
Untreated Brine
Untreated Brine
Untreated Brine
Treated Brine
Treated Brine
Fresh Unwashed Control After
Cucumbers Fermentation
.31 .43
.36 .34
.05 .06
.012
.013
0
Pasteurization
Treatment
.42
.35
.07
.005
.028
.002
.004
.017
Chemical
Treatment
.37
.31
.08
.019
.013
.001
0
0
-------
Data on the potential buildup of toxic substances in cucumbers as a result of
recycling is not yet completed. Results of 1975 experiments do not indicate
any buildup of compounds as a result of recycling which would limit the appli-
cation of recycling procedures. However, final conclusions cannot be made
until analysis of the second year of data is completed.
REFERENCES
1. Geisman, J.R. and Henne, R.E. Recycling food brine eliminates pollution.
Food Engr. 45(1):119 (1973).
2. Geisman, J.R. and Henne, R.E. Recycling brine from pickling. Ohio Report
58;: 76 (1973).
3. Palnitkar, M.P. and McFeeters, R.F. Recycling spent brines in cucumber
fermentations. J. Food Sci. 40:1311 (1975).
4. Etchells, J.L., Bell, T.A., Monroe, R.J., Masley, P.M. and Demain, A.L.
Populations and softening enzyme activity of filamentous fungi on flowers,
ovaries, and fruit of pickling cucumbers. Appl. Microbiol. 6:427 (1958).
5. Chavana, S. and McFeeters, R.F. Thermal inactivation of fungal pectinases
in cucumber brines. Lebensmittel-Wissenschaft Tech. (in press).
6, Bell, T.A., Etchells, J.L. and Jones, I.D. A method for testing cucumber
salt-stock for softening activity. U.S. Dept. Agr., ARS-72-5 (1955).
7. Etchells, J.L. and Moore, W.R. 1971. Factors influencing the brining of
pickling cucumbers-—Questions and answers. Pickle Pak Science 1:1 (1971).
8. Costilow, R.N., Bedford, C.L., Mingus, D. and Black, D. Purging of natural
salt-stock pickle fermentations to reduce bloater damage. J. Food Sci.
42:234 (1977).
185
-------
TREATMENT OF PACKINGHOUSE
WASTEWATER BY SAND FILTRATION
by
M. L. Rowe, Ph.D.*
INTRODUCTION
During the past five years the W. E. Reeves Packing. Plant in Ada, Oklahoma
has been used as the site for a number of wastewater treatment investigations.
These investigations have been conducted as a cooperative effort of the
Reeves' company, East Central Oklahoma State University, and the Environmental
Protection Agency. The research conducted at the site has been funded by
contracts and grants from the Environmental Protection Agency, Food and Wood
Products Branch, Corvallis, Oregon.
The investigations conducted at the Reeves' facility have been directed toward
finding treatment methods which are economically feasible and simple to oper-
ate and maintain, but which also produce a discharge which will meet effluent
guidelines. Earlier publications by Witherow (1) (2) reveal the results of
aerobic and anaerobic lagoon treatment, extended aeration lagoon treatment,
and overland flow irrigation.
However, there is a need for a variety of proven treatment methods which can
be reviewed by management, for each plant operator must consider the cost,
volume and concentration of wastewater, land requirements, climatic conditions,
etc. in order that the best method of treatment can be selected for each plant.
This paper will be concerned with the treatment of meatpacking wastewater by
intermittent sand filtration.
HISTORY OF SAND FILTRATION
The use of sand filtration for the treatment of water and wastewater is not a
recent innovation in the United States. A survey of the literature reveals
the existence of sand filtration for the improvement of drinking water supplies
in the United States as early as 1828 (3). Much of its use and further devel-
opment immediately following the introduction of the sand filter took place
in Europe. However, the population growth in the United States, especially
in the eastern cities of the United States, created a demand for larger volumes of
*Director, School of Environmental Science, East Central Oklahoma State
University, Ada, Oklahoma.
186
-------
drinking water and around the turn of the century, a number of slow sand fil-
ter units were in use in the United States for the treatment of drinking water
supplies.
Just as the population growth in the United States had created a demand for
methods of treating larger volumes of drinking water, the need arose for
methods of treating the increasing volumes of wastewater produced by the
municipalities. An experimental intermittent sand filter unit for the treat-
ment of domestic wastewater was built in Lawrence, Massachusetts in 1888 (4).
The operation of the intermittent sand filter unit proved successful. How-
ever, a rapid increase in the number of sand filter units for the treatment
of sewage was not experienced in the United States until the 1940's. The
limiting factors for the increased usage of intermittent sand filter units
were apparently the availability of natural sand sources meeting the desired
specifications and the requirement of large tracts of land.
Following World War II, the rapidly increasing number of subdivisions, mobile
home parks, and resort areas in Florida created a need for economical and
practical treatment systems which would produce an effluent of suitable qual-
ity. This need for simple and economical sewage treatment units for small
volumes of wastewater led investigators at the University of Florida to test
various designs for intermittent sand filter units (4). Much of the present
knowledge concerning intermittent sand filters has come from these early studies
at the University of Florida.
The early designs of intermittent sand filters have seen little change over the
years that they have been in use. The units usually consist of an underdrain
of open-jointed tile or perforated pipe. The underdrain network is covered
with approximately 18 inches of gravel ranging from 1/8 to 3 inches in diameter.
Filter sand is placed on the gravel at a depth that varies from 24 to 60 inches.
In the design of an intermittent sand filter, emphasis must be placed on sand
specifications. It must be a well graded sand with the proper effective
size and uniformity coefficient. The effective size is usually between 0.15
and .35 mm, and the uniformity coefficient is usually less than 3.0. Hydraulic
loading rates for sand filters vary depending on the filter media and the amount
of suspended solids in the raw wastewater. All of these factors must be
considered in order to design a filter unit capable of experiencing a feasible
period of operation before clogging of the filter media occurs.
In the past few years, workers have demonstrated the effectiveness of inter-
mittent sand filters in reducing the suspended solids values of domestic waste
which has received prior treatment in lagoon systems. Evidence of the effec-
tiveness of intermittent sand filters for the reduction of suspended solids
can be found in the published works of Reynolds (5), Marshall (6), and Walter
(7). Other supportive evidence for intermittent sand filters as a means of
lowering suspended solids values can be found in reports by Grantham (8) and
Furman (9).
PILOT SCALE OPERATION
A review of the work by the authors cited above led Witherow and Rowe to con-
sider the use of intermittent sand filters as a means of polishing the wastewaters
187
-------
from the lagoons at the W. E. Reeves Packing Company in Ada, Oklahoma. Other
investigations were being conducted at these treatment facilities at that time
which made evaluation of pilot scale intermittent sand filter units feasible.
Two pilot scale units were constructed in close proximity to the treatment
system. Each unit was formed by welding two 55-gallon barrels end-to-end.
Effluent lines were installed in the bottom of the units and each unit was
filled with gravel and sand to serve as the filter media.
The pilot scale filter units contained 18 inches of gravel, ranging from 0.25
to 1.25 inches in diameter, with the larger gravel particles in the bottom
of the unit. The gravel was then covered with 36 inches of sand which had an
effective diameter of 0.2 millimeters (mm). Investigations were then conducted
by applying wastewater from the extended aeration lagoon and the secondary
aerobic lagoon to the filter units at a rate equivalent to 0.5 million gallons
per acre per day (mgad).
Analysis of the influent and effluent samples collected during the pilot scale
study revealed a 5-day Biochemical Oxygen Demand (BODtj) removal of approximately
70 percent and a suspended solids reduction of approximately 50 percent. The
concentrations of BOD5 and suspended solids in the effluent from the pilot
units were converted to lbs/1000 Ibs of Live Weight Kill (LWK) and compared
to the Best Practical Treatment (BPT) and Best Available Treatment (BAT)
guidelines for small packinghouses. The results showed that the effluent
from the filter units could meet the suspended solids limits for BPT and BAT
guidelines. All BOD5 samples from the pilot units were below the BPT values and
only one was slightly in excess of the BAT guidelines. The results of the pilot
scale study led to the development of the large scale project at the W. E.
Reeves' facilities.
PRELIMINARY DESIGN
Since the treatment facilities had been used for previous investigations,
information pertaining to the daily volume of wastewater from the plant and the
concentration of various pollutants in the wastewater was readily available
to the investigator. The investigator also had access to information pertaining
to the average suspended solids concentration and average 8005 values of the
effluent from the extended aeration lagoon and the secondary aerobic lagoon in
use at the treatment facilities. Other information pertaining to sand and
gravel specifications, recommended loading rates onto the filters, filter
design, etc. were available from the literature and the pilot study previously
conducted by the investigator. All of this information was used in developing
the preliminary design of the filter units.
The preliminary design for the filter project was based on an average waste-
water flow of 18,000 gallons per day with a maximum daily flow of 30,000 gal-
lons. During preliminary design, a site adjacent to the existing lagoon system
was selected for the proposed sand filter units so that loading from either
the extended aeration lagoon or the secondary aerobic lagoon would be possible
by gravity flow. The proposed site for the sand filter units also had the
advantage of providing a means of diverting the flow from the filters to the
existing lagoons in case of an accidental spill or necessary filter repairs.
Figure 1 shows the layout of the facilities.
188
-------
manhole
I \$ from plant
extended
aeration
lagoon
secondary
aerobic
lagoon
stabilization
pond
to stream
Figure 1. W. E. Reeves' Treatment Facilities.
During the planning of the project, it was decided that instead of one filter
unit, two small units of unequal size should be constructed. This would offer
the advantage of having two different surface areas for the purpose of invest-
igating different hydraulic loadings and would have the added advantage of
providing one operable filter to be used while the second was being cleaned.
By following the recommendations for the bottom dimensions, slope of interior
dike, and depth of sand gravel, the completed structure consisted of two
filter units with sand surface areas of approximately 880 and 1220 square feet.
When a wastewater flow of 18,000 gallons was applied to the small sand filter,
the hydraulic loading would be 0.90 mgad, and with one-half of this average
day flow onto the small filter, the resulting hydraulic load would be 0.45 mgad.
Corresponding loading rates of 0.64 and 0.32 mgad could be obtained by using
the larger filter.
CONSTRUCTION
The filter units were formed by clay embankments on three sides, and the
fourth side of each filter unit was formed by a common concrete wall between
the two units. The purpose of the common wall was to reduce the amount of
land required for the filters. Each unit would consist of 36 inches of sand
over 18 inches of gravel and each filter unit would have a separate under-
drain system.
The bottom of the filter units was formed of six inches of compacted clay.
The bottom of each filter was sloped toward the effluent drain to insure
proper drainage from the filters.
189
-------
The embankments were constructed of clay and were built so that the interior
of the dikes were sloped at a ratio of 2.5 horizontal units to 1.0 vertical
units. The elevation from the bottom of each filter unit to the top of the
dike was a minimum of 9.5 feet, thus allowing a freeboard of at least 5 feet
above the surface of the sand. The tops of the embankments were 8 feet wide
so that vehicles and machinery could be used at the treatment site.
A factor which determined the dimensions of the bottom of the sand filter units
was the space required for the normal operation of heavy equipment, such as
a dozer, during the construction phase of the project. The final specifica-
tions for the bottom of the structure called for a wi, 1 r -f 12 feet and a
length of 42 feet. After the bottom was compacted, the 6 inch concrete wall
was to be poured to form two filter units with bottom dimensions of 12 feet
by 14 feet and 12 feet by 28 feet respectively.
The underdrain network of each filter consisted of a series of five inch dia-
meter perforated pipes on the bottom of the filter bed. Spacing of the pipes
was approximately 3 feet, and each lateral line was connected to a SJY Inch
diameter pipe serving as the main drain from the filter unit. The main drain
line projected through the walls of the dike and emptied into a sample col-
lection box. Effluents from the filters were discharged from the collection
box to the stream that runs from the Reeves' property.
A manhole already existed between the extended aeration lagoon and the sec-
ondary aerobic lagoon. A six inch line was installed from this manhole :o
a distribution box constructed in close proximity to the sand filters. A six
inch line was then installed from the distribution box to each of the filters.
By using this scheme, the wastewater from the extended aeration lagoon could
be discharged to the existing secondary aerobic lagoon or to the filters via
the distribution box. Wastewater from the extended aeration lagoon discharged
to the distribution box could be loaded into one or both of the filters.
OPERATION
The construction of the sand filters was completed in December, 1975, and op-
eration began in early 1976. During the early stages of operation, many
mechanical problems were experienced with the timers and valves, but these
difficulties were corrected and routine monitoring of the system began in
March, 1976.
The monitoring operation consisted of the collection and analysis of raw
wastewater samples, effluent from the extended aeration lagoon, and effluent
from the filter units. Composite samples of the raw wastewater and lagoon
effluent were collected, but grab samples were collected from the sand filters.
The extended aeration process was in a single lagoon and operated in a batch mode.
The aerator in the lagoon was operated from 9:00 a.m. to 11:00 p.m. Solids
in the lagoon settled for five hours and then at 4:00 a.m. a hydraulic valve
was activated discharging the supernatent from the lagoon to the filters. The
valve automatically closed after four hours.
During the first part of this investigation, March until September, the filter
media in the filters consisted of washed sand with an effective diameter of
190
-------
0.2 mm and a uniformity coefficient of 4. While this sand was in use the length
of filter runs was unsatisfactory. Loading rates of 0.55 mgad and 0.86 mgad
were evaluated in this portion of the study. For the loading rates of 0.55
mgad and 0.86 mgad, the average length of filter runs was 15 and 10 days
respectively. The filter run was the length of time from the first loading
of a filter to the time the filter was plugged and the filter was considered
plugged when the water loaded onto a filter remained on the filter surface
for more than 24 hours.
In an attempt to increase the length of filter runs, the original sand was
removed from the filter and replaced with sand having a uniformity coefficient
of 2.5 and an effective diameter of 0.35 mm. Also, the hydraulic loading was
reduced to 0.36 mgad. The increased sand size and the reduced hydraulic load
lengthened the filter run to 109 days. This portion of the study was conducted
during the period of time from October until late February. The results are
shown in Table 1.
TABLE 1. LENGTH OF FILTER RUNS
Loading Rate Sand Mean Filter Runs
(MGAD)
0.86
0.55
0.36
(Eff. Dia.)
0.20 mm
0.20 mm
0.35 mm
(Unif. Coeff.)
4.0
4.0
2.5
(Days)
10
15
109
(Number of Runs)
5
6
1
RESULTS
The wastewater samples collected during the investigation were analyzed for
a number of parameters. However, only the results for those parameters which
are of most concern to meat packers are presented in this paper. These para-
meters include 6005, Total Suspended Solids (TSS), Fats, Oil and Grease (FOG),
Ammonia Nitrogen (NH3-N), and fecal coliform. Table 2 lists the average
concentration for each of these parameters for the raw wastewater, extended
aeration lagoon, and sand filter effluent. This table also includes the reduc-
tion which was accomplished for each of these parameters by the treatment system.
The average flow for the project period was 19,756 gallons per day and the
average live weight kill was 24,617 Ibs/day.
191
-------
TABLE 2. SAND FILTRATION PROJECT
Wastewater Characteristics and Percent Removal
Parameters
BOD5 (mg/1)
TSS (mg/1)
FOG (mg/1)
NH3-N (mg/1)
Num.
46
27
16
16
Raw
Waste.
672.0
392.0
138.7
14.8
Num.
51
43
10
14
Ext. Aer.
Eff .
31.0
41.0
29.1
3.1
Num.
57
63
10
14
Filter
Eff.
10.4
11.1
<5.0
1.9
Percent
Removal
98.5
97.1
796.4
87.2
Fecal Coli.
(MPN/100 ml) 37 1.88 x 104
During the investigation, a sand with a uniformity coefficient of 4 was used
at hydraulic loading rates of 0.86 mgad and 0.55 mgad. Also, one sand source
with a uniformity coefficient of 2.5 was used with a hydraulic loading of
0.36 mgad. The average values for BOD5, TSS, NH3-N, fecal coliform, and FOG
are listed in Table 3.
Table 4 gives a comparison of the test results to the 1977 (BPT) and 1983 (BAT)
guidelines for small packinghouses. The pH and fecal coliform values are not
presented in the table. All pH values of effluent from the sand filters were
between the pH limits of 6 to 9 units with an average pH value of 7.4. A total
of 56 pH readings were taken. All coliform values for the sand filter effluent
exceeded the 400 MPN/100 ml limit with the exception of a one-day value.
The effluents from the sand filter units meet the 1977 (BPT) guidelines for
30-day average and maximum day average for BOD^, TSS, and FOG. Also, the
new source limitations for ammonia nitrogen were met.
The 1983 (BAT) limitations were met for FOG. However, the effluent did not
meet the 6005 and TSS limitations. An examination of Test B (Table 4) reveals
that the 30-day average and maximum day average for BOD5 were slightly exceeded.
Also, the TSS values of the effluent slightly exceeded the 30-day average, but
did meet the maximum day limitation. Test B was conducted between October 21,
1976 to February 23, 1977 and only eight TSS values and 10 6005 values were
collected. A longer evaluation is recommended.
COST
One of the objectives of the project was to develop a treatment system which
would be economical to construct and maintain, and one which would be simple
to operate.
192
-------
TABLE 3. RESULTS OF TESTS
to
00
Parameters
BOD5 (mg/1)
TSS (mg/1)
NHg-N (mg/1)
FOG (mg/1)
Fecal Coliform
(MPN/100 ml)
Num.
21
27
3
1
15
0.86 MGAD
Sand A Num.
10.2 26
10.6 28
2.4 6
£5.0 6
2.75 x 104 16
0.55 MGAD
Sand A Num.
10 . 5 10
11.8 8
1.9 5
-£5.0 3
•"
1.39 x 104 6
0.36 MGAD Combined
Sand B Num. Data
10.2 57 10.4
10.6 63 11.1
1.6 14 1.9
£5.0 10 £5.0
1.02 x 104 37 1.88 x 104
Sand A - uniformity coefficient of 4; effective diameter of 0.2 mm.
Sand B - uniformity coefficient of 2.5; effective diameter of 0.35 mm.
-------
TABLE 4. COMPARISON OF TEST RESULTS AND EFFLUENT LIMITATIONS
Parameter
1977 Limitations
BOD5
TSS
FOG
New Source Limitations
(include above + NH3)
NH3-N
1983 Limitations (BAT)
Packinghouse limits
30-day avg. max-day
Test Results A
30—day avg. max-day
Test Results B
30-day avg. max-day
.17
.24
.08
.24
.34
.45
.16
.48
.07
.07
<.03
.01
.17
.30
03
.03
.07
.07
<.03
.01
.10
.11
<.03
.02
BOD5
TSS
FOG
.04
.06
10 mg/1
.08
.12
20 mg/1
.07
.07
45 mg/1
.17
.30
45 mg/1
.07
.07
45 mg/1
.10
.11
45 mg/1
Test Results A - Data collected from filter at loading rates of 0.86 and 0.55 mgad and sand with uniformity
coefficient of 4 and effective diameter of 0.2 mm.
Test Results B - Data collected from filter at loading rate of 0.36 mgad and sand with uniformity
coefficient of 2.5 and effective diameter of 0.35 mm.
All values in Table 4 are in lbs/1000 Ibs LWK except 1983 FOG values.
-------
Earlier publications by Witherow (10) indicated that the cost of construction
of the extended aeration lagoon, sewer, and manholes was $14,000. The sand
filter units were constructed at a cost of $13,000. However none of these
figures include the cost of land. Additional expenses to be considered
are the cost of electricity for the aerator and maintenance cost of the
aerator.
The later part of the investigations revealed that the sand filter units could
be operated in excess of 90 days before clogging of the filters occurred.
Based on a continuous operation of 90 days for the sand filter, annual main-
tenance cost of the filter should not exceed $500. This figure allows for the
labor cost for four cleanings per year and also covers the sand replacement
costs. These cost figures are presented in Table 5.
TABLE 5. COST OF EXTENDED AERATION AND SAND FILTRATION
SYSTEM AT THE W. E. REEVES' PACKING PLANT
(a) Sewer and Manholes $ 2,000
(b) Extended Aeration Lagoon 12,000
(c) Sand Filter 13,000
(d) Annual Maintenance of Filter 500
Note: Items (a) and (b) taken from an earlier report (10). Also, cost of
land, electricity for aerator, and cost of maintenance of aerator
not included.
SUMMARY
The original objectives of the project were to develop an economical treat-
ment system capable of meeting the effluent guidelines. Cost figures presented
in this study indicate that the extended aeration lagoon and intermittent sand
filter system would meet the criteria of being an economical system to
construct.
During the early stages of operation, the test facility required much attention
and involved a great deal of manpower for frequent cleaning of the filter
units. However, after the filter media and hydraulic loading rates were
modified, maintenance and operational problems were minimal which indicates
that the system would also be economical to operate and maintain.
The effluent from the test facility met the 1977 (BPT) effluent guidelines
for all parameters, with the exception of fecal coliforms. Upon the incor-
poration of a disinfection system, the test facility would meet all the 1977
(BPT) effluent guidelines for small packinghouses. The effluent also met the
ammonia nitrogen new source limits.
195
-------
The 30-day average and maximum day average values for the 1983 (BAT) limita-
tions with respect to BOD5 and the 30-day average value for suspended solids
were exceeded slightly. The 1983 (BAT) limitations for FOG were met.
REFERENCES
1. Witherow, J. L. Small Meat Packers Waste Treatment Systems. Proceedings
of the 28th Industrial Wastes Conference, Purdue University, Lafayette,
Indiana, May 1973.
2. Witherow, J. L. Small Meat Packers Waste Treatment Systems - II. Proceed-
ings of the 30th Industrial Waste Conference, Purdue University, Lafayette,
Indiana, May 1975.
3. Daniels, F. E. Operation of Intermittent Sand Filters, Sewage Works
Journal 17: 1001-1006 (1945).
4. Reynolds, J. H., Harris, S. E., Hill, D. W., Filip, D. S., and Middlebrooks,
E. J. Presented at EPA Technology Transfer Seminar on Wastewater Lagoons,
Boise, Idaho, November 19-20, 1974.
5. Reynolds, J. H., Harris, S. E., Hill, D. W., Filip, D. S., and Middlebrooks,
E. J. Intermittent Sand Filtration for Upgrading Waste Stabilization Ponds,
Presented at Water Resource Symposium Number Nine, University of Texas,
Austin, July 22-24, 1975.
6. Marshall, G. R., and Middlebrooks, E. J. Intermittent Sand Filtration to
Upgrade Existing Wastewater Treatment Facilities, PRJEW 115-2, Utah Water
Research Laboratory, College of Engineering, Logan, Utah, February, 1974.
7. Walter, C. M. Progress Report, Blue Springs Lagoon Study, Blue Springs,
Missouri. Prepared for presentation at the Symposium on Upgrading Waste-
water Stabilization Ponds to meet Discharge Standards, Sponsored by EPA
and Utah State University, Logan, Utah, August 21-23, 1974.
8. Grantham, G. R., Emerson, D. L., and Henry, A. K. Intermittent Sand Filter
Studies, Sewage Works Journal 21: 1002-1014 (1949).
9. Furman, T., Calaway, W. T., and Grantham, G. R. Intermittent Sand Filters.
Multiple Loadings, Sewage and Industrial Wastes 27: 261-275 (1955).
10. Witherow, J., Tarquin, A., and Rowe, M. Manual of Practice - Waste Treat-
ment for Small Meat or Poultry Plants, Food Processing Waste Research,
EPA, Corvallis, Oregon, 1976.
196
-------
ECONOMIC RETURN ON POLLUTION CONTROL EXPENDITURES
FOR THE PICKLED FOOD INDUSTRY
by
James G. Meenahan, P.E.
Vice President
Johnson & Anderson, Inc.
BACKGROUND
Plants which ferment, store and pack pickled food products have a significant need for
wastewater control and treatment in order to comply with NPDES discharge permits
and municipal sewer ordinances.
The waste can be generally characterized as high in BOD^, TSS, Cl~ and Acid; also
during periods of ordinary rainfall there is a serious contamination of stormwater run-
off.
HOW DO YOU RESOLVE THESE PROBLEMS?
Detailed investigations were made of water usage and the sources of waste generation.
Wastewater treatment technology was reviewed as it might apply to the nature, volume
and frequency of the discharges.
INITIAL RESULTS
These studies showed that approximately 35% of water usage could be readily reduced
and through the use of improved housekeeping and screening, raw waste loading could
reduce by 40%.
ALAS, THE REAL SERIOUS CULPRIT SURFACES
As pickle processing plants perform more frequent and detailed wastewater characteri-
zations and as regulatory agencies perform more frequent and detailed monitoring of
receiving streams and municipal sewers, the plot thickens. It becomes abundantly clear
that there is a high level of total dissolved solids, principally in the form of chlorides.
The history of regulatory agencies, both State and Federal, shows little concern for TDS
or chlorides until the last two years. Even at this time there is little uniformity of
action. EPA has chosen to stay out of the controversy by not establishing Effluent
Limitation Guidelines for TDS or chlorides for the Flood Industry Categories. They say
that this is a matter for state regulation in accordance with each state's goals for
maintenance of water quality standards for individual receiving streams. This is
contrary to the basic EPA philosophy of "technology - based" effluent limitation
criteria.
197
-------
Most states have not made definitive analyses of the impact of IDS of chlorides upon
stream usage and have chosen to adopt the most restrictive use - drinking water. The
municipalities, in turn, have adopted the same limitation, that for drinking water, for
wastewater discharges to sanitary sewers.
Therefore, it has become apparent that several plants within the pickled food industry
must drastically reduce the chloride level in their wastewater.
HOW DO YOU ACCOMPLISH THIS?
Treatment is not the answer - distillation and reverse osmossis have high capital and
operating costs, land application is generally toxic to most vegetation and precipitation
with silver nitrate is a little ridiculous!
What's Left? Reduce it at its source. The area which contributes approximately 75% of
chlorides in the wastewater is the tank yard operation of fermentation and storage of
pickles.
The two papers proceeding this one described two major efforts to reduce the use of
chlorides: one by 3ames Harris recommending chloride reductions through the use of
controlled fermentation; the other by Dr. Roger McFeeters, recommending reuse of
spent brines by utilizing various conditioning processes to deactivate softening enzemes
and screen gross solids.
While each of these actions will drastically reduce the quantities of chlorides in waste-
water, they do not address two more sources of chloride loss - those of tank leakage and
stormwater overflow and run-off contamination.
Rainwater accumulating and overflowing the top of the open tanks not only causes
waste problems but can be disturbing to the fermentation process.
One of Johnson
-------
These process steps open the way for the following cost reduction features:
1. Clustering of tanks, thereby reducing real estate requirements by a factor of
3.0
2. Permanently sealed tops, eliminating spillage and rainwater contamination
3. Sluicing cucumbers and pickles eliminating damaged stock
b. Uniform, accelerated and controlled fermentation
5. Reuse of spent brine
6. Elimination of many forklift trucks and tote boxes and their associated
maintenance
HOW DID WE ASSESS THE ECONOMIC IMPACT?
We established the present costs for operating and maintaining the tank yard. A plant
audit was performed; all cost categories were identified; overhead and indirect costs
were allocated. This gave us a baseline for reference.
Then several cost categories were modified, eliminated or added, to reflect the
proposed conditions. The capital costs were distributed over a three year period;
depreciation and interest were distributed accordingly. The annual costs summaries
yielded the required cash flow and return on investment.
DO IT YOURSELF - BASIC PROCESS CHANGE
We feel this exercise can be repeated in many food industries by considering some of
the following:
1. Look at individual process steps
2. Ask, "Why do we do what we do?"
3. Should this operation be batch or continuous
4. What are the advantages of system monitoring
5. Where are the losses, the wastes
We all have a great reluctance to change. Change only occurs when there is a
significant potential difference between existing and projected conditions. The greater
the difference the greater the driving force, the greater the rate of change.
Crisis and subsequent threats to economic existence are excellent modivators for
change. EPA and FDA are presently providing liberal applications of these modivators.
Don't fold up your tent - try the water of process change, it can be rewarding and
exciting!
199
-------
AN EFFECTIVE WASTEWATER MANAGEMENT PROGRAM
FOR A FOOD PROCESSOR
by
George E. Wilson* and Jerry Y. C. Huang*
INTRODUCTION AND DILEMMA
The Federal Water Pollution Control Act Amendments of 1972, PL 92-500,
without doubt comprise the most far reaching pollution-control legislation
in history. The net effect of the new legislation in many cases means
increased costs to industry for treating its wastewater discharge. Under
PL 92-500, the industry is classified in two categories depending on the
final destination where the wastewater effluent discharged to. The first
and more complex situation is that in which industry discharges wastewater
effluent to a publicly owned treatment system. This type of industry is
called user industry. The other situation involves industry providing
treatment of its own effluents for direct discharge to a receiving body of
water. This type of industry is called direct discharger.
The comparison of the two effluent discharge categories is illustrated in
Table 1. For the user industry, in which the industry discharges its
wastewater effluent to a public wastewater treatment facility, there are
two constraints: industrial payback provisions and pretreatment standards.
Under this provision, no grant for any publicly owned treatment works has
been approved since March 1, 1973, unless the applicant has met the follow-
ing requirements:
a. Made provisions for payment to the applicant by industrial users
of that portion of the cost of construction which is applicable
to the treatment of industrial waste to the extent of the federal
share of the cost of construction.
b. Adopted a system of charges such that each recipient of waste
treatment services will pay its proportionate share for the cost
of operation and maintenance of waste treatment services.
In addition to the repayment of the federal share, pretreatment standards
are required by the local agency for introducing the pollutants into
publicly owned treatment works. Therefore, the permissible effluent limits
are varied depending on the requirement by the local agency.
For the direct discharger, who discharges its wastewater effluent to a
receiving body of water, there are also two constraints: effluent standard
* EUTEK, Inc., Process Development and Engineering, Sacramento, California
200
-------
TABLE 1. COMPARISON OF TWO EFFLUENT DISCHARGE CATEGORIES
Item
User Industry
Direct Discharger
ro
o
A. Effluent discharge destination
B. Constraints
C. Permissable effluent limit
Publicly owned system
Industrial payback provisions
Pretreatment standards
Varied
Receiving body of water
Effluent standards:
NPDES permit
BPT by July 1, 1977
BAT by July 1, 1983
Zero discharge by 1985
More stringent monitoring
Fixed
-------
requirements and stringent monitoring requirements.
The act calls for a National Pollutant Discharge Elimination System (NPDES)
which clearly states that any effluent discharge is illegal unless a permit
has been issued for it. By July 1, 1977, effluent limitations for existing
sources of liquid waste from industrial plants will require the application
of the best practicable control technology (BPT) currently available. By
July 1, 1983, effluent limitations require the application of the best
available control technology (BAT) economically achievable. The act
establishes zero discharge, the complete elimination of all discharges of
pollutants into navigable waterways, as a national goal for 1985. The act
contains stringent requirements on the establishment and maintenance of
records, waste stream monitoring, sampling, and analyses.
For either type of industry, the wastewater management program objective
is identical: namely,
The development of a performance guaranteed wastewater treatment
system at optimum cost.
For the user industry, the optimum cost represents the best combination of
in-plant treatment of certain waste material with the remainder treated in
the municipal treatment system. On the other hand, the optimum cost for
the direct discharger represents the least cost system which will produce
an effluent within the fixed allowable limitations.
This paper presents a phased approach for a carrot processor in developing
the most cost effective wastewater management program.
WASTEWATER MANAGEMENT ALTERNATIVES
A carrot processor, processing approximately 20 tons/hr of raw carrots,
cleans and transports the product hydraulically resulting in significant
amounts of soil solids in the wastewater. In the past, the carrot pro-
cessor discharged his wastewater directly into a nearby drainage ditch.
However, in 1975, the carrot processor was issued a cease and desist order
by the Regional Water Quality Control Board on the discharge of untreated
wastewater to a drainage ditch. The cease and desist order required the
plant effluent to have average biochemical oxygen demand (BOD) and
suspended solids (SS) concentration of 20 mg/1 with the maximum of 40 mg/1
each. The settleable matter was limited to an average of 0.5 ml/1 with
the maximum of 1 ml/1.
The management alternatives were:
1. Do nothing and shut down the plant;
2. Install treatment equipment without careful engineering
evaluation, thus risking failure of meeting effluent requirement;
3. Adopt a three-phase approach based on sound engineering evaluation
for developing a performance guaranteed system.
202
-------
The first alternative was unacceptable as the business was healthy and
worth preserving. The second alternative, however, risked the failure in
meeting the effluent requirement. The third alternative required time for
engineering evaluations, and was guaranteed to meet the requirement. The
management chose the third alternative as the one most consistent with the
long range growth plans.
OBJECTIVES AND PROGRAM DEVELOPMENT
Two objectives which had to be accomplished in the wastewater management
program were removal of settleable, suspended and colloidal materials from
the wastewater, and reuse of the reclaimed water in product cleaning opera-
tion. Three-phase program formulated for achieving these objectives was:
Phase I - Evaluation of soil solids removal and water reuse system
Phase II - Pilot evaluation of the system for suspended and colloidal
solids removal and water reuse
Phase III- Design, installation, and start-up of the full-scale solids
removal and water reuse system
Phase I - Evaluation of Soil Solids Removal and Water Reuse System
A flow diagram of the Phase I water recycle and solids removal facilities is
depicted in Figure 1. A submersible pump was placed at the bottom of the
washwater mud pit to pump the washwater through a teacup solid separator,
4 foot diameter and handling wastewater flow of 500 gpm, for soil solids
removal and hydra-vibe screen for removal of screenable vegetable material.
The water was then recycled back to the mud pit. The water in the mud pit,
which was free from the settleable soil solids, was reused in the product
cleaning and transporting operation.
The teacup solids separator, consisting of integrated teacup separator and
solid thickener compartments, has a flow regime similar to that of a
stirred teacup. Solids separation is efficiently achieved within the unit
by a combination of "teacup" secondary velocities, vortex, gravitational,
and inertial forces of the settleable particles. No hydraulic flow occurs
between the separator and the thickener, thus, the settling zone is
quiescent. Separated solids are sweeped in the quiescent zone by the
teacup secondary velocities. The teacup solid separator produced dense
settled solids and water discharged from the teacup solids separator and
the screen contained primarily colloidal material, both inorganic and
organic.
The soil solids removal performance of Phase I facilities is tabulated on
Table 2. The system was placed in operation on May 19 (Wednesday) at noon
and continued as a closed system with no addition of make-up waters through
noon Friday, May 21, 1976. Shift operations began with fresh water in the
mud pit on Saturday, May 22, and on Monday, May 24, 1976.
203
-------
HYDRA-VIBE SCREEN
ro
SCREENINGS
TEACUP DISCHARGE TO
^SCREEN (333 gpm)
TEACUP SOL IDS
SEPARATOR
GONDOLA SUPERNATANT
RETURN
SCREEN DISCHARGE SCREEN DISCHARGE
TO MUD PIT RESEVOIR
•SUBMERSIBLE PUMP
GONDOLA FOR
SETTLEABLESOLIDS
Figure 1. Flow diagram of Phase I facilities.
-------
TABLE 2. SOLIDS REMOVAL PERFORMANCE OF PHASE I FACILITIES
ro
o
Duration
of Closed Accumulated Mud Pit
Mud Pit Run SS and Colloidal Solids , SettTeable Solids Removed by Teacup
Date (8 hr shift) (mg/1) (Ibs/shift) (ft3/shift) (* solids) (Ibs/shift) («Total solids)
5/19/76 (Wed), 0.5 12.000 1.000
5/20/76 (Thurs) 1.5 23.000 640 150 74 12,745 95.2
5/21/7« (Fri) 2.5 30.000 500 150 74 12,745 96.2
5/22/76 (Sat) 0.5 1,200* 100 -
5/24/76 (Won) 1.0 400* 17 -
Ibs/ton products
-
84
83
-
-
* Fresh water was added to the cleaning operation.
All solids weight are expressed in dry weight.
-------
The soil loading contributed to the wastewater varied daily depending on the
amount of soil solids associated with the incoming product. As shown in
Table 2, the accumulated suspended and colloidal solids per shift varied
considerably, ranging from 17 to 1,000 Ibs/shift. The portion of settleable
solids removed by the teacup represented approximately 95% of the total
solids contributed to the wastewater flow. The amount of total solids (dry
weight) was about 83 Ibs/ton of product, representing approximately 5% of
the weight of the product processed.
The soil solids removal system effectively removed almost all the settle-
able solids from the system, which eliminated the discharge of the soil
solids to the drainage ditch and the water free from the soil solids was
reused for product transport and cleaning.
Phase II - Pilot Evaluation of the System for Suspended and Colloidal Solids
Removal and Water Reuse
The objective of the Phase II pilot study was to determine the most cost
effective means for producing water for high quality water reuse and for
meeting the effluent requirement.
The pilot apparatus consisted of chemical feeding pump and diffuser, decayed
gradient flocculator, gravity sedimentation tank, and granular media filter.
The screen discharge water was coagulated by decayed gradient flocculator,
settled in the gravity sedimentation tank, and finally polished by the
filter. The summary of the pilot filtration performance is presented in
Table 3.
Two sizes of filter media were investigated, one passing through 1.2 mm and
retained by 0.8 mm openings, and another passing through 3.2 mm and
retained by 1.6 mm openings. In filtering wastewater of high solids
loading rate, it was found that the media size of 1.6/3.2 mm (retaining/
passing) was more effective in producing high quality filtrate with a
reasonably long filter run. It was also determined that the most effective
mode of chemical addition was to inject the chemical coagulant immediately
upstream of the filter media. The coagulation-flocculation and filtration
of the colloidal solids occurred within the filter bed, eliminating the
chance of surface cake formulation on top of the filter media. Therefore,
excessive pressure drop across the filter media was eliminated.
The characteristics of the filtered wastewater varied, from recirculated
wastewater to wastewater with fresh water added. The filtration rates
tested ranged from 2.5 to 10 gpm/ft2- The coagulant used was cationic
polymer, trade name Cat-Floe (product of Calgon) and the chemical feeding
rates varied, ranging from 20 to 100 mg/1. The influent turbidity ranged
from 40 to 300 FTU (Formazin Turbidity Unit) and the filtrate turbidity
ranged from 2 to 50 FTU. A significant reduction in turbidity was achieved.
Therefore, with the proposed clarification and filtration system, the
reclaimed water should be suitable for reuse as cull pit and/or mud pit
make-up water.
206
-------
TABLE 3. SUMMARY OF PILOT FILTRATION PERFORMANCE
Coagulant Cone. (Cat-Floe) Filter Performance
Filter Filtration Wastewater Coag.-Floc. Contact Pressure Run Reasons for Run
Date Media Rate, Characteristics -Sediment Filtration Influent Effluent Drop Length Termination
(urn) (gpm/ftz) '(mg/D (mg/1) (FTU)a (FTU)a (ft. water) (hr)
5/21/76 0.8-1.2 5 2-3 day's 100
recirculated water
ro
0
50 170-210 10-28 37 1 deteriorated effluent
quality and
excessive pressure
drop
5/22/76 0.8-1.2
fresh water added
50
150
17-24
32
deteriorated effluent
quality and
excessive pressure
drop
5/22/76
5/24/76
0.8-1.2
1.6-3.2
2.5
5
10.5
15
fresh water
first day's
reclrculated
added
water
0
20
20
20
50
6
6
0
290
42-58
70
52-90
2-10
9-15
9-20
20-50
41
0
0
16
2
2
1/2
1 1/2
excessive pressure
drop
deteriorated
effluent
quality
Formazln Turbidity Unit
-------
Phase III - Design, Installation, and Start-up of the Full Scale Solids
Removal and Water Reuse System
Based on the results of the Phase II pilot studies, the full scale solids
removal and water reuse system included:
1. Decayed gradient flocculator/gravity clarifier; and
2. Coarse media contact filter
The flow diagram of the complete system is shown in Figure 2. The portion
of final polishing was designed to handle a continuous flow of 100 gpm of
screen discharge wastewater. The layout of the proposed facility takes
best advantage of the limited available space without infringing on traffic
patterns.
The decayed gradient flocculator is a series of enclosed square channels
coiled around the 6 foot diameter gravity clarifier. The size of the
enclosed channels increases gradually downstream such that the most inten-
sive chemical mixing occurs at the upstream at the point of chemical
injection and the mixing intensity decreases as the floe travels downstream
to enhance the flocculation. The flocculator/clarifier was elevated to
gravity discharge the settled liquid slurry to haul equipment.
The contact filter was designed as a constant flow device, capable of
accommodating a 10 foot pressure drop over the bed prior to backwashing.
A novel backwash system allowed use of coarse filter media, thereby
assuring effective solids removal with minimum buildup of pressure drop.
Following the completion of each filtration cycle, the filter would be
backwashed with transfer of filter solids to the flocculator/clarifier for
final thickening and disposal.
The typical performance of flocculation/clarification and filtration system
is presented in Table 4. The turbidity, SS, BOD, and COD of the screen
discharge wastewater were 600 to 1,000 FTU, 1,200 to 5,300 mg/1, 20 to
55 mg/1, and 100 to 240 mg/1, respectively. In comparing the wastewater
characteristics of the effluent from the flocculation/clarification with
that of screen discharge, a significant reduction in all the parameters was
noted. Further reduction of all the parameters was achieved by the filtra-
tion. The turbidity, SS, BOD, and COD of the effluent from the filter were
7 to 20 FTU, 10 to 30 mg/1, 5 to 10 mg/1, and 25 to 50 mg/1, respectively.
The quality of the filtrate meets the discharge requirement and is suitable
for reuse in the carrot processing operation.
SUMMARY
This plant, a direct discharger, installed a process water recycle and final
effluent polishing system which produces a treated water quality commensur-
ate with the plant's potable water. By taking the phased approach in
developing an effective wastewater management program, the carrot processor
obtained a performance guaranteed system. An approximate 35% savings in
the total investment was realized relative to the comparable system.
208
-------
COAGULANT
FLOCCULATOR/CLARIFIER
BACKWASH FLOW
COAGULANT
CONTACT FILTER
SUMP &CHLORINATOR
(OPTIONAL)
TEACUP SOLIDS
SEPARATOR
.-SCREENINGS
FRESH WATER
MAKE-UP AS
REQUIRED
MAKE-UP
QUID FLOW
•SOLID/SLURRY FLOW
'OPTIONAL
OVERFLOW TO
DRAINAGE DITCH
Figure 2. Flow diagram of Phase III facilities.
209
-------
TABLE 4. TYPICAL PERFORMANCE OF FLOCCULATION/CLARIFICATION AND FILTRATION SYSTEM
Influent to Clarifier
Turbidity
(FTU)
600-
1,000
SS
(mg/1)
1,200-
5,300
BOD
(mg/1)
20-
55
COD
(mg/1 )
100-
240
Effluent from
(Influent to
Turbidity
(FTU)
40-
150
SS
(mg/1 )
50-
180
Clarifier
Filter)
BOD
(mg/1 )
6-
11
COD
(mg/1 )
28-
50
Effluent from Filter
Turbidity
(FTU)
7-
20
SS
(mg/1)
10-
30
BOD
(mg/1)
5-
10
COD
(mg/1)
25-
50
-------
RECOVERY OF SOLUBLE SERUM PROTEINS
FROM MEAT INDUSTRY WASTES
by
Richard W. Greiling*
INTRODUCTION
Wherever animals are slaughtered, decisions must be made on how best to
dispose of the blood. For most meat and poultry packers the options are few.
There was a time when the blood could be discharged into any receiving body
of water along with other wastes. That time is gone. As effluent limita-
tions became stricter, blood wastes in most abattoirs were retained for
treatment with other renderable materials.
Blood is the largest single waste originating on the killing floor and, in
terms of BOD, the strongest pollutant in the entire meat packing industry.
Some studies suggest that 5-day biological oxygen demand (BOD,.) reductions
of greater than 40 percent can be achieved by shifting from a no-recovery
to a recovery of all blood wastes. In 1967, 80 percent of all packing plants
in the United States were recovering blood. On a live-weight-kill (LWK)
basis, 96 percent of the blood from slaughtered animals was recovered and
treated in some form(l).
Standard sizes for blood recovery equipment are 5,000 Ibs/hr (2,275 kg/hr)
and 10,000 Ibs/hr (4,550 kg/hr). Most recovery systems operate at a blood
feed rate of 10,000 pounds per hour. In the most common blood recovery
system blood is pumped through a steam-jacketed heat exchanger ranging in
length from 10 to 20 feet (3 to 6 meters). To obtain a blood temperature
of 90 C, at which rapid coagulation occurs, requires a blood retention time
of 15 to 20 seconds, and 550 pounds (250 kg) of steam per hour. The blood
clots are then centrifuged out of suspension for drying (spray drying is
common) and packaging; the centrate being set to evaporation facilities or
discharged to the sewerage system.
The biochemcial behavior of blood proteins in pure solutions has been under-
stood for several years. The biochemical relationships, and influencing
factors such as pH, ionic charge, temperature, and solids concentrations
have been reported. What the literature fails to present is the biochemical
behavior of blood proteins in a heterogeneous solution that has been sub-
jected to the thermal and physical treatment mechanisms associated with
blood recovery facilities.
*Washington Department of Ecology, Olympia, Washington
211
-------
DESIGN CONSIDERATIONS FOR PROTEIN RECOVERY
A minimum of constraints were placed on the development of a serum protein
removal process. Aware of capital limitations that confront many abattoirs,
certain factors concerning any final recommendation for a pre-treatment and
recovery facility were of fundamental importance and necessitated constant
consideration throughout the investigation.
First, the initial cost of facility construction must be low. Approximately
25 percent of the nation's slaughtering is done in small establishments(2)(3).
A truly effective treatment scheme will be one that can be utilized by all
meat packers. Low capital investment is a constraint on such a facility
design.
Second, the process must be simple and relatively "foolproof". Operation
should not require careful control or unpleasant tasks. Good treatment
results should be visable to encourage the operator and convince him that his
plant is really accomplishing its purpose.
Third, mechanical equipment should be held to a minimum to prevent shutdown
due to failure and to cut maintenance costs. Duplicity of equipment should
be avoided. Processes utilizing chemical and energy inputs should operate
at the most optimum conditions to minimize operational costs.
OBJECTIVES OF THE INVESTIGATION
The literature suggests that increasing the temperature of a protein solution,
particularly at its isoelectric point, enhances protein precipitation. Can
the increased yield, however, justify the increase in thermal inputs? In a
heterogeneous protein solution is there an optimum pH at which precipitation
is most likely to occur? Is there a quantitative limit to protein recovery
due to protein solubility relationships? And how close can a protein
recovery system approach a limit of solubility and still be justified by
economics?
The stimuli for effective blood recovery are a need for product recovery,
and a determined need on the part of industry for effluent abatement. Perhaps
the single most motivating force which will induce industry to consider
further improved protein recovery is the energy costs associated with present
recovery facilities. Most recovery systems are designed such that the liquid
fraction of blood is mixed with other tank waters and dried by evaporation.
Threatened unavailability of natural gas and fuel oil and rising energy costs
associated with evaporators and driers are already forcing several abattoirs
to change their rendering operations. One system improves centrifugation of
blood particulates yet still discards the centrate to the sewer system(4).
212
-------
What his facility fails to recover is the high quantity of dissolved serum
proteins which represent a salable product.
Another objective of this research was to produce as pure and natural as
possible a protein by-product which can be marketed. Fractionated proteins
in a pure state yield the highest market price. However, facilities for such
a treatment process are inaccessible to all but a few of the nation's meat
packers. To meet the constraints previously discussed it became apparent
that singularly or in combination heat, chemical, and physical methods may be
necessary for protein extraction. The resultant protein solid is no longer
in its natural state and losses the qualities necessary for pharmaceutical
and highly specialized protein applications. It is cautioned, too, that the
heat and pressure conditions found in an autoclave may make certain labile
amino acids unavailable for nutriative purposes.
The markets of animal feed and fertilizer could be the receiver of such a
controlled recovery product. Because of the pressing need for protein, it
was hoped that a product could be recovered which can be marketed as an
animal feed supplement. This objective is constrained because the Federal
Food and Drug Administration does not allow the feeding of ruminants or
other animals any feedstuffs chemically removed with synthetic polyelectro-
lytes(5). This constraint led to the investigation of the serum protein
recovery capabilities of the organic polymer chitosin.
PRELIMINARY INVESTIGATIONS
Preliminary investigations were performed on a blood serum waste to asess
the physical and chemical dependency of protein denaturation upon pH,
temperature, protein concentration, and the presence of a coagulant aid.
The intent of the investigations was to reduce the number of variables in
question. A final design could be performed which would then provide data
that could be used for the development of an operational protein recovery
facility.
This blood serum waste stream, pH 6.7 to 6.9, has an organic nitrogen
concentration of approximately 1,630 mg/1. Total solids are about 23,000
mg/1 of which 15,000 mg/1 are volatile. Assuming serum protein is 6.4 times
the measured nitrogen assay at least 10,500 mg/1 of the volatile solids
fraction is attributable to serum proteins(6),
Variable Selection
The literature suggests that pH and temperature most strongly influence
protein denaturation. Either variable alone can induce denaturation and
coagulation, but no data have been reported on their combined influence on
protein denaturation in a blood serum waste stream.
Protein denaturation occurs most rapidly at the isoelectric point of the
solution. Consequently it was necessary to determine that pH point and
then bracket it for subsequent analyses of protein recovery. The isoelectric
point of the serum waste stream was to be determined by microelectrophoresis.
213
-------
Denaturation is accelerated at elevated temperatures. Several temperature
levels were used for preliminary investigations to determine the relation
between temperature, protein solubility, and protein removal.
The literature suggests that protein stabilization against denaturation is
inversely proportional to concentration. The effect of protein concentration
on removal efficiencies was studied. In a recovery operation will the
addition of wash-down waters sufficiently dilute the serum waste stream
therefore hindering protein recovery?
Chitosin, an organic polymer composed of glucosamine residues, was used to
help remove the suspended protein aggregates from solution. Chitosin has
been shown to be an effective coagulant aid with protein-containing wastes(7).
Approval for its use as a polymer for industrial applications and subsequent
marketing in feedstuffs is pending before the Food and Drug Administration(S).
Determination of the Blood Serum Isoelectric Point
Protein mobility within an electric field was determined by. measuring the
time required for the aggregate to travel across a Howard counter microscope
grid. Protein mobility is dependent upon the pH. of the solution for the
net ionic charge of the protein molecule changes with pE.
Microelectrophoresis was used to determine the isoelectric point of this
heterogeneous solution. The sample was diluted with distilled water to make
a 2.5 percent solution. Samples of the serum dilution were drawn and the
pH randomly adjusted by using either 0.1 N KOH or 0.1 N HC1. The sample
was then placed in the Briggs cell and a portion drawn thru the cell. Voltage
was applied to the cell electrodes and the time of travel for a protein
aggregate was recorded. After each set of observations, p'H and specific
resistance of the sample, and line current were recorded. For each pH the
electrophoretic mobility was calculated.
The observed electrophoretic mobilities for a 2.5 percent serum solution are
plotted against ph (fig. 1). The isoelectric point is where protein mobility
is zero. For this 2.5 percent solution it occurs near pH 4.6 to pH 4.7.
Denatured Protein Settling Tests
Three settling tests were conducted to determine if denatured proteins would
settle out of suspension. A stock solution of the serum waste and distilled
water (for dilution) were heated to 60°C. Protein solutions were made up in
serum : distilled water ratios of 1:0.25, 1:0.5, and 1:1. Each solution was
adjusted to pH 4.6 and poured into a graduated cylinder.
Organic nitrogen and volatile solids tests were conducted on the supernatant
in each column over a 24 hour period. All reported concentrations are
adjusted for dilution. Solids concentrations were also determined on the
settled solids at 24 hours.
A plot of residual organic nitrogen (fig. 2) suggests that blood serum protein
removal by pH denaturation is independent of the concentration to at least a
214
-------
ro
en
-1.0
o
0)
CO
e
u
- .5
•§
G + .5
•H
-------
1800
1600
1400
so
e
\~s
a 1200
a)
t>0
O
S-i
4J
O
•H
§1000
oo
j-i
O
T3
•i-l
CO
a>
*« 800
100
Serum Dilution Tests:
A = (1:0.25)
B = (1:0.50)
C = (1:1.00)
34567
Settling Time (hrs.)
Figure 2. Organic Nitrogen Residuals for Three Serum
Dilution Ratios at pH 4.6,
216
-------
dilution of one. Within one. hour all settleable solids have been removed
from suspension.
At the end of 24 hours, a sample of the supernatant was centrifuged for five
minutes at about 1,500-g. No additional removal of residual nitrogen
concentration was observed. It was concluded that gravity settling for
one hour will remove nearly all of the denatured protein aggregates.
Centrifugation of the chemically treated serum waste stream would not be
necessary to remove the denatured proteins.
At 24 hours, the settled solids are approximately 7 percent solids by weight
and are 93 percent volatile. Visual examination of the graduated cylinders
would suggest the settled solids to approximate 10 percent of the initial
blood serum waste stream volume. The solids data indicate that few of the
inorganic dissolved solids are being removed by either chemical complexation
or entrainment in the protein aggregate. This suggests that dewatering of
the settled solids could lead to a highly pure product with very little
ash content.
Frotein Recovery in a 2-Level Factorial Design
In an attempt to reduce the number of variables an investigation was conducted
to determine the effect of pH, temperature, and protein concentration on
recovery efficiency. The settling tests suggested that recovery was
independent of concentration at low dilution. It was felt a factorial design
would statistically varify that fact.
f\
A 2 factorial design was developed. Temperature levels were 20 C and 60 C.
The lower temperature was selected for convenience and the higher temperature
approximated the temperature of the serum waste stream of the centrifuge from
which the samples were obtained. pH 4.6 was chosen for its proximity to the
isoelectric point and pH 5.1 as a point in the range of high anionic mobility.
Protein concentration levels selected were raw serum (dilution 1:0) and a 50
percent concentration (1:1).
Table 1 presents nitrogen and volatile solids concentrations and the observed
standard errors for the design corner points. Standard errors for both
nitrogen and solids concentrations are within the accuracy of the analytical
procedures. To determine the significance of the main effects, an F-test
with a 95 percent confidence level was conducted (table 2) . At that level,
Both organic nitrogen and volatile solids data suggest that the main effects
of pH and temperature are statistically significant. The third level inter-
action is significant in nitrogen assays. The pH and temperature second
level interaction and the third level interaction are significant in volatile
solids data. Neither the main effect nor the second level interactions of
protein concentration is significant. This lack of significance suggests the
variable may be dismissed from subsequent investigations.
Both nitrogen and volatile solids results suggest that for the levels under
investigation protein recovery is much greater near the isoelectric point
217
-------
"TABLE 1. ORGANIC NITROGEN AND VOLATILE SOLIDS CONCENTRATIONS FOR 2 FACTORIAL DESIGN
ro
NITROGEN RESIDUAL
No. Average St. Error
%St.E.
Test Obsv. mg/1
Raw
I
2
3
4
5
6
7
8
pH
T
I)
pHx!
pHxD
TxD
PJJKTXD
3
3
4
3
3
4
3
3
3
1,570
1,070
907
1,200
985
1,210
111
1,280
1,200
S 2 = 1,950
P
Analysis of
-223
+175
+ 76
+ 75
- 34
+ 71
+101
40
46
52
12
54
53
29
26
60
St. Error
Effects
2*5
4.3
5.7
1.0
5.5
4.4
3.7
2.0
5.0
- 44.1
No.
VOLATILE SOLIDS RESIDUAL
Average St. Error
ZSt. E.
Obsv. mg/1
3
3
4
3
3
4
3
3
3
12,500
9,560
8,390
10,400
9,050
10 , 700
7,200
10,900
10,500
S 2 = 103,960
P
Analysis of
-1,610
+1,250
+ 475
+ 730
- 345
+ 500
+ 820
350
510
510
120
250
170
310
110
330
St. Error
Effects
. 2.8
5.3
6.1
1.1
2.8
1.6
4.3
1.0
3.1
- 322
-------
than at higher pH. Both tests suggest that a temperature of 60°C decreases
the amount of protein recovered compared to that at 20°C. This result is
opposite that expected. It may be hypothesized that protein solubility has
increased with temperature and that the temperature selected was not high
enough to induce thermal denaturation. Factorial design analysis suggested
that thermal denaturation must occur above 60°C. This hypothesis is
supported by later investigations.
The recovery of soluble proteins has again been shown to be independent of
concentration up to a dilution of 50 percent. For this reason, the protein
concentration was dismissed as an experimental variable. This conclusion
suggests that small volumes of wash waters can be added to blood recovery
operations without adverse effects to any subsequent protein recovery. From
an operational viewpoint, volumes should be held to a minimum. Subsequent
investigations were conducted using undiluted blood serum wastes.
TABLE 2. SUM OF SQUARES AND F-TEST FOR 23 FACTORIAL DESIGN
ORGANIC NITROGEN
Effect
j£
!
p_
£HxT_
£HxD_
TxD
pHxTxD
S
T
Sum of Squares
49,729
30,625
5,776
5,625
1,156
5,041
10,201
2
1,950
D.F.
1
1
1
1
1
1
1
F
25.50
15.70
2.96
2.88
0.59
2.59
5.23
VOLATILE SOLIDS
Sum of Squares
2,576,025
1,562,500
225,625
532,900
119,025
250,000
672,400
2
S 103,960
D.F.
1
1
1
1
1
1
1
F
24.78
15.03
2.17
5.13
1.14
2.40
6.46
Establishment of the Polymer Dose
Investigations were conducted to determine the effectiveness of chitosin as a
coagulant aid for the removal of suspended denatured proteins. Two investi-
gations were conducted at 75°C and pH 4.65 and pH 5.4 Organic nitrogen and
volatile solids analyses were performed on all samples. A subsequent investi-
gation was conducted at 80 C and pH 4.65 to look at chitosin dosage levels and
to determine any change in protein recovery due to a slightly higher tempera-
ture. Only organic nitrogen concentrations were measured.
Results for nitrogen concentrations plotted against chitosin dose (fig. 3)
suggest that protein removal is enhanced with increasing polymer dosage up to
219
-------
Temp
1 75°C
Trial 2 75°C
Trial
Trial 3
80°C
pH
5.4
4.65
4.65
50 100 150' 200
Chitosin Dose (mg/1)
250
Figure 3. Organic Nitrogen Residual vs. Chitosin Dose.
220
-------
a chitosin concentration of about 100 mg/1. Beyond that dosage there is a
leveling off of the yield of protein recovered. The proteins remaining in
solution appear no longer influenced by the presence of the chitosin.
In order to establish the range of pH values over which any final analyses
should be performed, a series of investigations was conducted at 90 C.
pH ranged from 5.8 to 3.8 at intervals of 0.4 pH units.
Figure 4 suggests that protein recovery is strongly influenced by both pH
and chitosin dose. Two significant observations may be noted. First, at
low chitosin concentrations, the maximum recovery of proteins occurs well
below the isoelectric point of the protein solution; protein yield is
greatest at pH 3.8. Second, at a chitosin concentration of 100 mg/1
protein recovery is maximized over a range of pH 5.0 to pH 3.8. The
recovery has been maximized for the given variables and that the residual
protein concentration is most likely limited by protein solubility.
EXPERIMENTAL DESIGN AND CONCLUDING INVESTIGATIONS
Selection of Design Variable Levels
It has been determined that a minimum temperature of about 60 C is required
for thermal denaturation of proteins. Theoretically, thermal denaturation
of all proteins occurs at 100 C. The variable levels for temperature were
thus limited by these two temperatures. Complications arise with operating
temperatures near the boiling point of the serum solution. Excessive steam
would create operating problems. Heat inputs would be nearly as great as
present evaporation facilities. As such, a maximum operating temperature of
90°C was hypothesized. Temperature levels to be investigated were 60 C, 75 C,
and 90°C.
Preliminary investigations suggest that protein recovery is greatly improved
with polymer doses of 20 to 100 mg/1. Chitosin doses of 20, 60, and 100 mg/1
were used in the final design.
A strong interaction between pH and chitosin dose suggest that a wide pH
range can yield maximum protein recovery. Preliminary investigations suggest
this pH range to be from pH 5.0 to at least pH 3.8. In an attempt to fully
bracket the pH range, final investigations were conducted from pH 5.4 to
pH 3.4. Six levels 0.4 pH units apart were selected for observation.
Procedure
The final design is a matrix with variable levels of 3 x 3 x 6. The design
was to be run in duplicate, with any remaining serum samples being used to
triplicate the experiment. The order in which samples were run was only
partially randomized. Because of equipment restraints, only one temperature
level could be conducted at a time. Both pH and polymer dose levels were
randomized for a particular temperature level.
A large serum sample was heated to the desired temperature. A 300 ml sample
was withdrawn and pH adjusted at random. A 100 ml sample was then added to
221
-------
Organic Nitrogen Residuals
10
§ 9
,
2 6
o
•H
20
Raw Serum Org.N. = 1,580 mg/1
3.8 4.2
4.6
5.0 5.4
5.8
Volatile Solids Residuals
10
l»
•3 6
•H
0)
H
•H
>
Raw Serum V.S. = 12,500 mg/1
3.8 4.2 4.6 5.0
5.4
5.8
FIGURE 4. Residual Organic Nitrogen and Volatile Solids Concentrations
as a Function of pH and Chitosin Dose at 90°C-
222
-------
each of three bottles containing the amount of chitosin to yield the desired
polymer concentration. The bottle was stoppered and placed in the warm air
oven for a reaction and settling time of one hour. After clarification the
supernatant was tested for organic nitrogen and volatile solids.
A total of 134 tests was conducted. This provides for triplication of about
one half of the design matrix. All other test conditions were duplicated.
Results
Protein recovery was determined as percent of nitrogen or volatile solids
removed from solution and suspension. Protein recovery efficiency varies
from 31 to 62 percent. Efficiencies are slightly higher for organic nitrogen
assays than for volatile solids. Earlier mass balance approximations
suggested that proteins accounted for about 70 percent of measured volatile
solids. The solubility of dissolved salts will limit the efficiency of
volatile solids removal. This limitation will tend to depress the percent
reduction of volatile solids independent of the reduction of dissolved
organic nitrogen.
Regression Analysis of Protein Recovery Efficiencies
Isometric projections of protein recovery efficiencies (as measured by
organic nitrogen) were drawn to assist data interpretation (figs. 5-7).
The projections suggest recovery is highly dependent upon pH and polymer
dose and nearly independent of temperature.
Recovery is most effective in the range of pH 3.8 to pH 4.6. Within that
range recovery is nearly independent of the chitosin dose. At pH values
above 4.6 the recovery is highly dependent upon polymer dose. Both of these
observations confirm earlier investigations into the effectiveness of pH as
a denaturant and chitosin as a coagulant aid. That recovery efficiencies
do not significantly improve with increasing temperatures is contrary to
earlier hypotheses. Protein solubility in the high dissolved salt solution
must be limiting protein recovery efficiencies.
To facilitate ease in using experimental results an attempt was made to
develop a model which would predict protein recovery efficiency as a function
of pH, temperature, and chitosin dose. Isometric projections of organic
nitrogen removal efficiencies suggest a second order response to both pH and
dose. A linear response to temperature is suggested by the three projections.
Several models were used in a multiple regression analysis. The model was
to fit an equation through the 54 data points collected in the test matrix.
The parameters (T) and (T x Dose) were insignificant. A six-parameter model
(linear in the parameters) for predicting the removal efficiency for organic
nitrogen is:
F = 0.54(T) + 52.0(pH)2 - 6.56(pH)2 - 0.44(Dose) - 0.09(pH x T)
+ 0.12(pH x Dose) - 59.87
where F is organic nitrogen removal efficiency as percent (see table 3).
223
-------
60
Figure 5. Organic Nitrogen Removal Efficiencies as a
Function of pH and Chitosin Dose at 60 Ct
224
-------
T)
Q)
>
O
0)
00
O
t-1
4J
c
0)
a
M
0)
PM
30
Figure 6. Organic Nitrogen Removal Efficiencies as a
Function of pH and Chitosin Dose at 75 C.
225
-------
60
Figure 7. Organic Nitrogen Removal Efficiencies as i
Function of pH and Chitosin Dose at 90°C.
226
-------
F = 02 (T) + 93(pH)
04(pH)2
05(Dose) + 06(pH x T) + 07(pH x Dose) + 0,
ro
ro
Variable
No.
2
3
4
5
6
7
Mean
7.500QE+01
4.4000E+00
1.9827E+01
6.0000E+01
3.3000E+02
2.6400E+02
Standard
Deviation
1.2363E+01
6.8954E-01
6.0814E+00
3.2967E+01
7.5529E+01
1.5251E+02
Correlation
X vs Y
2.5340E-01
-5.6012E-01
-5.8526E-01
4.1808E-01
-2.1404E-01
3.0550E-01
Regression
Coefficient
5.3847E-01
5.1984E+01
-6.5528E-01
-4.4068E-01
-9.1489E-02
1.1927E-01
Std. Error
of Reg.Coef.
1.7002E-01
7.6656E+00
8.0027E-01
6.3758E-02
3.8184E-02
1.4319E-02
Computed
T Value
3.1670E+00
6.7815E+00
-8.1883E+00
-6.9117E+00
-2.3960E+00
8.3293E+00
Dependent
1 5.4172E+01
Intercept
Multiple Correlation
Std. Error of Estimate
6.6312E+00
-5.98778E401
9.42796E-01
2.34752E-WO
Analysis of Variance for the Regression
Source of Variation
Attributable to Regression
Deviation from Regression
Total
Degrees of
Freedom
6
47
'53
Sum of
Squares
2.07156E+03
2.59010E+02
2.33057E+03
Mean
Squares
3.45260E+02
5.51085E+00
F Value
6.26510E+01
TABLE 3. REGRESSION ANALYSIS FOR ORGANIC NITROGEN REMOVAL EFFICIENCY
-------
The correlation between the variables and the removal estimates are
intuitively satisfying. There is a small, but significant, positive
correlation with temperature. The greatest correlations for the model
related estimated protein recovery with pH. For the pH range under study,
the lower the pH the greater the recovery. There is a fairly large
positive correlation between chitosin dose and product recovery for both
chitosin in the first order and its interaction with pH.
Dissolved Protein Removal by Carbon Adsorption
A laboratory investigation was conducted to determine if dissolved protein
removal, beyond that recovered by denaturation and polymer coagulation,
could be achieved with activated carbon. The extension of carbon capacity
may be very significant under certain conditions. Decreasing pH and
increasing temperatures increases the adsorptive characteristics of
activated carbon. These two conditions are attained in the chemically
treated blood serum waste, and the high dissolved organic content should
enhance surface adsorption.
A preserved serum waste was heated to 60 C, pH adjusted to 4.6, and mixed
with chitosin at a dose of 100 mg/1. After one hour of settling the
supernatant was withdrawn and 300 ml samples were added to six test vessels
containing measured amounts of powdered carbon. The stirrer and carbon
suspensions were maintained at 60 C in a heated room. Carbon suspensions
were stirred at 40 rpm for a two hour activated carbon residence time.
A five minute centrifugation at about 3,600-g provided excellant solid/
liquid separation.
Organic nitrogen, chemical oxygen demand (COD), and volatile solids analyses
were conducted on the raw serum waste, the waste chemically treated for
protein recovery, and on the six carbon adsorption centrates (table 4).
Analyses were conducted to determine carbon adsorption capacities for
protein removal.
The adsorptive capacity for the powdered carbon was determined by comparing
the amount of COD removed per unit of carbon versus the COD loading rate.
The total COD used in the adsorptive capacity test was determined by
measuring carbonaceous COD and adding to that the theoretical oxygen
demand of the organic nitrogen concentration.
When the adsorptive capacity of the activated carbon is plotted against
residual COD (fig. 8), the curve shows that carbon adsorption removal
improves with increasing concentration of dissolved organics. It is noted
here, too, that as the concentrations of dissolved organics increase, the
rate of unit increase in adsorptive capacities is less that unity.
That protein removal by activated carbon is significant leads to the
development of unit processes beyond protein recovery. Spent granulated
carbon from removal operations involving low organic concentrations may yet
have sufficient adsorptive capacity to polish the organic-rich supernatant.
Following protein recovery by denaturation and coagulation, the supernatant
228
-------
ro
ro
vo
Sample
Raw
Serum
Treated3
Serum
1
2
3
4
5
6
a.
b.
c.
Carbon
Dose
g/1
4
8
20
40
80
160
pH = 4
Total
mg COD
COD Org. N
mg/1 mg/1
12,400
7,780
7,080
5,860
3,150
2,300
1,090
804
1,760
680
407
274
209
149
71
52
Total COD T.S. V.S. mg COD1
mg/1
20,400
10,900
8,940
7,720
4,110
2,980
1,410
1,040
mg
mg/1 mg/1
23,800 16,200
8,420 3,990
8,630 3,560
7,770 2,680
7,160 2,040
6,670 1,530
6,570 1,150
7,130 1,080
Carbon
2.73
1.36
0.55
0.27
0.14
0.07
mg COD
mg Carbon
0.49
0.40
0.34
0.20
0.12
0.06
.6; T - 60°C; Chitosin = 100 mg/1
COD - Carbonaceous COD + 4. 57 (Org. N concentration)
/mg Carbon = 10,900 mg COD/(X mg Carbon)
TABLE 4. CARBON ADSORPTION COD REMOVAL
-------
0.6
0.5
a 0.4
o
-a
cfl
u 0.3
TJ
-------
protein structural deformation and also because of the hazards associated
with working with strong acids. Thus both pH and polymer mixing tanks and
feed systems will be required. Some heat exchange unit (steam jacket around
the tank) will be necessary to maintain elevated temperatures sufficient
to induce thermal denaturation. The tank can be covered to minimize the
release of odors and loss of heat.
Cost estimates assume a large slaughterhouse kills 5,000 hogs per day and
each hog loses one gallon (3.7 liters) of whole blood(8). Following
centrifugation of coagulated blood cells, the serum is sent to a 1,000 cubic
foot (29 cubic meter) holding tank. Temperature is maintained between 60 C
and 70 C. After slaughtering is complete, the serum temperature is raised
to 70 C, pH adjusted to 4.2, and chitosin added to enhance coagulation.
After a reaction and settling time of four hours (as determined by particle
settling velocity) settled proteins are withdrawn for evaporation and the
supernatant is wasted. The tank and lines are cleaned, wash waters being
sent to the sewer.
Because of corrosive acidic conditions, the entire system is made of
stainless steel. Both pH and chitosin solutions are made once a week. The
small daily requirements make this possible. Chemical treatment is done on
a batch basis at the end of the work day. The operator reads the level of
serum in the settling tank and manually adjust the polymer feed for the
desired dosage. pH is monitored and is either manually or automatically
adjusted. Valves are located in several lines to direct waste and product
stream flows.
A cost estimate for the capital investment suggests this design to cost
approximately $17,000. Operation costs (excluding labor) for the heat
exchange unit, evaporation unit, and chemical precipitation have been
generated for cost comparison purposes. Conventional evaporation costs
were computed on a daily loading rate of 5,000 gallons (18,700 liters) of
serum. Capital investment was amortized over ten years (at 10 percent) and
daily costs established on the basis of 250 work days per year.
Table 5 shows operational costs and benefits associated with the protein
recovery system. Total operating expenses (excluding labor and carbon
adsorption) are about $41 per day. Recoverd proteins have a value of
$34 per day. When cost savings associated with present evaporation systems
($140/day) are added to the value of the protein by-product, a net savings
in treatment costs results with the protein recovery system. These savings
could be used to polish the chemically treated serum supernatant in an
activated carbon unit. This would reduce the organic load to the sewer
system.
Simplicity of design affords the use of this system in all industries that
require treatment of blood wastes. Energy needed to evaporate blood water
is the greatest cost associated with present blood recovery operations. By
precipitating dissolved proteins and reducing blood waste volume by 90
percent, costs of evaporation are greatly reduced. That savings, plus the
marketing of a protein by-product, can most probably cover the costs of the
recovery system for both a small and large slaughtering operation.
231
-------
TABLE 5. PROTEIN RECOVERY OPERATION COSTS AND SAVINGS
(5,000 gpd Raw Serum)
Chemical Supplies
Chitosin (0.8 Ibs/day @ $5/lb) = $ 4.00
Acid (6.5-f cone. H2S04 @ 90c#0 = 5.80
Steam
Heat Exchanger (3,200 Ibs/day @ $3.50/1000 Ibs) = 11.20
Evaporation (3,400 Ibs/day) = 11.90
Capital Recovery (10 years, 10 percent)
$16,670 x 0.06275 = $l,050/yr
$1,050/250 work days = 4.20
Neutralization and Treatment of Supernatant ($1/1000 gals) 4.50
Total Daily Cost = $ 41.60
Salable Protein Feedstuffs
340 Ibs/day @ 10
-------
4. Both organic nitrogen and volatile solids tests suggest protein
removal to be approximately 60 percent efficient. Both analytical
procedures yield similar efficiencies for given variable levels.
5. Protein removal efficiency as measured by organic nitrogen can be
predicted with a six parameter model. Removal efficiency is more
dependent upon pH and polymer dose than on temperature.
6. Denatured protein aggregates are sufficiently large to settle out of
suspension be gravity. The volume of settled solids is about 10
percent of the volume of the original waste stream.
7. Settled protein solids approximate 7 percent solids by weight. Over
90 percent volatile, the solids can be withdrawn, dried, and
marketed as a protein-rich feed supplement.
8. The supernatant from chemical treatment of the serum waste stream can
be polished further by using activated carbon to adsorb dissolved
organics. COD reductions of 90 percent can be accomplished when the
serum waste stream is treated to recovery dissolved proteins and then
treated with activated carbon.
9. Dissolved salts in the serum waste stream are not significantly
reduced by either chemical protein coagulation or activated carbon.
10. The high dissolved salt concentrations may tend to limit protein
recovery because of protein solubility in saline solutions.
11. Because daily operating costs are low, protein recovery from blood
wastes should be an attractive alternative to present blood recovery
techniques. The estimated net worth of the dried protein product
plus cost savings from present treatment systems should be more than
sufficient to amortize the required capital investment and also cover
preliminary cost estimates.
12. Chemical treatment for the recovery of serum proteins appears even
more attractive in light of the fact that rising energy costs are
making evaporation operations prohibitively expensive.
REFERENCES
1. Jones, H.R. Pollution Control in Meat, Poultry, and Seafood Processing.
Noyes Data Corporation, Park Ridge, New Jersey (1974).
2. Steffen, A.J. "Waste Disposal in the Meat Industry: A Comprehensive
Review". Proc. of the Meat Industry Research Conference, American Meat
Institute Foundation, Chicago, Illinois, pp 115-144 (March, 1969).
3. Stiemke, R.E. "Disposal of Wastes from Small Abattoirs". Proc. of the
4th Industrial Waste Conference, Purdue University, West Lafeyette,
Indiana, pp 178-202 (1948).
233
-------
4. Anonymous. "Blood System Solves Processing Problem". Reprint from
Meat Processing, April (1971) .
5. Bough, W., Landes, D., Miller, J., Young, C., and McWhorter, T.
"Utilization of Chitosin for Recovery of Coagulated By-Products from
Food Processing Wastes and Treatment Systems". Reprint from the
Department of Food Sciences, Univ. of Georgia College of Agriculture
Experiment Station, Experiment, Georgia (1975).
6. Watson, D. "Factors for Calculating Serum Albumin and Total Protein
from the Nitrogen Conten". Clinica Chimica ACTA, 16:322-333 (1967).
7. Bough, W. Personal Communication (1975, 1977)
8. Ullmann, J.E., editor. Waste Disposal Problems in Selected Industries.
Hofstra University Yearbook of Business, Series 6, Vol. 1. Hofstra
University (1969).
234
-------
IMPROVED BIOLOGICAL TREATMENT OF FOOD
PROCESSING WASTES WITH TWO-STAGE ABF PROCESS
by
B. W. Hemphill* and R. G. Dunnahoe*
INTRODUCTION
Biological treatment of food processing and combined food processing/municipal
wastes has historically presented unique problems because of: 1) highly
erratic flows and wastewater strengths; 2) high soluble organic concentrations;
and 3) tendencies to contribute to operational problems such as bulking of
activated sludge due to the presence of high levels of carbohydrate materials
(1). Traditional fixed-film treatment systems such as trickling filters have
often been used on these wastes because of their relative stability under
changing loads and low operating costs (2)(3) . However, they are limited
because of high land area requirements, high capital costs, and inferior
effluent qualities. Suspended growth systems (activated sludge) have been
used because of their lower first cost, higher effluent quality, and lower
land requirements. The success of these systems has been hindered by their
operational complexity, their sensitivity to shock loadings and tendency to
be unstable under conditions such as those typically imposed by food pro-
cessing plants.
The activated bio-filtration (ABF) process has been developed to make use of
the advantages of both the fixed-film and suspended growth treatment systems
while minimizing the disadvantages. A superior process stability is realized
over a wide range of wastewater characteristics and treatment problems. This
paper will summarize the development of the ABF process in treatment of food-
processing waste and will present operating data from both pilot scale
studies and full-scale plants.
ABF PROCESS DESCRIPTION
Figure 1 shows the flow schematic for the ABF process. After primary
treatment, the wastewater is combined with return sludge from the secondary
clarifier and Bio-Cell recycle to form a mixed liquor which is then pumped
to the Bio-Cell. The Bio-Cell reactor contains horizontal bio-media, to
which fixed-film organisms attach. The suspended growth organisms combine
with the fixed-film to oxidize the biodegradable organics. The Bio-Cell
underflow is split, with a portion being returned as Bio-Cell recycle and the
remainder proceeding to a short-term aeration basin. The aeration basin acts
as a complete mix activated sludge basin, which further oxidizes organic
materials and provides a flocculent mixed liquor prior to final sedimentation.
Settled biological solids are recycled to the Bio-Cell lift station with a
portion going to waste. For roughing applications, the aeration basin is
often deleted from the flow scheme.
*Neptune Microfloc, Incorporated, Corvallis, Oregon
2,55
-------
ABF Process Flow Schematic
ro
CO
cr>
PROCESS
INFLUENT
BIO-CELL
LIFT STATION
Fixed -Film
Bio -Cell
tPROCESS
'EFFLUENT
Return Sludge
^ Waste
r Sludge
Figure 1.
-------
The horizontal ABF bio-media consists of individual racks made of wooden lath
fixed to supporting rails, as shown in Figure 2. The horizontal configur-
ation of the media provides turbulent conditions, which are necessary to main-
tain a high dissolved oxygen environment and particle-micro-organism contact.
Oxygen transfer is provided by the dual action of wastewater continually
moving in a film across the biota, and splashing between layers. The open
design permits free flow of air in all directions, and prevents ponding or
bridging.
Sludge Recycle
One of the keys to the operation of the ABF process lies in the application
of biological solids to the fixed-film Bio-Cell, Numerous studies have
verified the advantages of this. These include studies at Corvallis, Oregon
(4), Turlock, California (5)(6), and Tracy, California (7).
At Corvallis, direct comparisons were made between identical fixed-film Bio-
Cells, with one unit utilizing sludge recycle (ABF), run next to a unit
operated as a conventional high-rate trickling filter (HRTF). Horizontal
media were used in both systems. The wastewater consisted of domestic and
combined beet processing/domestic waste. Comparative data for the two
systems are shown in Figure 3. The soluble removal efficiency for the ABF
tower was approximately twice as great as for the HRTF tower. Similar
measurements were taken when the Bio-Cells were followed by high rate acti-
vated sludge systems. Further evidence of the increased Bio-Cell removal in
the ABF mode was found in the measurement of the aeration basin oxygen uptake
rate (OUR) for the two systems. The OUR of the ABF system was approximately
50% less than that of the parallel HRTF-activated sludge system.
Similar tests were made at Tracy, California (7), where side-by-side studies
were run of ABF-activated sludge and HRTF-activated sludge systems. Measure-
ments taken during the study indicated a significantly greater soluble COD
reduction across the Bio-Cell with the application of return sludge.
An additional process benefit that has been associated with recycling of
sludge to the fixed-film unit is that of increased process stability. This
was demonstrated in the Tracy, California (7), and Turlock, California (5)(6) ,
pilot studies. Both studies were run on combined food processing and domestic
wastes. Both included side-by-side comparisons of the ABF process with hori-
zontal media and the HRTF-activated sludge process with vertical media. At
Turlock, mechanical problems early in the study caused a period of dissolved
oxygen deficiency in both systems, resulting in development of filamentous
organisms. After the problem was corrected, the ABF system recovered readily,
reaching full treatment efficiency in 7 to 10 days. However, the HRTF-
activated sludge mixed liquor settleability did not improve as did the ABF
mixed liquor. Throughout the study, the HRTF system continued to have
problems with solids settleability and carryover of solids into the final
effluent. The HRTF-activated sludge system did not meet the discharge
requirements of 30 mg/1 BOD and TSS during the test periods with yam canning
contributions, while the ABF system averaged 16 mg/1 BOD and 11 mg/1 TSS.
Similar results for both systems were observed at Tracy.
237
-------
Figure 2. Horizontal media.
238
-------
Comparison of
Soluble Organic Removal
ABF (Without Aeration) &
HRTF Processes
0 50 100 150 200 250
BIO-CELL SOLUBLE BOD LOAD
(Ibs. BOD/1000 ftVDay)
Figure 3.
239
-------
Further details of the studies of combined food processing/domestic waste-
water at Tracy and Turlock are given later in this paper.
EXPERIENCE WITH ABF ON FOOD PROCESSING WASTES
Pilot Plant Studies
Numerous pilot plant studies have been conducted to demonstrate the effective-
ness of the ABF process on various types of food processing wastewaters.
Some of the results of these studies are shown in Table 1 and are discussed
in further detail below.
Corvallis, Oregon (1974); Beets/Corn/Domestic (4): The wastewater composi-
tion tested during the study was 46% corn, 26% beets, and 28% domestic, based
on BQD5. As shown in Table 1, the system loadings were extremely high in
this study, with the Bio-Cell load over 240 Ib/BODs/day/1000 cu.ft., and
aeration detention time less than 45 minutes. Despite the high loadings of
highly soluble carbohydrate wastes, the ABF system maintained excellent set-
tling characteristics, with the SVI below 60 for all three runs. Sludge
production ranged from 0.5 to 0.6 Ib VSS/lb 6005 removed.
Tracy, California (1972); .Domestic/Tomato/Potato (7); During the canning
portion of the study, the primary industrial contributor was from tomato
processing and the BOD of the waste was 92% soluble. ABF was compared with
HRTF-activated sludge and pure oxygen activated sludge systems. ABF with
horizontal media was selected as the most efficient system studied, and con-
sistently provided an effluent with less than 20 mg/1 BOD, sludge settling
velocities greater than 6.0 feet per hour, and SVI less that 75 ml/gm.
Turlock, California (1975); (5)(6); This 1975.study included domestic wastes
combined with poultry processing, tomato cannery, apricot, peach, and yam
processing wastes. The results in Table 1 are broken down into five phases
representing different mixes of these wastewaters. In this study, as in
Tracy, ABF was compared with pure oxygen activated sludge and HRTF-activated
sludge. ABF produced the highest quality effluent, and based on an economic
analysis, has been selected for the full-scale plant.
Corvallis, Oregon (1975); Beets/Corn/Domestic (8); During the study of a
prototype 50,000 gpd ABF package plant at Corvallis, seven weeks of data were
collected, during which a significant portion of the waste was contributed by
a local canner processing corn and beets. The wastewater characteristics
were similar to those studied in Corvallis in 1974. The aeration basin
detention time was about 2 hours. The Bio-Media depth was varied from 6 feet
to 9 feet and the Bio-Cell organic loadings ranged from less than 50 to over
400 Ib BOD5/day/1000 cu.ft. Excellent effluent quality and stable mixed
liquor characteristics were observed throughout the range of loadings tested.
Full Scale Plants
Table 2 contains a partial listing of full scale plants utilizing ABF for
treatment of combined food processing/municipal or straight food processing
wastes. Summaries of operating data from the Idaho Falls, Forest Grove,
Madera and Hood River plants are summarized in Table 3.
240
-------
TABLE I
OPERATING CONDITIONS AND RESULTS
PILOT PLANT STUDIES
Influent
Corvallis, Oregon (1974)
Beets (26%), Corn (46%),
Domestic (28%)
Tracy, California
Toma to/Potato/Domes tic
Turlock, California
Domestic/Poultry
Domestic/Poultry /Apricot
Domestic/Poultry/Tomato/
Peach
Domes tic/Poultry /Tomato
Domestic/Poultry/Yam
Corvallis, Oregon (1975)
Beets/Domestic
Corn/Beets/Doraestic
BOD
349
415
344
437
180
311
314
180
295
170
337
SBOD
278
302
271
400
72
177
193
69
121
119
241
TSS
161
194
143
163
98
154
193
211
260
158
153
Bio-Cell
Effluent
BOD
49
49
48
16
25
23
11
7
15
14
14
SBOD
31
24
25
4
5
4
3
3
3
6
5
TSS
44
53
51
23
25
23
11
10
19
16
16
SBOD/ 1,000
cu ft/day
300
281
242
192.
72
116
219
122
157
118
302
Depth
ft
14
14
14
14
19
19
19
19
19
9
6-9
Aeration Basin
MLSS MLVSS
4,110 3,210
3,500 2,860
3,110 2,520
3,650 2,290
1,430
2,530
3,120
2,790
2,580
3,220 2,160
3,860 2,970
ISR*
6.2
5.5
8.3
7.8
_
5.9
5.3
3.2
2.7
7.2
2.8
SVI
45
56
50
61
_
73
69
74
105
51
57
Time
His
0.55
0.73
0.67
6.20
2.80
6.20
3.40
3.40
4.20
2.00
1.90
System
• F/M
4.74
4.75
4.91
0.77
1.07
0.80
0.75
0.47
0.61
1.01
1.48
tto-,
//BODs
0.19
0.15
0.19
0.63
1.10
0.70
1.10
0.70
0.50
0.30
*ISR - Initial settling rate, ft/hr.
TABLE 2
PARTIAL LIST OF ABF PROCESS INSTALLATION
FOR FOOD PROCESSING WASTE
Same/Location
Madera, California
Aberdeen, Idaho
Idaho Falls, Idaho
Ore-Ida Foods
Ontario, Oregon
Crete, Nebraska
Hood River, Oregon
Forest Grove, Oregon
Twin Falls, Idaho
Tracy, California
Oscar Mayer & Company
Perry, Iowa
Consolidated Badger
Marshfield, Wisconsin
Process*
ABF (w/o Aer.)
ABF
ABF
ABF
ABF (w/o Aer.)
ABF
ABF
ABF
ABF
ABF
ABF (w/o Aer.)
Design Flow
7.0 MGD
0.6 MGD
17 MGD
3.5 MGD
1.0 MGD
3.5 MGD
5.0 MGD
8 MGD
10.2 MGD
1.2 MGD
0.15 MGD
Waste Description
Olive/Domestic
Potato/Domestic
Potato/Domestic
Potato
Corn/Domestic
Fruit/Domestic
Fruit/ Vegetable/
Domestic
Potato/Domestic
Tomato/Potato/Domestic
Meat Packing
Dairy
Industrial
Contribution
25%
85%
80%
100%
80%
85%
90%
80%
85%
100%
100%
Startup
Mar 1973
Oct 1973
Jan 1974
Apr 1973
Jan 1975
Apr 1975
May 1975
May 1976
Feb 1977
Jun 1977
*ABF: Bio-Cell followed by aeration
ABF (w/o Aer.): Bio-Cell with Sludge recycle; no aeration.
241
-------
TABLE 3
ABF INSTALLATIONS
PLANT OPERATING CONDITIONS AND RESULTS
ro
-fs>
ro
Idaho Falls, ID
Forest Grove, OR
Hood River, OR*
ABF Without Aeration
No.
Months
34
15
19
Q
MGD
8.5
2.2
1.3
Influent
BOD
255
440
470
TSS
207
120
300
Effluent
BOD
12
12
18
TSS
8
15
20
Bio-Cell Load
Ib BOD/Day/
1,000 Cu Ft
50
37
67
Madera, CA**
21
2.3
246
135 35
14
51
*Three days of data omitted when sludge disposal system was inoperable in
November, 1975; four days omitted when aeration basin off-line in September,
1976.
**Includes data during olive processing season only (9 months/year).
-------
PROCESS CHARACTERISTICS AND DESIGN CONSIDERATIONS
The factors influencing ABF system performance are numerous. As a result,
classical fixed-film and suspended growth analyses are not easily applied to
system design. Fortunately, empirical approaches have been used with success.
A convenient analytical tool which will be used herein quantifies the system
organic loading in terms of System F/M (9). This is defined as:
, _ Ib Influent BOD/day to Bio-Cell
bystem J-/M lfe MLVSg ±n Aeration Basin
This equation does not include the mass of biological solids contained in the
Bio-Cell or clarifier. Because of this, the numerical value of the System F/M
will be several times greater than those normally used in activated sludge
process design.
The following sections deal with effect of loading on process performance.
Effluent Quality
In Figure 4 the effects of System F/M on ABF effluent quality for a variety of
wastewaters are shown graphically. No marked deterioration of final effluent
quality can be seen over the wide range of loadings represented. More bio-
logically resistant organic materials may affect effluent quality.
Mixed Liquor Settleability
As discussed previously, the design of conventional activated sludge systems
are often limited by the Settleability of the mixed liquor. This is
especially true for food processing wastes in which high soluble organic
content wastes contribute to sludge bulking (1). In Figure 5, the relation-
ship of system loading in terms of System F/M is shown as it affects the
settling characteristics of the mixed liquor in the ABF process. The values
of the initial settling rate and sludge volume index remain excellent even at
System F/M's greater than 4.0. Sludge Settleability, therefore, does not
limit the allowable System F/M for the ABF process.
Aeration Basin Oxygen Requirements
Oxygen requirements for the ABF aeration basin can be identified as follows:
Ib. 0? to Aeration Basin
Aeration Basin Oxygen Demand =
Ib. BOD Removed in System
As in the System F/M definition, the influence of the Bio-Cell is not consid-
ered in this equation. The denominator is computed as the total BOD to the
Bio-Cell minus, the clarifier effluent soluble BOD. For ease of analysis,
oxygen demand data are tabulated and presented as shown in Figure 6,1 in which
System F/M is plotted versus aeration basin oxygen demand. It can be seen
-that the oxygen demand decreases as the System F/M increases. The data
indicate that aeration power requirements are minimized by selection of the
System F/M at a value above 0.8. Nitrification is observed to occur at
243
-------
Effect of System
on ABF Effluent
O)
1
^_
cO
OB s
3 5
go
"o
w
Q 30
8 3°
a
on
+•_
fl>0> 10
3 c '^^
«!
N-
•D
0)
0)
I» 3O
8-
w|>20-
^E 2°
2 • 10
2«
«^T3
«»•«
il o
o j| iPo
o oo Tf o
"S" T
0
m 0
r*JO
o
• 1
l.O 2O 30 4 O 5.0
SYSTEM F/M
lbs.BOD/lbs. MLVSS
0 g
%
a>
e
o
o
1O 2.0 3O 4O 5
SYSTEM F/M
lbs.BOD/lbs. MLVSS
%»
M O O
A X
wT> *^
c
0°
4>°
AM
OO
O
o
e
l.'o 2.O 3O 40 5
SYSTEM F/M
Ibs. BOD/ Ibs. MLVSS
Figure 4.
244
-------
Effect of System F/M on
Settleability of ABF Mixed Liquor
»
I
r*
^^A
^m
o]
A
f$o*
o o
9
W 0
°
_
o^
00
o
o
o
0
O 2O 3.0 4 O
SYSTEM F/M
Ibs.BOD/lbs.MLVSS
Figure 5.
245
5.0
-------
PO
ui
u
So-
il
35 cf
UI
1.8
1.7
1.6
1.5
1.4
Aeration Basin Oxygen Demand
vs
System F/M
• Corvallis.Ore. - NITRIFICATION
O Corvallis,0re. -CARBONACEOUS
O Bend,Ore. - CARBONACEOUS
Bend,Ore.-NITRIFICATION
Rochester, Minn. - NITRIFICATION
O CD Turlock, Ca. - Mun./Yams
1=1 Turlock, Ca. — Mun./Food Proc.
Tracy, Ca. — Mun./Food Proc.
Potato Waste
Corvallis, Ore. - Mun./ Food Proc.
1.5 20
SYSTEM
Ibs. BOD/lbs. MlVSS
Figure 6.
-------
System F/M less than 0.6, hence the increased oxygen demand and greater degree
of data scatter.
Bio-Cell Loading and Media Depth
The ABF aeration basin oxygen requirements, effluent quality, and mixed liquor
characteristics have not been found to be appreciably affected by the Bio-Cell
organic loading. The relationships observed in Figures 4, 5, and 6 hold true
for loadings ranging from 50 to 300 Ib BOD/day/1,000 cu. ft. Normal design
values range from 100 to 350 lb/day/1,000 cu. ft, with a value of 200 at
average day loading being most commonly used. Media depth can range from 7
feet to 22 feet, again with no deleterious effect on performance.
Typical ABF System Design Criteria
Table 4 shows typical design criteria for the ABF process which may be applied
to most domestic/food processing waste, treatment situations. It should be
noted that design criteria are specific to the type of wastewater and that
some difficult-to-treat wastes may call for more conservative design in order
to meet stringent effluent standards.
Power Requirements For ABF Process
The ABF process can offer significant power savings compared to conventional
processes. The reason for this is the high organic removal that occurs in
the Bio-Cell in relation to the power input required for pumping.
Table 5 shows a typical comparison between ABF and activated sludge for a
food processing waste. All assumptions are indicated and reflect typical
design values. For this example, the ABF system required about only half of
the power input required for the activated sludge process. The magnitude of
the power savings is dependent upon the particular conditions but generally
falls in the range of 25 to 55 percent.
Operational Flexibility
Quite often, consideration must be given to treatment of wastes not only
during design conditions but also for time periods during which reduced
organic loadings are expected. Because the ABF process consists of two
stages, a great deal of operational flexibility is available without the need
for construction of multiple Bio-Cells, aeration tanks, etc.
A good example of the operational flexibility which can be designed into the
plant by simple piping installations is shown in Figure 7, which shows the
alternatives available at the Forest Grove, Oregon, Wastewater Treatment
Plant. For low organic loadings, the system can be operated as either a
standard high-rate trickling filter or a conventional activated sludge system.
For moderate loadings, the conversion can be made to ABF without aeration or
high-rate trickling filter with activated sludge. At high loadings, a stan-
dard ABF with aeration system is utilized. Thus, operation costs and opera-
tional complexity can be matched to the incoming loading.
247
-------
TABLE 4.
Typical ABF Design Criteria
Effluent Criteria
BOD5
Suspended Solids
NHoN (Nitrification Design Only)
Units
mg/1
mg/1
mg/1
Typical Value
20
20
1.0
Range
10-30
10-30
0.5-2.5
Bio-Cell Parameters
Organic Load lb BOD/Day/ 200 100-350
1,000 cu.ft.
Media Depth Feet 14 7-22
Hydraulic Parameters
Sludge Recycle (1) 0.5Q 0.3-l.OQ
Bio-Cell Hydraulic Load gpm/sq.ft. 3.5 1.0-5.5
Bio-Cell Recycle - as required to maintain media wetting rates.
Aeration Parameters
System F/M (2)
Carbonaceous Design
Nitrification
Oxygen Utilization
Carbonaceous Design
Nitrification (3)
MLVSS Concentration
MLSS Concentration
Oxygen Uptake Rate
Clarifier Parameters
Overflow Rate
Solids Loading
Sludge Production
Carbonaceous Design
Nitrification
lb. BOD/day/lb.VSS
it
lb 02 /lb BOD rem.
it
mg/1
mg/1
mg02/l-hr
gpd/sq.ft.
Ibs/hr/sq.ft.
lb VS/lb BOD rem.
lb VS/lb BOD rem.
1.4
0.3
0.37
0.65
3,000
4,000
65
600
1.25
0.65
0.45
0.8-3.5
0.2-0.6
0.2-0.8
1500-4000
2000-5000
50-100
300-1200
0.5-2.0
0.5-0.75
0.3-0.55
(1) Based on average flow
(2) Based on primary effluent BOD5 (process influent) loading to
MLVSS in aeration basin
(3) Total 02 required = Carbonaceous +4.6 Ibs. 02/lb NH3~N removed
248
-------
TABLE 5.
COMPARATIVE ANALYSIS OF POWER
REQUIREMENTS FOR ABF AND ACTIVATED SLUDGE
TREATMENT CONDITIONS AND REQUIREMENTS
Flow =2.0 MGD
BOD = 500 mg/1 (8,340 Ib/day)
Effluent Required:
30 mg/1 BOD
30 mg/1 TSS
Bio-Cell Pumping Requirement,
gpm (TDK = 22 feet)
Bio-Cell Horsepower
Aeration Oxygen Supply, Ib/hr
Aeration Horsepower
Total System Horsepower
ABF*
2,980
22
125
63
85
Activated Sludge**
348
174
174
*Assumed: Bio-Cell loading = 200 Ib BOD5/1,000 cu ft/day
Media depth = 14 ft; pump efficiency = 75%
Flow sufficient to provide 1.0 gpm/sq ft wetting rate
System F/M = 1.5; 02 required =0.37 Ib/lb BOD5 removed
**Assumed: 1.0 Ib 02/lb BOD5 required
Aerator field transfer efficiency of 2.0 Ib 02/hp-hr
249
-------
PRIMARY
EFFLUENT
PRIMARY
EFFLUENT
AERATION
BASIN
T
— (
RETURN \
ACTIVATED V^
SLUDGE |
(T) HIGH RATE TRICKLING FILTER
'SECONDARY
CLARIFIEB
SECONDARY
EFFLUENT
WASTE
ACTIVATED
SLUDGE
) ACTIVATED SLUDGE
PRIMARY
EFFLUENT
40 HIGH RATE TRICKLING FILTER-ACTIVATED SLUDGE
ACTIVATED BIOFILTER - ACTIVATED SLUDGE
FIGURE 7
FOREST GROVE WASTEWATER TREATMENT PLANT
SCHEMATIC DIAGRAMS
SECONDARY TREATMENT ALTERNATIVES
250
-------
SUMMARY
The ABF process offers a viable alternative for secondary treatment of food
processing wastes. The system offers a high level of process stability, an
important criterion in treatment of such wastes. Operational flexibility is
inherent in the two-stage process.
System design criteria can be expressed in terms of Bio-Cell organic loadings
and aeration basin loadings expressed as System F/M.
Significant power savings are achievable, due to the energy-efficient fixed-
film Bio-Cell. Overall cost-effectiveness has been verified by others (10) .
251
-------
BIBLIOGRAPHY
1. Pipes, W. 0., "Bulking of Activated Sludge." Advances in Applied
Microbiology, 9 (1967), 185.
I
2. Waste Reduction in Food Canning Operations. Water Pollution Control
Research Series No. 12060. FWQA, 1970.
3. Trickling Filter Treatment of_ Fruit Processing Wastewaters. Water
Pollution Control Research Series No. 12060 EAE, U.S.E.P.A., 1971.
r
4. Owen, W. F., and Slechta, A. F., Soluble Organic Removal with the
ABF Process, presented at the Third Annual Industrial Water and
Pollution Conference and Exposition by WWEMA, Chicago, Illinois
(April, 1975).
5. 1975 Wastewater Treatment Pilot Study for the City of_ Turlock,
California, CH2M/Hill, (June, 1976).
6. Mattl-i, G. P., Turlock, California, 1975, Pilot Plant Study of the
ABF System, Final Report for the City of Turlock, California (1975).
7. Williams, C. R., et al, Results of Pilot Studies on Biological
Treatment of Combined Food Processing/Domestic Wastewater at Tracy,
California, presented at the Annual Industrial Water and Pollution
Conference and Exposition by WWEMA, Chicago, Illinois (March, 1973) .
8. Booty, W. F., and Slechta, A. F., Performance Evaluation 50,000 ggci
ABF Package Plant, Technical Publication No. KT-7272, Neptune
Microfloc, Incorporated, Corvallis, Oregon (1976).
9. Slechta, A. F., and Mattli, G. P., Activated Bio-Filter Process for
Biological Wastewater Treatment, Wastewater Treatment and Reuse Seminar,
South Lake Tahoe, California (October, 1976).
10. Benjes, H., Jr., Evaluation of Biological Wastewater Treatment
Processes, Wastewater Treatment and Reuse Seminar, South Lake Tahoe,
California, (October, 1976).
252
-------
SINGLE CELL PROTEIN FROM FOOD WASTES BY THE DEEP TANK PROCESS
by
M. L. Jackson* and C. C. Shen**
ABSTRACT
Recent papers describe the development of deep tank aeration-flotation for
fermentation processing including the activated sludge process for waste-
water treatment. Extension of the concepts, which include design factors for
scale-up from laboratory-size equipment, to the production of single cell
protein (SCP) is straightforward. The procedures are reviewed briefly and
provide a design basis for oxygen supply and solids separation for the pro-
duction of biomass.
The economics of SCP production presently are marginal when the carbon source
is provided by petroleum derivatives. In contrast, the use of a suitable
waste stream for the substrate is shown to offer a potentially profitable
operation for the production of a protein material and simultaneously solve
both liquid and solid waste disposal problems. An economic estimate of the
conversion of 1.25 million gal/day of a potato waste on a continuous basis,
and sale of SCP for animal feed indicates that the profit margin should be
favorable, with income about double the annual cost of production. The
largest unknown in the operation is that of the method and cost of drying, if
indeed a dried product is desired.
Following the suggestion of another investigator, the economics of producing
SCP by batch processing, possibly to avoid the need for sterilization of the
incoming waste, were evaluated for the deep tank. Costs for tankage were
increased two to three times over that for continuous flow, depending on
processing variations, with capital costs increased 30 to 40%. However,
operating costs remained unchanged with annual costs increased somewhat.
This confirms a report for a similar process, in one case employing a potato
waste, where potential profit was shown to be substantial.
SCP AND THE DEEP TANK PROCESS
The production of single cell protein (bacteria, yeast, fungi, termed SCP) is
currently of much interest as it is expected to ultimately fulfill a major
food need for mankind (1), either directly for human consumption or as an
animal feed. Carbon sources for cell growth have been proposed from petro-
leum derivatives such as methane, alcohol, methanol and other compounds. The
*Professor and **Ph.D. candidate, Department of Chemical Engineering,
University of Idaho, Moscow, Idaho 83843.
253
-------
cost of these materials is such that the protein produced by microbial growth
does not yet appear to be competitive although a large plant for the utiliza-
tion of methanol is under construction. The use of a waste can provide a
cheap substrate particularly, as with potato wastes, where appreciable
amounts of phosphorus and nitrogen are also present for cell growth, thus
saving the cost of added nutrients. Simultaneously treating a waste stream
and solving a discharge problem by producing a product with a monetary value
is especially attractive, and producing at a profit is an evaluation to be
made.
The Environmental Protection Agency, acting on the Federal Water Pollution
Control Act Amendments of 1972, imposed limitations for the discharge of
wastewaters. These regulations apply to industrial and municipal sources
alike and are particularly severe for situations where land availability is
limited for conventional treatment. Further, capital investment and energy
consumption are currently major considerations in new plant construction.
Federal requirements are for best practical technology currently available
(BPTCA) by July 1977, and best available technology economically achievable
(BATEA) by 1983. These requirements impose a need to develop processes which
are applicable to sma1!! land areas and are economical of capital and energy.
Past practice for the growth of organisms, aside from activated sludge and
similar waste treatment plants, have not been directed toward large tonnage
biomass production on a continuous basis. The problems are those of meeting
the large oxygen demand, controlling the temperature of the fermenting
medium, and separating, dewatering and drying the cell solids. Most fer-
mentors currently in use employ mechanical agitation as part of the aeration
and mixing process in the fermentor, usually in the form of a draft tube.
Mechanical agitation appears to contribute to the heat generation problem,
along with that from the oxidation process occurring in the fermenting
medium.
Recent development of the deep tank aeration-flotation process offers a
simple, less costly system which supplies oxygen at needed requirements,
effects rapid mixing, and when the liquid depth is sufficient, provides dis-
solved gas for the flotation separation of solids (2,3,4,5). It also appears
that some cooling can be effected by evaporation of water to the air input.
This would be particularly useful for situations which operate at higher
temperatures. The small liquid surface area of a deep tank facilitates
temperature control, especially in cold climates.
The aeration tank is as tall as is feasible for ground support and space
availability, preferably 50 feet or higher, with simple nozzle air inlets
(holes in pipes) distributed uniformly over the bottom of the tank (3). The
greater liquid depths do not add to air compression costs, but rather reduce
the air volume requirement markedly because of the greater amounts of oxygen
transferred during the longer rise times. A rough rule of thumb is that 1%
of the inlet oxygen in air will be transferred to clean water for each foot
of liquid depth as shown to be valid for depths to 70 feet in narrow columns
(6,7). The depletion of the oxygen in the gas phase with increasing height
is compensated by a corresponding increased pressure at the lower tank levels.
254
-------
Mixing of the contents of tanks at 6- and 25-foot diameters has been shown
to be rapid (2,3).
The dissolved air flotation method separates cells from the liquid and takes
advantage of the inherent supersaturation of dissolved gases. Auxiliary
equipment for a pressurized air-liquid contacting system, as required in
conventional dissolved air flotation processes (8), is expensive and unneces-
sary. Reduced capital costs, compressor horsepower and pumping energy for
extending tank height, as shown later, are reduced for the deep tank process
which also has a much smaller land space requirement.
Problems associated with the recovery of SCP cells and the actual isolation
of protein from the cells have been discussed (9). For cell separation, it
is indicated that bacterial cells are more difficult to settle, in comparison
to yeasts, and that centrifugal separation may not be suitable for a bacterial
organism. The possibility of utilizing air flotation, where large cell size
is not a necessary factor and may be less desirable, was not discussed. Dis-
solved air flotation creates a buoyancy of the cells which results in a
separating force substantially greater than that for settling. In deep tank
aeration, the gases have time to diffuse to the innermost portions of the
cell floes, and when the pressure is released, appear to effect an expansion
which increases buoyancy.
-Sf
The aeration-mixing-flotation process was demonstrated for the treatment of
0.7 million gal/day of a pulp-paper mill waste stream by the growth of sus-
pended bacteria (3). Although the settling of bacteria by conventional
gravity settlers is normally considered to be a slow process, the separation
by flotation using the supersaturated dissolved nitrogen and carbon dioxide
gases was demonstrated to be a rapid process giving an underflow stream of
very low solids content. Flotation tank performance exceeded conventional
design criteria and is attributed to gases diffusing to the floe interiors
and expanding upon pressure release. Conventional air dissolving arrange-
ments, where liquid-gas contact times are only a few minutes, do not permit
such diffusion to occur. Supersaturation is also absent. Further, the cells
were readily filtered on rotary belt equipment when admixed with settled
fiber and wood fragments from a clarifier. The activated sludge contained
some fiber from the input waste stream and this also was filtered directly
on occasion without admixture to give a firm cake. This suggests that mixing
fibrous food processing wastes with cell solids would facilitate dewatering
and drying where these are available and have merit as an animal feed
supplement.
FERMENTATION PROCESSING
The kinetics of cell growth, types of organisms which can be employed to
produce protein, conventional design of fermentors, nutrients required, and
related topics are beyond the scope of the present discussion. Such topics
are treated in cogent form by Humphrey and others (10,11,12,13). Of interest
here are the economics of producing protein materials from food processing
wastes by relatively large continuous or batch processes. The deep tank
aeration-flotation design, though employing higher compression and pumping
255
-------
pressures has been shown to be economical of both energy and capital for
waste treatment (4). The costs of further processing and recovery of the
cell solids in dry form for an animal feed supplement are of additional
interest. In some cases, the cake produced may be sufficient as an animal
feed without the need for drying and storage. It is assumed that the solids
will have a protein content of about 50%, that they are bacterial in nature
and therefore show the fastest growing rate, that the organism utilized
either is developed from contaminating organisms or is from innoculation,
and that growth conditions are controlled such that the desired organism
predominates. This avoids the somewhat costly procedure of sterilization
of the incoming waste liquid and air. It has been suggested that batch
processing and proper control of pH and other conditions can be employed to
avoid the need for aseptic culture (14).
The continuous process has not been employed extensively on a large scale and
requires that the rate of growth of cells equal the dilution resulting from
the incoming waste; otherwise washout of cells occurs and growth declines.
Alternatively, the recycle of some cell solids to maintain high solids con-
centrations and high utilization rates can be employed. However, bacterial
growths of mixed cultures have been carried out in the activated sludge
process for many years although studies of the types of organisms adapting
to particular waste and the related protein content have not been pursued
extensively. Filamentous organisms occur is some situations under some
conditions and lead to what is termed bulking or floating in settling equip-
ment. With flotation separation this should not be a problem. A high pro-
tein content has been reported for some such organisms in the aeration basin
treatment of potato wastes from a potato processing plant (personal communi-
cation) .
DESIGN OF A CONTINUOUS DEEP TANK FERMENTATION PROCESS
To provide an approximate evaluation of the growth of a cell protein on
potato waste solubles, calculations were made for a deep tank aeration-
fermentation continuous process. The basis selected was a waste stream from
a potato processing plant, which has been reported in detail, and for which
an activated sludge lagoon system with mechanical surface aeration was con-
structed and employed as a demonstration project (15). Thus, a direct cost
comparison can be made with a deep tank design. Potato wastes contain
readily available carbon as the food source along with quantities of nitrogen
and phosphorus sufficient for cell growth.
Table 1 gives the general conditions assigned for the production of single
cell protein. The carbon:nitrogen:phosphorus ratios were reported to average
100:6.2:1.1 and such that no additional nutrients were needed because 80-90%
BOD reductions were obtained.
The calculations employed a computer program which has evolved through
several stages and gives deep tank designs in considerable detail (4,16), and
for a wide range of variables. It provides tank and related costs, as a
function of tank depth, and includes operating costs. The flotation,
filtering and drying operations are constant factors and were added as a lump
256
-------
sum which was considered to be on the conservative (high) side of actual
costs.
TABLE 1. CONDITIONS FOR CONTINUOUS SCP PRODUCTION
Waste flow
BOD concentration
BOD loading
Cell solids produced
Operation
Alpha (reduced Q~
transfer rate)
Dissolved oxygen residual
BOD reduction f
Motor-compressor efficiency
/
Motor-pump efficiency
Tank material
PH
1.25 million gal/day
1350 mg/1
14,100 Ib/day
0.5 Ib/lb BOD
9 months/year
0.8
2 mg/1
80%
0.75
62%
Carbon steel, rubber lined
Neutral or acidic
Previous calculations (4, 16) indicate that tank and aeration costs do not
increase significantly after a minimum annual cost condition is attained at
about 40-50 feet. Table 2 gives the initial capital costs for tank, pumps,
compressor, and piping as determined for tank heights from 22 to 82 feet.
Certain operating costs, including the cost of electrical power for pumps
and compressor, are also shown and are seen to vary little with depth with
a minimum at 50 ft. The oxygen transfer efficiency increases with liquid
depth, and less air needs to be compressed even though the pressure
required is higher. Capital costs reach a minimum at 30-40 feet and then
increase with tank depth, because compressor and piping costs decrease but
tank and pump costs increase. However, the differences in costs are not
large being only 15% greater at 82 ft as compared to 32 ft.
Table 3 summarizes the details of capital equipment costs, gives equipment
sizes and indicates the results for one tank depth, taken as 80 feet to
indicate that deep tank aeration is not a costly procedure. Acid resistant
materials and lining for the tanks were specified.
257
-------
TABLE 2. PARTIAL CAPITAL AND OPERATING COSTS FOR VARIOUS TANK HEIGHTS
(1976 Dollars)
Tank Height
Feet
22
32
42
52
62
72
82
Pump, Compressor, Tank
and Piping Costs
$165,300
162,700
166,500
169,300
175,800
182,500
189,200
12-month Power
and Related Costs
$16,900
16,800
16,700
16,400
16,500
16,600
16,700
TABLE 3. CAPITAL EQUIPMENT AND RECOVERY FOR SCP PRODUCTION UTILIZING
A POTATO WASTE STREAM COMPARED TO LAGOON WASTE TREATMENT
Aeration volume
Liquid depth
Aerator size
Solids separator
Aeration time
Solids separation time
Tanks, pumps, compressor
piping
Flotation unit, filter,
drying
Added for contingency
escalation,
instrumentation
Total capital investment
Deep Tank Aeration-
Flotation
(1976 Dollars)
260,000 gal
80 ft
24 ft diameter
450 sq ft (24 ft
equivalent diameter,
depth of 8 ft)
4 hr
21 minutes
$189,000
$215,000
35%
$545,000
Lagoon Surface Aeration-
Clarification (14)
(1970 Dollars)
3,750,000 gal
16 ft
70 x 150 ft plus
150 x 150 ft
3850 sq ft (70 ft
dia.) plus 710 sq ft
(30 ft dia.)
3 days
$646,000
ANNUAL FIXED COSTS
$ 67,600 (at 9% and
15 years)
$ 61,000 (at 7% and
20 years)
258
-------
The total costs for equipment in Table 3 include added factors for contin-
gency (20%), escalation (12%), and instrumentation (3%). Flotation tank
eosts were from an EPA publication (8), were much higher than needed for the
deep tank system, and were therefore assumed to include eosts for the filter
and dryer. Capital costs and annual charges are seen to be similar for the
SCP production facility and the lagoon treatment system even though the
collection and drying of the cell solids is an-added cost for an SCP product.
Table 4 gives estimated operating costs on the high side, and the total costs
indicated are believed to represent a conservative estimate of the costs of
production. Again, annual operating costs are similar for the two processes.
TABLE 4. OPERATING COSTS FOR SCP PRODUCTION UTILIZING A POTATO
WASTE STREAM COMPARED TO LAGOON WASTE TREATMENT
(1976 Dollars)
Installed power
Electrical power
Cost of electrical power
Heat for drying
Operating labor
Repair labor, materials
Miscellaneous
ANNUAL OPERATING COSTS
Deep Tank Aeration-
Flotation
120 hp
0.01 $/kwh
$6500
$7100
$30,000
$1400
$10,000
$55,000
Lagoon Surface Aeration-
Clarification
500 hp
0.01 $/kwh (est.)
$23,500
$18,100
$13,500
$10,000
$65,200
Table 5 shows total annual costs for the production of 810 tons of SCP for
the nine-month period considered in both cases. The selling price of the
dried product would need to be $150/ton to meet production costs but at this
figure there would be no cost for pollution control although at 80% BOD
reduction the discharge would be higher than may be allowed. 90% BOD reduc-
tion could be achieved with little added costs as was demonstrated (3). With
a selling price of $300, indicated to be a current reasonable value, the
profit margin over production costs can be considerable, taxes excluded.
Operation for 12 months per year, rather than nine as used for comparison
with the lagoon process, would increase profit margins assuming a suitable
substrate to be available.
259
-------
TABLE 5. COMPARISON OF ANNUAL COSTS FOR SCP PRODUCTION AND
LAGOON WASTE TREATMENT
f
(1976 dollars, 9 months)
Annual capital recovery $ 67,600 $ 61,000
Annual operating costs 55,000 65,200
TOTAL ANNUAL COSTS 122,600 126,200
Disposal of Cell Solids 810 tons at 14,000
$150/ton =
production costs
NET COST 0 $140,300
PROFIT AT $300/ton $122,600/year (-$140,300)
Church, Erickson and Widmer (17) give costs for the growth of a fungus on a
corn canning waste at conditions similar to the above for the potato plant
waste: 1 million gal/day and 12,000 Ib BOD/day loading for an income of
$140,000 per year and a small profit with nutrient costs being high. The
amount of solids produced and the assigned value were not indicated. A
larger operation (17) showed a profit of 0.75 of the production costs
(income = 1.75 x total annual production costs). This value compares
favorably with the income ratio in Table 5 of about twice the cost of pro-
duction. The possibility of producing a profit by growing single cell
protein on a waste substrate appears to be favorable for some situations.
BATCH TANK OPERATION
The suggestion was made by Tomlinson (14) that non-aseptic operation on a
batch basis could be an alternative to the need for sterilization to produce
a given organism. Using the above potato waste example, a calculation of
tank costs only was made for a fixed tank height of 50 feet with the number
of tanks, and hence diameter, determined from cycle and fill times. For
multiple tanks, the cycle time is the sum of the fill time, aeration time,
and drain time (equal to the fill time) where
\
Number of tanks = (cycle time)/(fill time)
The design was such that the input to the series of tanks and the passage of
the cell-liquid stream to the flotation tank were continuous but with the
required number of tanks being filled, aerated or emptied as the sequence
requires.
260
-------
The basis of calculation was the continuous process with a residence time of
four hours, waste feed of 1.25 million gal/day and a recycle rate of 25% of
the float containing 2% solids. Thus, the flotation unit for batch process-
ing would be the same as for the continuous process, the compressor duty
would be about the same, and an additional pump would be required to maintain
the pressure drop at the flotation cell inlet during final stages of drainage.
Anticipating operation at low pH values, to suit certain types of organisms,
the tanks were selected as rubber lined, and the additional pump made acid
resistant.
Table 6 shows the number of tanks, and relative tank and total capital costs
for various fill-drain and aeration times. Some aeration would also be
accomplished during the fill time which is not included in the aeration times
indicated. Tank costs increase from twice to nearly four times over the
continuous arrangement. Total capital cost increases range from 1.2 to 1.7,
the latter requiring ten tanks for a half-hour fill time, and the minimum
being for a three-hour aeration time and 1.5 or 3.0 filling times. Certain
advantages might accrue from the use of a combination of a larger number of
tanks and a reasonably long filling time, as experience might dictate modi-
fications of procedures, and surge storage could be accommodated by a varia-
tion of conditions.
TABLE 6. RELATIVE COSTS FOR BATCH TO CONTINUOUS FERMENTATION
Aeration Filling Number Diameter Relative Relative Total
Time Time of Tanks of Tanks Tank Cost Capital Cost
Continuous Fermentation
4 hr 1 30 ft 1.0 1.0
($97,000) ($544,000)
Batch Fermentation
4 hr 0.5 hr 10 11 ft 3.9 1.7
1.0 6 15 3.0 1.5
2.0 4 21 2.5 1.4
3.0 3 30 2.7 1.4
3 hr 0.5 hr 8 11 ft 3.1 1.5
1.0 5 15 2.5 1.4
1.5 4 18 2.3 1.3
3.0 3 26 2.3 1.3
261
-------
The capital costs for batch treatment are seen to be considerably higher than
for a continuous process, but if the need for sterlization is thus avoided,
the problem of cost and availability of energy is reduced. Operating costs
for the fermentation-separation step should be about the same for the batch
and continuous processes.
EUROPEAN PRACTICE
Two recent papers (14) describe the laboratory investigation of a
variety of food wastes and the growth of several types of "yeasts and other
fungi" for protein yields. Wastes considered were malting, brewing, distil-
lation, carmelization, beet sugar processing, canning (two types) and two
different potato processing wastes. Sixteen different organisms were
investigated but not all organisms for each waste. Conditions and results
are reported briefly for operating pH, carbon to nitrogen ratio, carbon to
phosphorus ratio, cell yield, percent substrate removal, final total organic
carbon content, percent crude protein, and batch growth times. Also reported
are the initial flow rates of the waste, BOD and COD concentrations, organic
carbon content, nitrogen and phosphate levels and initial pH. For present
purposes only the potato processing wastes are considered.
Certain details not described in the papers (14) but developed through per-
sonal correspondence, are that one potato processing waste, PP-1 is that
from the production of frozen chips (U.S. terminology = french fry) and PP-2
is that for potato crisps (U.S. = potato chips). Both streams were "from
the primary production lines" and were relatively clean and free from soil.
PP-1 had passed through a crude starch removal stage.
Waste PP-1 at a flow of 420 gal/min had an initial BOD of 3150 mg/1 and COD
of 3800, organic carbon of 1510, pH 4.5, and substantial nitrogen and some
phosphate with C/N =10.3 and C/P = 250. The reduction of the substrate was
from 60 to 80%. Additions of phosphate to give ratios of C/P of 35 and 70
made little difference in substrate reductions or in crude protein contents
of 40 to 56% on cell yields of 300 to 600 mg/1.
The waste designated PP-2, at a flow of 290 gal/min had a BOD - 3250, COD =
5800, organic carbon = 2280, C/N = 162, C/P = 36 and pH = 4.8. Laboratory
results for three organisms gave substrate removals of 60 to 80%, cell
yields of 2000 to 3500 mg/1, and crude protein contents of 20 to 45%.
A personal communication concerning the wide difference in cell solids from
PP-1 and PP-2, a factor of 10 in some cases, indicated that no reason was
known. This was indicated as especially difficult to understand because
of the high utilization of BOD and the similar reductions in the total
organic carbon. A toxicity problem might be suspected if the substrate
reductions were substantially less for PP-1.
In a second paper (14) the non-aseptic batch culture was proposed and tested
on a laboratory scale of six wastes each with an organism selected from the
previous screening procedures. The procedure was to investigate whether
contamination of organisms could be minimized by close control of operating
conditions rather than by sterilization of the feed waste. Low contamination
262
-------
was demonstrated by imposing three conditions: at innoculation the desired
organism greatly out-numbered the contaminating organisms, the batch opera-
tion was completed in a relatively short time, and the pH was optimum for the
desired organism. Operating at a low pH and the method of innoculation were
important factors in minimizing contamination.
The results of an economic feasibility study were also presented for an
arrangement similar to that for aeration in an activated sludge treatment
plant with added costs for dewatering and drying which were on a conservative
basis. Operating costs were based on an oxygen transfer efficiency of
1.6 Ib 02/hp-hr at 2.2 cents/kwh. The life of the plant was taken as 10
years and interest rates on capital costs were computed at both 10 and 15%.
A survey of the value of protein solids was stated to vary from $200 to "at
least" $400 per ton (18) with $300/ton selected as a reasonable figure for
estimation purposes.
Estimates of annual costs were made for the six wastes for conventional
activated sludge treatment and for plants for the production, drying and sale
of the protein solids. Design details were not provided. In all cases,
receipts for SCP production paid for the annual costs which were in excess
over that for conventional treatment except in one case where costs were just
met. However, considering annual receipts and annual costs for SCP produc-
tion, without an allowance for needed conventional treatment, a profit was
shown in only two cases but PP-2 wastes showed an annual income more than
twice that for the annual costs of production. Based on 1975 costs in
British pounds, and using an approximate exchange of $2.00/pound (now about
1.75) the capital investment was $740,000, and annual costs $250,000 for the
SCP production plant at 15% interest. The estimated annual receipts were
$540,000.
Imperial Chemical Industries, in England, has been interested in the growth
of single cell protein on methanol, and recently recognized the merit
of deep tank aeration for fermentation and growth of SCP (19). This confirms
the results of present work as to the feasibility of the procedure, the merit
for large scale processing, and quite important, the ability to supply large
quantities of molecular oxygen from air and meet the demand of the organisms.
The simplicity of the procedure is recognized which does not require the use
of mechanical agitators.
LITERATURE. CITED
1. Scrimshaw, N. S. and D. I. C. Wang, "Protein Resources and Technology:
Status and Research Needs", National Science Foundation, No. NSF RA-T-75-
037, Dec. (1975).
2. Jackson, M. L. and C. C. Shen, "Aeration and Mixing for Deep Tank Fermen-
tation", presented at the 69th Annual meeting, American Institute of
Chemical Engineers, Chicago, Dec. 1976 (under review for publication in
the AIChE Journal).
263
-------
3. Jackson, M. L., C. C. Shen and C. Plopper, "Deep Tank-Flotation Biologi-
cal Treatment: Groundwood Paper Mill Wastewater", Water Pollution
Control Federation meeting, Seattle, October 1976 (under review for
publication).
4. Shen, C. C. and M. L. Jackson, "Economics of Deep Tank Aeration-
Flotation for Wastewater Treatment", a report to the Northwest Pulp
and Paper Assn., Feb. 1977 (to be submitted for publication).
5. Jackson, M. L., "Continuous Fermentation Process and Apparatus", patent,
assigned to the Idaho Research Foundation, May, 1976.
6. Urza, I. J. and M. L. Jackson, "Pressure Aeration in a 55-foot Bubble
Column", I&EC Proc. Des. and Develop., 14, 106 (1975).
7. Jackson, M. L., James, D. R. and B. P. Leber, "Oxygen Transfer in a
23-Meter Bubble Column," AIChE Symp. Ser., WATER-1975, No. 151, Vol. 71,
159 (1975).
8. Environmental Protection Agency, "Sludge Treatment and Disposal," (1974)
and "Suspended Solids Removal," (1975).
9. Wang, D. I. C., "Protein Recovery Problems," in "Engineering of Uncon-
ventional Protein Production," Chemical Engineering Progress Symposium
Series, No. 93, Vol. 65, p. 66 (1969).
10. Humphrey, A. E., "Current Developments in Fermentation", Chemical
Engineering, Dec. 9 (1974), p. 98.
11. Humphrey, A. E., "Starvation: Chemical Engineering Can Help Fight It",
Chemical Engineering, July 18, (1966), p. 149.
12. Aiba, A., A. E. Humphrey, N. F. Mills, "Biochemical Engineering", second
ed., Academic Press, 1973.
13. Humphrey, A. E., "Engineering of Single Cell Protein: State of the Art,"
Chem. Engr. Prog. Symposium Series, No. 93, Vol. 65 (1969), p. 60.
14. Tomlinson, E. J., "The Production of Single-Cell Protein from Strong
Organic Waste Waters from the Food and Drink Processing Industries -
1. Laboratory Cultures; 2. The Practical and Economic Feasibility of
a Non-Aseptic Batch Culture:, Water Research, 10, 367, 372 (1976).
15. French, R. T., Co., "Aerobic Secondary Treatment of Potato Processing
Wastes," EPA Program, 12060 EHV, WPRD 15-01-68, 1970.
16. Edwards, L. L., Leber, B. P., Jr., and M. L. Jackson, "An Economic
Evaluation of Deep Tank Aeration," AIChE Symp. Ser., WATER-1975, No. 151,
Vol. 71, 154 (1975).
264
-------
17. Church, B. D., E. E. Erickson, and C. M. Widmer, "Fungal Digestion of
Food Processing Wastes", Food Technology, 27, 36 (1973).
18. Smith, R. N., P. Houslay and A. Whitaker, "Recycling of Food Wastes -
Microbiology and Economics," J. Applied Chem. and Biotechnology, 24
376 (1974).
19- Anon., "ICI to Scale Up Single Cell Protein Process," Chemical Engineer-
ing News, p. 25, Oct. 11, (1976).
265
-------
EVALUATION OF INSTANT NOODLES PROCESSING
WASTEWATER CHARACTERISTICS AND TREATMENT ALTERNATIVES
by
P. Y. Yang* and Victor S. Luis, Jr.**
INTRODUCTION
An instant noodles processing plant located on the industrial estate at
Minburi, Thailand, produced instant noodles at the rate of 1,670 packages per
8-hour day of operation, with each package containing 30 x 60 gram bags of
instant noodles.
Water consumption in the factory was primarily for soup-making, process equip-
ment washing, removing spent alkali from the oil used for baking and boiler
feed. Most of the washings inside the factory were channelled into a common
drainage system which in turn discharged into a rectangular concrete collect-
ing tank. The contents of this tank was pumped out daily to a nearby canal.
Toilet wastewaters and floor washings from the administrative office were
flown to a separate sewer.
Factory wastewater, containing substantial quantities of biodegradable organic
material, was discharged without treatment to the nearby canal at locations
SP1, SP2 and SP3 shown in Figure 1. The present study was proposed to deter-
mine the quality and quantity of the wastewater and recommend possible
treatment alternatives to remove effectively oil, solids and organic matter.
It was expected the results of this study would provide basic criteria for
functional design of wastewater treatment processes.
EXPERIMENTAL METHODS
Wastewater characteristics and treatment studies were included in the present
study.
Wastewater Characteristics
The factory washing contained a. mixture of soap, grease (from chicken dressing
and oil refining), flour and noddle strands.
Daily samples of wastewaters from SP1, SP2 and SP3 in Figure 1 were collected
continuously for a period of 2 weeks, but not on Sundays (the factory was
closed on Sunday). The quantity of the wastewater was determined by measuring
the depth in the receiving tank (fixed volume) at SP1, SP2 and SP3. The
quality of the wastewater, including solids, 8005, COI>> nitrogen, phosphorus,
grease and pH, was analyzed using the procedures recommended in Standard
Methods (1).
*University of Hawaii at Manoa, Honolulu, Hawaii
**University of Philippines, College Laguna, Philippines
266
-------
•»
*
!P 1
(
Spillage
Gondensate^
drippings.
1
1
Spillage*
Oil^Drip.
pings & ^
Tannings
Itf
Brismings*
Ingredient
. Mixer
(Flour, water
other)
1
Dough
Breaker
\
Roller
1
Steaming
Cutting
1
Souping
*
M
_ Fry ing
--f Salts Vv
Cooling
I
t
Packing
,
,
Storing
Steam
-N
»_
Sashwa
From Boiler
Floor and Ma
Washwater
•i
Oil
Refining
i
i
Cold Storage
Chicken Soup
and - — .
Seasoning
Preparation
chine
i
^- t
Ukali
Udltion
ter^-J
Marketing
sr i
^ Spillage
!
i
Spillage and
^" Washwater
t
SP 3
V
1
1
1
1
1
1
1
I
1
•Figure 1. Plant Layout Showing Process Flow, Sources of Washwater and
Location of Sampling Points (SP 1, SP 2, SP 3).
267
-------
Wastewater Treatment Studies
The studies were conducted in three phases:
1. Sedimentation/Flotation Study - Experiments were carried out using a
sedimentation column 127 cm high and with square cross sectional surface
area of 161.3cm2, as shown in Figure 2. The column was equipped with
sampling points, spaced at 20.3 cm intervals with the first sampling
point located 4 cm below the overflow pipe at the top of the column.
Simultaneous samplings were made from all points immediately after
filling the column and then at 5, 10, 20, 40, 60, 90 and 120 minutes
lapsed times. Withdrawn samples were analyzed for suspended solids,
total COD and grease.
2. Coagulation Study - The jar test method was adopted for the coagulation
study. Alum was the chemical coagulant used and flocculation was carried
out with a stirring machine manufactured by Phipps and Bird Laboratory,
Richmond, Virginia. The optimal pH and alum dosage to achieve suitable
removal of turbidity were evaluated by using the Hach Laboratory
Turbidimeter, manufactured by Hach Chemical Co., Ames, Iowa. Color, COD
and grease were the other parameters whose removals were determined in
this phase of the work.
3. Biological Treatment Study - This phase of the work was divided into two
parts dealing with batch and continuous-flow biological treatment processes.
In the batch study, acclimatized organisms were developed using instant
noodle wastewater as substrate. Two batch activated sludge units were
then fed with different forms of the factory wastewater. One unit
received the raw waste, while the other unit received wastewater which
had been subjected to a process whereby 60-86 percent of the grease
content had been removed. Analyses for COD, BODj, suspended solids and
grease were carried out at different time intervals. In the continuous-
flow completely mixed biological treatment study, the extended aeration
activated sludge process shown in Figure 3 was used. Two aeration tanks
were used and fed with two forms of the factory wastewater which had been
used in the batch treatment study. Again, COD, 8005, suspended solids
and grease were analyzed in the feed, mixed liquor and effluent.
RESULTS
1. Wastewater Characteristics
Since the factory only operates 8 hours a day, 6 days a week and uses small
quantities of water for washing, the average daily flow rate was considered
suitable for treatment process design. This average daily rate was observed
to be small, with a mean of 1.24 +0.17 m-Vday.
Originally, washings from the factory were discharged at three locations,
designated SP1, SP2, SP3 in Figure 1. A pond near the
factory received the wastewater from SP2 and SP3. At that time, the waste-
water characteristics at the three points were as shown in Table 1.
Subsequently it was found that the flow rates at SP2 and SP3 were small
(< 0.022 m3/day) compared with the flow rate at SP1. Moreover, from August
-------
3 on O.D. Overflow For
0.7 cm I.D. Sampling Points
@ 20 cm c-c
Sedimentation Column
0.5 cm Plexiglass
•«— 1.5 cm I.D. pvc Tube
Details of Sedimentation Bottom
Open Top 10
Feed Tank
0.5 cm Plex
All Dimensions in Centimetres
Scale: 1 cm • 10 cm
Figure 2. Sedimentation Apparatus.
269
-------
Pump
22 Liter Feed Reserve!
Glass Jar
Feed Tank
0.5 Thick Plexiglass
Reactor
1cm Thick Plexfg
Note:
All Dimensions In Centimetres
Scale: 1cm * 6 cm
Figure 3. Extended Aeration Apparatus.
270
-------
TABLE 1. WASTEWATER CHARACTERISTICS
Characteristics Unit
BOD5 mg/fc
Filtrate
Total
COD mg/jl
Filtrate
Total
Total Phosphate mg/fc as P
Kjeldahl Nitrogen mg/£ as N
BODF/BODT
CODF/CODT
BODF/CODF
BODT/CODT
BODT:N:P
BODF:N:P
Suspended Solids (SS) mg/£
Volatile SS mg/£
Temperature °C
pH
Grease mg/£
Flow Rate m3/day
Sampling Points
SP1
303
845
1000
3013
4.88
60.0
0.36
0.40
0.25
0.28
173:12:1
61:12:1
640
590
30.0
7.0
3700
SP2
76
380
300
1184
5.31
24.5
0.20
0.25
0.25
0.32
72:5:1
15:5:1
185
150
28.0
7.0
3500
1.24 + 0.17
SP3
257
738
755
2147
5.49
20.0
0.35
0.35
0.34
0.34
134:3:1
43:3:1
1000
880
38.0
7.2
-
BOD, COD, Nitrogen, Phosphorus, Temperature and pH values are 50 percentile
probability levels.
271
-------
1975, a central drainage system in the factory channeled all factory waste-
waters to the discharge point SP1. The characteristics of the combined
wastewater at SP1 were considered representative of the future discharge and
this was used in subsequent treatability studies.
Table 1 presents 50 percentile probability values of wastewater character-
istics. The BOD^ to COD ratios of the wastewater are lower than expected for
a food industry waste (usually about 0.5), probably due to the poor degrada-
bility of the grease in the analysis of BOD5. Other characteristics of the
waste, such as temperature, pH, and nutrient (nitrogen and phosphorus) concen-
trations are found to be optimal for the growth of microorganisms which is
necessary for the functioning of biological treatment processes.
2. Wastewater Treatment Studies
2-1. Sedimentation/Flotation Process - The sedimentation/flotation
process was studied for removal of suspended solids, COD, and
grease. In Figures 4 and 5, they can be seen that total COD and
suspended solid removal are less than 35% and 58% respectively
with the time and depth of column applied in the present study.
Since the higher percentage of volatile portion (about 92.2%)
presented in suspended solid (Table 1), the removal of suspended
solid by using sedimentation process become unnecessary. How-
ever, grease could be removed by the sedimentaiton/flotation
process, as shown in Figure 6. It can be seen that if 1.25 m
depth of tank is provided, it takes only 2 hours to achieve
90% grease removal by flotation.
2-r2. Chemical Coagulation Process - The purpose of this study was to
optimize the coagulating chemical dosage and pH so that maximum
total COD and grease removal could be achieved. A summary of
the results is given in Table 2. It can be seen that removals
of 57-61 percent COD and 3.7-14 percent grease are possible using
chemical coagulation. Optimal conditions are 80 mg/& of alum
dosage and pH of 6.5. However, using these conditions, the
quality of the effluent from chemical coagulation still cannot
meet the effluent standards set by the Ministry of Industry of
Thailand.
2-3. Biological Treatment Process - Batch and continuous-flow system
studies were carried out to investigate the 8005 and COD removal
efficiencies of biological treatment processes since both physical
and chemical methods were not found to be able to remove oxygen
demand materials (8005 or COD) to the required levels to meet the
effluent standards.
The results of batch studies on the factory wastewater with and without grease
skimming are shown in Figures 7 and 8. In general, the removal efficiency of
BODs (filtrate) and COD (Filtrate) are within the ranges of 96-97 percent and
84-93 percent, respectively. Residual BOD5 contents in this study range from
7.5 to 20 mg/£, which meet the required effluent standards (< 20 mg/fc). The
effect of the presence of grease in the wastewater on the BODs or COD removal
efficiency of the batch activated sludge process is found to be negligible.
272
-------
ro
-j
CJ
0.125
0.33
0.42
(X
0)
n 0.58
0.79
0.96
Percent of SS Removed In Test
of Indicated Time and Depth
12
14 (
O
17 O 19 O 0 O 6 O
J L
)55
J I
10
20 40
Time minutes
60 90 120
Figure 4. Suspended Solids Removal vs Time and Depth,
-------
no
0.96 -
O Percent of Total COD Removed in Test
at Indicated Time and Depth
10 20
40 60
Time, minutes
90 120
Figure 5. Total COD Removal vs Time and Depth.
-------
PO
•~4
(Jl
100
Flotation Column Depth
• -125.0 cm
+ -103.7 "
o - 85.4 "
a - 64.0 "
20 30 40 50 60 70 80 90 100 110 120
Detention Time, MIN
Figure 6. Grease Flotation Efficiency as Influenced by Detention Time.
-------
TABLE 2, COAGULATION STUDY
I. Sample Characteristics;
pH = 6.66
Turbidity = 38.5 FTU
II. Determination of Optimum Dosage:
CODx = 1247 mg/&
Grease = 267 mg/£
Alum Dose, mg/£
20
40
60
80
100 120
pH 7.02 6.98 6.98 6.92 6.89 6.84
Turbidity (FTU) 37.0 36.0 33.5 23.0 16.0 9.0
Alkalinity (mg/£) 510 460 450 440 410 430
Color Dull >• Clear
CODT (mg/£) 415 415 405 324 405 283
Opt. Dose +
III. Determination of Optimum pH at Optimum Dosage of 80 mg/£ Alum:
PH
Turbidity (FTU)
pH
Color
Opt. pH
5.5
7.0
5.35
Clear
6.0
7.5
5.89
6.5
17.5
6.45
+
7.0
33
7.09
7.5 8.0
40 38.5
7.18 7.39
'* greyish
IV. Grease and COD.]-, Removal at Optimum pH and Dosage:
Optimum Dosage and pH
Turbidity (FTU)
PH
CODT (mg/£)
% CODT Removed
Grease (mg/fc)
% Grease Removed
1
27
6.79
487
61
238
10.8
2
27
6.84
555
55.5
257
3.7
3
32
6.91
536
57
242
9.3
4
25
6.9
526
57.8
230
14
Untreated
Sample
33
6.59
1247
267
1 and 2 adjusted to pH 6.5; 3 and 4 unadjusted pH.
276
-------
Legend:
O—O Total COD
»'• ' • Filtrate COD
D O Suspended Solids
Filtrate BOD5
0 24 6 8 10 12 14 16 18 20 22 24
Aeration Time, hr
Figure 7. Batch. Biological Process Without Grease
Skimming Wastewater
Legend:
O——O Total COD
• 4 Filtrate COD
p__n Suspended Solids
Filtrate BOD5
6 8 10 12 14 16 18 20 22
Aeration Time, hr
Figure 8. Batch Biological Process With Grease
Skimming Wastewater.
24
277
-------
Based on the wastewater characteristics and batch activated sludge studies,
the extended aeration activated sludge process was selected for further study
which could be possible for future practical application. By applying
averaged organic loading rate of 0.413 and 0.23 of Kg Total BODs/KgVSS-day
and Filtrate BOD5/KgVSS-day, the BODs removal efficiency was always above
90% and the BOD5 content of the effluent was consistently lower than 20 mg/£.
However, the suspended solid concentration in the effluent was generally
higher (160-300 mg/£) than the effluent standard of the Ministry of Industry
(less than 60 mg/£). The loss of suspended solids in the effluent would be
that the grease accumulated on the sludge. In the present study, it was also
observed that the range of grease removal was varied from 3% to 54%. Appa-
rently, the removal of grease becomes necessary before further biological
process is proceeded. Certainly, it will improve the effluent quality based
on the suspended solid concentration.
For further studies of biological treatment processes (including extended
aeration, low-rate trickling filter, rotary tube filter and anaerobic contact
processes) by using the same wastewater employed in the present study, Luis
(2) concluded that the order at which high COD removal was observed was
extended aeration > rotary tube filter > trickling filter > anaerobic contact
process within the range of organic loading of 0.3882 to 0.9705 Kg COD/m3-day
as shown in Figure 9. Also, it was concluded that the order of overall per-
formance for grease removal was trickling filter > rotary tube filter >
anaerobic contact > extended aeration process as shown in Figure 10. Both
rotary tube and trickling filters exhibited higher grease removals which may
be that the slime (fixed film) is able to "adsorb" the grease from wastewater
and which is finally slowly degraded. Relatively, the trickling filter shows
a higher grease removal than the rotary tube filter because the specific
surface area of the medium (corrugated PVC) used in the trickling filter is
195 m^/m3 compared to 76.4 rn^/m3 for rotary tube filter.
RECOMMENDATIONS AND SUMMARY
1. The results of the studies suggest grease should be removed from the
wastewater before other treatment processes are applied. This can be
done by providing a grease separator as first stage in the process or,
if batch or semi-continuous flow treatment is preferred, by arranging
for surface skimming of grease from a holding tank.
2. Since the quantity of wastewater is very small, averaging 1.24 m^/day,
and the plant operates and produces wastewater only 8 hours per day,
semi-continuos flow operation of a treatment plant is possible.
3. Raw wastewater sedimentation, even with chemical coagulation, is not
recommended.
4. Biological treatment is necessary because of the high BOD5, COD and
volatile solids contents of the factory wastewater,
5. Considering the low flow and high grease content of wastewater, the
extended aeration activated sludge process incorporating with prior
removal of grease is recommended as most suitable for this wastewater.
278
-------
ro
^i
UD
9<
100-
90-
80-
3 W
§51
4tf
30
20
10
Q
Aeration
* Rotary Tube Filter
Trickling Filter
-A Anaerobic
J7 Anaerobic
* Average Values of Anaerobic Contact
0.5
COD Applied, kg COD/m5-day
1.5
Figure 9. Comparison of COD Removal Efficiency of the Biological
Process with COD Applied.
-------
70
A
A
A
A
A
ro
CO
o
60 1
50
40
30 -\
20
10 H
A
B
B
O
O
A
O Activated Sludge
Q Rotary Tube Filter
A Trickling 'Filter
If Anaerobic Contact : 30-day SRT
Contact: 20-day SRT
Average Values of Anaerobic Contact
0.5
*j
Grease Applied, kg Grease/m -day
Figure 10. Comparison of Grease Removal Efficiency of the
Biological Process with Grease Applied.
1.0
-------
6. Either a semi-continuous or continuous flow activated sludge process
would be appropriate forms of biological treatment.
7. The semi-continuous flow extended aeration process with grease skimming
is outlined in Figure 11. This form of operational process would cost
less in terms of construction cost but would involve more operational
attention than a continuous flow process.
8. The continuous flow process alternative is outlined in Figure 12. This
process would operate continuously with intermittent removal of grease
from the holding tank and occasional removal of excess sludge from the
activated sludge aeration tank. The sedimentation tank should be an
integral part of the aeration tank, with the settled sludge returning
by gravity to the aeration compartment.
At the present time, the factory has been treating its wastewater with the
semi-continuous flow operation process and giving a satisfactory result. The
present treatment facility was constructed by two times the figure recommended
because the factory processing plant was expanded two times the original
plant.
ACKNOWLEDGMENT
This work was supported in part by Wan Thai Foods Corporation, Thailand,
through the Asian Institute of Technology, Bangkok, Thailand.
REFERENCES
1. STANDARD METHODS FOR THE EXAMINATION OF WASTES AND WASTEWATER. 13th Ed.,
Amer. Pub. Health Assn., New York (1971).
2. Luis, Victor S. Biological treatment of noodle wastewater. Masters
Thesis, Asian Institute of Technology, Bangkok, Thailand, No. 981 (1976).
281
-------
Combined Westwater j Flow Rate - 1.24+0.17 m3/day
Grease
Skimming
Holding Tank., (Existing Tank)
Volume is 3.0 m3 with depth of 1.3 m
BOD5 (Total) - 845 mg/1
BOD, (Filtrate) - 303 mg/1
Suspended Solids - 640 mg/1
Volatile Suspended Solids = 590 mg/1
Excess Sludge
Wastage when the
MLSS Concentration*
Reaches 8000 mg/1
Pumping Capacity to Pump
Daily Flow Over 2 hours
Activated Sludge Aeration and
Settling Tank with Volume of 3 m3
Aeration « 20-22 hr
Sedimentation » 2-4 hr
Supernatant
. Final Effluent
BOD5 < 20 mg/1
Suspended Solids < 60 mg/1
Grease < 50 mg/1
Figure 11. Recommendation of Semi-Continuous
Flow Treatment Process.
282
-------
Combined Westwater
Flow Rate - 1.24+0.17 m /day
Holding Tank., (Existing Tank)
Volume s 3.0 m with depth of 1.3 m
BOD5 (Total) - 845 mg/1
BOD5 (Filtrate) - 303 mg/1
Suspended Solids - 640 mg/1
Volatile Suspended Solids - 590 mg/1
Grease
Skimming
Pumping Capacity = 1.5 m/day • 0.062 m /hr
Extended Aeration Activated Sludge
Process (Combined Aeration and Settling
Tank with 100* Sludge Recycle)
Supernatant
Detention Time:
Aeration - 20 hr
Settling » 6 hr
Volume of Tank:
Aeration =1.24 m3+30%
Settling - 0.37 m3
•Occasional Sludge Wasting
Final Effluent
BOD5 < 20 mg/1
Suspended Solids < 60 mg/1
Grease < 50 mg/1
Figure 12. Recommendation of Continuous
Flow Treatment Process.
283
-------
POTATO JUICE PROCESSING*
by
**
J. R. Rosenau, L. F. Whitney, and J. R. Haight
ABSTRACT
Various options are available for processing of potato juice into animal
feeds. These include reverse osmosis, ultrafiltration, heat and acid coag-
ulation of protein, vacuum evaporation, drum or spray drying, and acid solu-
bilization and alkaline precipitation of solanine. The most promising flow-
chart is discussed in light of economics, solanine toxicity, lysine availa-
bility, trypsin inhibiter inactivation, and product characteristics.
INTRODUCTION
Traditionally, potato starch processes have involved grinding cull potatoes
followed by sieving and refining operations which used large amounts of
water (1) (2) (3) (4) (5). The water was used to separate the starch from
the pulp and to refine, i.e., wash, this starch to a high degree of purity.
At the Seventh National Symposium on Food Processing Wastes, we presented a
process (6) that reduced water usage, which has been as high as 1000 pounds
per cwt. of input potatoes, to 25 pounds of water per cwt. We have since
tuned and modified that process (see Figure 1) to eliminate an air classi-
fication step. The starch processing system (discussed in the following
section since it affects options available to juice processing) produces a
juice stream of about 5% solids. This paper summarizes factors influencing
possible processing schemes for the juice.
STARCH PRODUCTION PROCESS
Figure 1 depicts the current modified starch production process. Potatoes
are ground by hammer milling and the slurry wetsieved with recycled juice.
The pulp remains above the 140 mesh screen while juice and starch pass
through. The slurry is pumped to a liquid cyclone wherein the starch is
spun to the underflow while the overflow is largely recycled to the screen.
Excess juice at 5% total solids (w.b.) passes to the juice processing
*This investigation has been supported by the U. S. Environmental Protec-
tion Agency (Grant No. R-803712-02-0), Agway, Inc., the Main Potato Com-
mission, and the University of Massachusetts.
**Department of Food and Agricultural Engineering, University of Massachu-
setts, Amherst, MA 01003.
284
-------
CULL POTATOES
SO,
Grinder
Cyclone
JUICE
Sieve
pulp
Press
i
CONC. DEPROT. JUICED
Tank
I
Defoamer
Pump
**-
Pump
*
Cyclone
Drier
PULP
WATER
I
Centri f.
Pump
i
cake
Drier
STARCH
Fig I. Starch and Pulp Processing Flowchart.
285
-------
system. A defearning device such as the Cornell Machine Company's "Vesator"
can be used as shown to control foaming - a problem of major importance if
frozen or especially off-grade tubers are to be processed. The underflow
from the liquid cyclone is pumped to a second liquid cyclone. The overflow
from the second liquid cyclone is recycled to the tangential elutriation
inlet of the first liquid cyclone (the design is a patented feature of the
Bird Machine Company**). The underflow from the second cyclone passes to
the basket centrifuge for final purification. Wash water in the amount of
only 25 pounds for every 100 pounds of potatoes processed is introduced at
this point. All of the liquid spun from the starch in the centrifuge is
recycled to the liquid cyclone system.
The second liquid cyclone as shown, overcomes the problem of incomplete
starch removal from the juice by a single liquid cyclone if the flow rates to
the device are not very carefully tuned.
Centrifuged starch (60-65% T.S. (w.b.)) is dried - preferably in a flash
drier to obtain maximum whiteness. Reflectance values in the range of 93.5-
96% are obtained, however, even with simple tray drying at 60C. The crude
protein content is about 0.06%. Air classification as previously used is
unnecessary in the revised process.
ULTRAFILTRATION
The first obvious system for processing juice would include ultrafiltration
to concentrate a protein stream in a similar manner to the now widely used
system for cheese whey. Unfortunately, however, potato proteins, like soy-
bean proteins, contain trypsin inhibitors which must be inactivated with
heat. With little more heat than required for inactivation, the proteins
are coagulated such that they can be collected and dewatered to 20% solids
by centrifugation. It thus seems pointless to include a superfluous ultra-
filtration step.
WHOLE JUICE CONCENTRATION
Juice can be concentrated by evaporation (with or without a reverse osmosis
preconcentration step) and spray or drum dried to produce 50% crude protein
meal. The Eastern Regional Research Laboratory of the USDA (ERRL) has
recently investigated this option at some length (7). The dried juice is
very hygroscopic unless lime is added before the drying step. Sufficient
heating again must be done to inactivate the trypsin inhibitor. Chick feed-
ing trials performed by R. Gerry at the University of Maine have shown that
inclusion of this material at levels above 6% of the diet slow growth rates.
*Cornell Machine Company, 45 Brown Avenue, Springfield, NJ 07081.
**Bird Machine Company, Inc., South Walpole, MA 02071.
286
-------
E. 0. Strolle (ERRL), in attempting to pinpoint the cause of this poor per-
formance, has noted up to 80% losses in available lysine, apparently due to
simple Maillaird browning, at evaporation temperatures as low as 60C (140F)
(7). Armed with only the above facts, one would conclude that potato juice
is only good for ruminant feeding. However, E. 0. Strolle, in earlier
trials, has conducted rat feeding trials with potato protein obtained by
very rapid heat and acid precipitation followed by filtration and drum dry-
ing and has obtained PER's equal to casein. This suggests that the protein
can be made to coagulate fast enough into a form wherein browning is inhib-
ited. Braverman's text on food chemistry (8) supports the idea that reac-
tion rates are very slow with materials not in solution. Moreover, acidic
conditions also tend to slow browning reactions. This suggests a third
general approach to juice processing.
HEAT PRECIPITATION
By rapid (i.e., by steam injection or infusion) heat and acid precipitation,
potato juice proteins can be coagulated, centrifuged to roughly 20% solids,
and dried (9) (10) (11). While the approach only collects about 35% of the
crude protein of the juice, it yields a protein that is nonhygroscopic,
concentrated (i.e., about 70% protein (d.b.)), and without antitrypsin ac-
tivity. The 35% figure includes nearly all the TCA precipitable protein.
The rat feeding tests mentioned above suggest that the protein is well
balanced. The deproteinated juice can be concentrated (with or without
reverse osmosis preconcentration) to 70% solids without serious foaming pro-
blems. Since there is no way to prevent lysine destruction with this mate-
rial, the obvious way to dispose of it is to add it back to the pulp (which
serves as a drying aid) and dry it for ruminant feeding. (Ruminants do not
require a balanced proteinL) Where feed lots are in close proximity to the
factory, this drying step can be eliminated. A final question with respect
to the heat coagulation process (see Figure 2) is the distribution of glyco-
alkaloids naturally occurring in cull potatoes.
SOLANINE
Solanine is the principal component of the toxic glycoalkaloids found in
potatoes. (The word "solanine" is also used to denote the whole family of
these materials (12) (13).) Dutch researches (14) have suggested that up to
half of the solanine is associated with heat coagulated protein. Our recent
results with juice heated at pH 4 do not confirm this. Rather, the solanine
seems to act as if it were soluble and can be found in the various output
streams in rough proportion to the water content of these streams at the
point of separation. In particular, the solanine content of heat and acid
precipitated protein is about the same, on a dry-weight basis, as the par-
ent potatoes. The deproteinated juice is rich in solanine but by blending
this fraction with the pulp, the solanine is diluted. The resulting pulp
mixture contains about twice the solanine content (again on a dry-weight
basis) of the parent potatoes. T. J. Fitzpatrick and S. F. Osman (ERRL),
who have worked extensively with solanine, feel that, at this level, solanine
should not pose a problem with feeding of the pulp to ruminants which are
the only animals that can utilize it. This is a reversal from the viewpoint
287
-------
JUICE
.ACID
Heater
Nozzle Centrif.
sludge ^
Spray Drier
T
PROTEIN MEAL
Rev. Osmosis
WATER
Evaporator
I
WATER
CONG. OEPROT. JUICE
(to Pulp Stream)
Fig. 2. Juice Processing Flowchart.
288
-------
that solanine would have to be solubilized by acid extraction and then pre-
cipitated and removed by raising the pH followed by centrifugation with a
disk type automatic desludging centrifuge.
FOAMING
Potato juice has a great tendency to foam. Juice from partially frozen pota-
toes is especially troublesome. Defoaming chemicals, if used in sufficient
quality, can control the problem but are expensive.
The Cornell Machine Company "Versator" has been used under plant conditions
and has proven effective in breaking foam. The machine is based on the prin-
ciple of sheeting the foam onto a spinning curved disk within a vacuum chamber.
The defoamed liquid, of course, maintains its foaming characteristics and, if
again agitated with air, will refoam. The "Versator" is of no use in handling
foaming problems during evaporative concentration.
An ultrasonic probe was tested to see if high energy sound could be used to
break up the foam but the results were totally negative.
Heating to the point of coagulation greatly reduces the foam stabilization
properties of the protein. Juice treated in this manner can be evaporated
without serious foaming problems.
YEAST FERMENTATION
The last remaining starch plant in the Northeast has installed, in a coopera-
tive venture with Bio-kinetics, Inc.*, a yeast fermentation system. They are
experiencing some start-up problems with foaming since they are aerating, in
open tanks, whole juice without heat treatment.
An additional problem with fermentation systems is that starch plants use
high levels of S02 to overcome enzymatic browning. We feel that these levels
pose problems for any single cell process unless it is precipitated or
stripped.
PRELIMINARY ECONOMICS
The present economic picture remains satisfactory although the current de-
pressed price of cornstarch is not helpful. One hundred pounds of incoming
potatoes should produce 12.2 pounds of 18% moisture starch which, at 11
-------
It is difficult to place a value on cull potatoes. A recent Canadian pub-
lication mentions $0.27 per cwt. In some seasons, this value might increase
to $2.00 per cwt. Current Maine prices are approximately $0.75. Energy
costs will be $0.16 per cwt processed assuming $3,00/10& Btu. This leaves
a gross margin of $0.93 per cwt processed if we assume miscellaneous supplies
will be $0.05 per cwt.
If we assume an average of 400 ton of potatoes are processed each day of a
200-day campaign, that we have a labor force of 10 persons with an average
cost of $20,000/year each, that the plant costs about $2,000,000 and that 20%
of this must be used per year to cover maintenance, taxes, depreciation, etc.,
we find a 42% R.O.I. Obviously a more detailed cost analysis must (and will
be) be performed to confirm the economic feasibility of the processes pre-
sented.
SUMMARY
In light of the apparent retention of lysine availability in heat coagulated
protein, its loss in soluble protein processing, the apparent solubility of
solanine, and the reduction in foaming ability by heat coagulation, a logical
juice processing scheme becomes obvious. Juice is heated quickly and centri-
fuges. The resulting protein sludge (roughly 20% solids, 70% crude protein
(d.b.)), which accounts for about 35% of the crude juice protein, is spray
dried as a high protein meal. The remaining juice is concentrated to 15%
solids by reverse osmosis followed by evaporation to 70% solids. It is then
mixed with the pulp for drying which acts as a carrier for the otherwise dif-
ficult to dry hygroscopic material.
REFERENCES
1. Hemfort, H., Huster, H. and Heimeier, 1975. Low water consumption in
preparing potato starch. Reviewed by Peterson, N.B. Edible Starches
and Starch-Derived Syrups. Noyes Data Corp., Park Ridge, NJ,
2. Hicks, C.P., 1970. Starch refining 2--quality, yields and equipment.
Process Biochemistry 5 (7) : 30,
3. Howerton, W.W. and Treadway, R.H., 1948. Manufacture of white potato
starch. Industrial and Engr. Chem. 40(8): 1402.
4. Treadway, R.H., 1959. Potato starch. Potato Processing, ed. by
Talburt, W.F. and Smith, 0., AVI Publishing Co., Westport, CT.
5. Treadway, R.H., 1967. Manufacture of Potato Starch, ed. by Whisler,
R. L. and Paschall, E.F., Starch: Chemistry and Technology Vol. II,
Academic Press, NY.
6. Rosenau, J.R., Whitney, L.F. and Elizondo, 1976. Low wastewater pota-
to starch/protein production process - concept, status, and outlook.
Proceedings of the Seventh National Symposium on Food Processing Wastes.
290
-------
7. Strolle, E.O., 1977. Verbal communication. Eastern Regional Research
Laboratory, 600 East Mermaid Lane, Philadelphia, PA 19108.
8. Braverman, J.B.S., 1963. Introduction to the biochemistry of foods.
Elsevie Publishing Company.
9. Anon., 1968. Protein. Protein recovery from potato starch, Process
Biochemistry. May, 1968, p. 51.
10. Stabile, R.L., Turkot, V.A., and Aceto, N.C., 1971. Economic analysis
of alternative methods for processing potato starch plant effluents,
Proceedings of the Second National Symposium of Food Processing Wastes,
Denver, CO.
11. Strolle, E.O., Cording, J., Jr., and Aceto, N.C., 1973. Recovering
potato proteins by steam injection heating. J. Agr. Chem. 21 (6): 974.
12. Jadhav, S.J., and Salunkhe, D. K., 1975. Formation and control of
chlorophyll and glycoalkaloids in tubers of Solanum tuberosum L. and
evaluation of glycoalkaloid toxicity. Advances in Food Research, Vol.
21. ed. by Chichester, C.O. Academic Press, NY.
13. Alvarado, R., and Fagerson, I.S., 1976. Personal Communication. Dept.
of Food Science and Nutrition, University of Massachusetts, Amherst,
MA 01003.
14. deNoord, K.G. 1976. Personal communication. Avebe, P. 0. Box 15,
Veendam, Holland.
291
-------
RECOVERY AND APPLICATION OF ORGANIC WASTES
FROM THE LOUISIANA SHRIMP CANNING INDUSTRY
by
B. E. Perkins and S. P. Meyers*
INTRODUCTION
Since 1968, investigations from our laboratory have documented the value of
"waste" proteinaceous substrates from Louisiana shrimp and crawfish process-
ing industries(1) (2) (3). Overall efforts have been directed toward material
recovery and development of a diversity of applications of proteinaceous by-
products. With regulatory requirements necessitating screening of solids from
effluent discharges, shrimp canning operations are experiencing an abundance
of solid waste materials. Efforts to upgrade the quality of meals from shrimp
and other crustacean wastes are in progress, especially in terms of the effect
of processing conditions on nutritional and pigment value. This is particu-
larly important in view of the demonstrated variability in analyses and poten-
tial nutritional composition of shrimp meals(4). Further attention is being
given to reduction of BOD loads of discharge streams and economically sound
use of recovered proteinaceous material.
The magnitude of shrimp waste in the Gulf Coast area for recovery and applica-
tion is by no means insignificant. The total heads-on landings for the Gulf
Coast canned and frozen shrimp industry, including both brown and white shrimp,
amounted to 179.5 million Ibs in 1976. Based upon a loss of 75.9% of the
heads-on weight, the total potential waste, including materials such as heads,
shells, legs, etc., from both the Gulf canned and frozen shrimp industries is
136.3 million Ibs (Fig. 1).
The calculated waste generated in the various effluent streams from the pro-
cessing operations is noteworthy (Fig. 2). Based on an approximate 3.3 gal
water/lb shrimp processed, as much as 600 million gal are used in shrimp pro-
cessing in the Gulf. Included in this volume is a total potential dissolved
and suspended microscopic waste load of 4.7 million Ibs. Approximately 3.3
millions Ibs dry wt (16.5 million Ibs wet wt) of this waste is from the frozen,
peeled tail operations, while the canning portion (raw and blanch wastes) con-
tributed 1.4 million Ibs dry wt (7.0 million Ibs wet wt). In the main, recov-
ery and use of this dissolved shrimp protein has been neglected with prime at-
tention being given to macroscopic solids recovery as a meal substrate.
*Department of Food Science, Louisiana State University, Baton Rouge, Louisi-
ana 70803.
292
-------
GULF SHRIMP
LANDINGS
1976
(HEADS-ON)
150
CD
J
U_
o
V)
z
2 100
i-
X
o
yj
50
C
A
N
N
E
D
POTENTIAL
WASTE
YIELDS
PROCESSED AS
FROZEN CANNED
FROZEN CANNED
Figure 1. Gulf shrimp landings, yields and potential waste, 1976.
293
-------
500
o: 400
UJ
t-
300
o
200
IOO
BLANCH
WASTES
RAW
WASTES
PROCESSED AS
FROZEN CANNED
1.4 X 10 LBS.
(DRY WT.)
Figure 2. Amount of water used in Gulf Coast shrimp processing, 1976.
294
-------
The general waste flow from shrimp canning involves largely three separate
components: a) material from the peeling, separating and deveining opera-
tions; b) that from the blanch tanks; and c) the shrimp meat and debris dur-
ing the several inspection phases. Primary attention has been given to the
blanch portion of the process in terms of recoverable proteinaceous material
from the hot brine liquor. Aspects considered in this study have included in-
plant characteristics, notably ratio of blanch load to total discharge and to
volume and meat/liquid ratio, load fluctuations and seasonal peak features;
variations between plants, i.e., process flow, blanch time, salinity and tem-
perature, and shrimp load and characteristics; analyses of potential biochemi-
cal/nutritional application of product in terms of facility logistics. Changes
in dissolved organic and suspended meat levels as related to averages or
trends in shrimp size also are important and must be tabulated in overall
analyses of recovery variability.
Many facets of the shrimp canning operation have been presented in earlier re-
ports (5) (6) directed toward development of an economical, practicable method
of effectively and efficiently treating waste waters from such plants.
Studies involved characterization of the BOD-5 and suspended solids, includ-
ing that of the water flume dumps, canning, and retort cooling and inspection.
It was calculated that the peeling operation contributed approximately 70% of
the BOD-5 (4.89 lbs/100 Ibs shrimp) and suspended solids (2.63 lbs/100 Ibs
shrimp) to the total discharge, compared with <5% for the blanching phase.
Data generated in the previous and current studies will be used for evaluation
and application in design, construction and operation of wastewater treatment
systems for the shrimp canning industry, particularly on the Gulf Coast.
MATERIALS AND METHODS
Sampling Techniques
Samples of effluent streams were taken from three major shrimp canneries in
Westwego and Harvey, in the greater New Orleans vicinity. More than 20 such
facilities of various sizes are located in the Louisiana-Mississippi Gulf re-
gion, the majority of which are found in the New Orleans and Houma, Louisiana
area. Logistics of areal plant concentrations are important in any projection
of economics of by-product recovery and ultimate use.
Blanch water was taken directly from the blancher overflow valves. While a
batch process was used in early stages of the industry, most plants now employ
a continuous system in which shrimp are passed through the tank via a screw
conveyor and brine water is continually added, with surplus washed from the
tank. In blanching or pre-cooking operations(7) shrimp are processed in a
boiling brine which extracts moisture and solubles, curls the meat, and devel-
ops the characteristic pink to red color of the final product. During this
treatment, both particulate and dissolved shrimp protein is concentrated in
the liquor which is discarded usually at the end of the daily processing.
The three facilities studied all use continuous blanchers in which the greater
portion of the water is recirculated within the tank and a lesser amount dis-
carded through an overflow valve. As noted, overflow loss is compensated for
by addition of fresh brine, procedures for which vary from plant to plant.
Since processing equipment used in the three canneries sampled is more or less
295
-------
similar in operation, differences in by-product loading rates as a result of
dissimilar equipment may be discounted.
Raw wastewater was collected from effluent streams carrying wastes from the
washing, peeling, and deveining operations. All liquid material was collected
and stored for transport in 5-gal plastic carboys. Samples of solid waste
composed of detritus remaining after initial picking, bits of shell not re-
moved during peeling, and broken pieces of cooked shrimp were also collected.
Samples were collected from the discharge chute of the after-blanch air
cleaner. All solid samples were packed in plastic bags and stored in ice
chests. Subsequently, all material was transported to the LSU Department of
Food Science at Baton Rouge and frozen at -5 C until analyzed.
u
Shrimp Blanch Water Analysis
Blanch water samples were filtered through cheesecloth to remove suspended
meat fragments, dried in a microwave oven (Amana Touchmatic Radarange Model
#RR-6-W), and the resulting dry weights recorded. Unless otherwise indicated,
all dry weight determinations were made in this manner.
The protein liquor obtained after removal of suspended meat was adjusted to
pH 4.2 via 12 M reagent-grade (37-38%) HC1, thus allowing for isoelectric pre-
cipitation of dissolved proteins. After precipitation was complete (60-75
min), the supernatant layer (sugar liquor) was removed and the precipitate
centrifuged at 10,000 rpm for 10 min. The supernatant was added to the
previously-siphoned sugar liquor. Dry weight of dissolved protein removed was
determined.
The sugar liquor obtained after removal of dissolved protein was adjusted to
pH 9.5 with technical-grade Ca(OH)a. Adjustment to pH 9.5 in the presence of
Ca"^" ions permitted isoelectric precipitation and salting out of the dissolved
sugar. Following this, the supernatant layer (final effluent) was siphoned
and the precipitate centrifuged at 10,000 rpm for 10 min. The supernatant was
added to the previously-siphoned final effluent. Dissolved sugar removed was
dried and weighed.
To determine the amounts of dry solids in the unused blanch water (brine
water) and the final effluent, samples of each were vacuum-dried at 85 C/30 in.
Hg pressure.
Raw Shrimp Processing Water Analysis
Raw shrimp processing water was filtered through cheesecloth to remove sus-
pended meat fragments and shells which were dried and weighed. The remaining
protein liquor obtained was adjusted to pH 4.2 with 12 M HC1, permitting pre-
cipitation of dissolved proteins. The supernatant layer was siphoned and dis-
carded, and the precipitate centrifuged at 10,000 rpm for 10 min and dried and
weighed.
296
-------
Determination of Total Carbon
in Shrimp Blanch Water
Samples of each of the previously-dried components, i.e., dry material from
fresh brine solution untreated blanch water, protein liquor, sugar liquor, and
final effluent, were ground in a mortar and pestle and weighed in 0.1 g
amounts. All of these were taken uniformly at 1 hr after start-up of process.
Each sample was analyzed for total carbon with a Hoskins electric furnace
(1000 C). Total carbon was converted to COa gas, which was captured in an
Ascarite-containing vessel. Changes in weight.in the Ascarite-containing ves-
sel indicated the weight of the COa absorbed. From these values, total car-
bon contents of each of the components was calculated and recorded.
RESULTS AND DISCUSSION
Isoelectric Point Precipitation
Major attention has been given to isoelectric point precipitation with HC1 and
preliminary centrifugation, as well as use of specific coagulants such as chi-
tosan (Hercules Co., Kytex H®). In initial tests (Table 1), concentrated
HzSOij, concentrated HNOa, and glacial acetic acid appeared to be good pre-
cipitating agents, although none were as effective as concentrated HC1.
TABLE 1. EFFECTIVENESS OF DIFFERENT ACIDS
ON ISOELECTRIC POINT PRECIPITATION
Precipitated shrimp
Acid protein/gal1
Concentrated HC1 303 g
Concentrated HzSOi* 152 g
Concentrated HN03 175 g
Concentrated CH3COOH 230 g
Precipitated within 2 hr.
Isoelectric precipitation of suspended and dissolved proteins is more rapid
when fresh blanch water is used. Samples frozen and stored at -5 C for <2 wks
also showed good precipitation. Those stored at -5 C in excess of 2 wks re-
quired additional levels of concentrated HC1 for effective recovery of pro-
teins. This may be due to molecular destruction caused by ice crystal forma-
tion or chemical modification under prolonged storage. Precipitation occurs
more rapidly at low temperatures.
Addition of Ca(OH)2 at concentrations of 2.5 g/L acidified blanch water re-
duced overall flocculation time by 50%, from several hours to less than one
hour. Neither OH~ alone (added as NaOH) or Ca"1"1" (added as CaClz, Ca(No3>2 or
CaHPOO achieved a similar effect. In one test, using a 2% chitosan-acetic
297
-------
acid (100 ml of a 10% solution) mixture, flocculation of the blanch water was
not observed, even at the isoelectric point of pH 4.2. Comparable results
were obtained when varying amounts of CaClz, Ca(NOs)2, and CaHPCK were added
to the chitosan-acetic acid blanch water mixture. However, supplementation
with 0.5% Ca(OH)a resulted in excellent flocculation, comparable to that ob-
tained with HC1 and 0.5% Ca(OH)2. Noteworthy differences in hardness and tex-
ture of the proteinaceous precipitate were seen with the various treatments.
The addition of calcium hydroxide or chitosan in conjunction with isoelectric
point precipitation caused a change in the appearance of the precipitated pro-
tein. With increasing concentration of Ca(OH)2, the sediment becomes some-
what harder or "cake-like," rather than the usual consistency, much like that
of toothpaste. Chitosan, on the other hand, causes the sediment to become
softer, with the consistency of mucus. It appears that Ca(OH)a is superior
to chitosan as an additional clearing agent of the supernatant layer. Fur-
ther testing is needed to see whether texture of the final product can be
manipulated through use of agents such as Ca(OH>2 or chitosan to suit a par-
ticular application. Additional evaluation of other acids and coagulants for
protein precipitation is also worthwhile.
Dry Weight Determination—Shrimp Blanch Water
Dry weight determination of the various components or groups in shrimp blanch
water represent averages from fifteen collections over the period May through
December, 1976 (Table 2). This period includes samples of both the brown
shrimp (Penaeus azteaus) and white shrimp (Penaeus setiferus).
TABLE 2. CHARACTERIZATION OF SHRIMP BLANCH WATER
Average concentration (dry wt)
g/gal lb/100 Ib
Component
Suspended meat fragments
Dissolved protein
Dissolved sugar
Other solids*
Total
H20
9.5
33.5
4.6
194.0
241.6
raw shrimp
0.11
0.39
0.05
2.27
2.82
*Includes salts, trace elements, miscellaneous organics.
Mauldin and Szabo(6) reported suspended solids in shrimp blanch water as much
as 0.19 lb/100 Ib raw shrimp processed. Currently, discounting that component
noted as "other solids" (Table 2), total suspended and dissolved solids are
0.55 lb/100 Ib^raw shrimp processed. On a gallon basis, recoverable dissolved
protein is 7.5% of the total. Both suspended meat fragments and dissolved
protein are in sufficient amounts for economical recovery, and while the
298
-------
dissolved sugars are not present in large concentrations, their removal may be
necessary to comply with pending EPA "zero discharge" regulations. It might
be possible to accumulate the remaining sugar liquor following removal of the
proteinaceous fractions, concentrate the dissolved solids, and remove these
together.
Dry Weight Determination—Raw
Shrimp Processing Water
Results obtained on the various portions of the raw shrimp processing water
are averages from the same fifteen collections made for the shrimp blanch
water (Table 3).
TABLE 3. CHARACTERIZATION OF RAW SHRIMP WASTEWATER*
Average concentration (dry wt)
g/gal lb/100 Ib
Component
Dissolved protein
Shells, legs, heads
Total
H20
3.39
21.10
24.49
raw shrimp
2.44
15.16
17.60
*Includes peeling, washing, deveining, and miscellaneous uses.
Mauldin and Szabo(6) report suspended solid loads in rawwastewaters of 3.68 Ib/
100 Ib raw shrimp processed. Table 3 shows an amount of total suspended and
dissolved solids of 2.44 lb/100 Ib raw shrimp processed. Grams of substrate/
gal recovered, less solids, is approximately 50% of that found in the blanch
water effluent. Applications for raw dissolved shrimp protein have not been
fully investigated. Possibly, the nutritive value of the extracted raw pro-
tein is superior to that of the material extracted from the blanch tanks,
since the former has not been subjected to heat treatment which might cause
protein denaturation and destruction of thermolabile amino acids such as
lysine.
Waste loads in the various effluent streams may vary on a daily or even hourly
basis. Analyses of variability in dissolved proteins in random collections of
shrimp blanch water are shown in Figure 3. Waste load variability also is
found in effluent blanch streams of the three canneries examined on the same
day (Table 4). Dissimilarities exist not only in total dissolved and sus-
pended wastes, but also in the relative amounts of the different components
comprising the effluent loads. Variability may be due to such factors as size
of shrimp, whether shrimp have been deveined, blanch period, temperature, and
salinity. Nevertheless, plant-to-plant and day-to-day dissimilarities, de-
pending on the tonnage being processed, ultimately can be calculated to pro-
vide an average amount of potential waste product recovery.
299
-------
o
5 3O -
QC
O
CO
t-
o
a:
a.
a
UJ
3
O
MAY 14 MAY 19 JUNE 23 JUNE 30
COLLECTION DATES, 1976
Figure 3. Variability of dissolved protein in shrimp blanch water (Plant A)
300
-------
TABLE 4. VARIABILITY IN CONCENTRATION OF PROTEINACEOUS MATERIAL
IN SHRIMP BLANCHING WATER*
Component
Suspended meat fragments
Dissolved protein
Total
Plant A
40.8
12.5
53.3
Plant B
0.8
25.2
26.0
Plant C
7.4
25.0
32.4
*Expressed as g (dry wt)/gal blanching water.
Another source of valuable shrimp meat is that from the discharge section of
the forced-air apparatus used to remove debris (small meat pieces and shell
fragments) originating from the turbulence of the blanching operation. In
another plant, this material was separated by hand. In both facilities, the
debris, including the food-grade shrimp meat pieces, are traditionally treated
as waste and either are incorporated into the meal or are discarded entirely.
Hand-separation of meat particles and shells revealed as much as 82% edible
food-grade meat present. Based on a discard solid shrimp load weight of 500-
1000 lb/8 hr, as much as 410-820 Ib of valuable shrimp meat are discarded
daily. Considering that often peak-load days run as long as 15-16 hr, along
with the total number of processing days in a season and number of plants in-
volved, clearly, multi-tons of valuable shrimp meat may be recovered.
Problems in load variability as noted earlier make any attempt to establish an
average treatment of shrimp blanch water extremely difficult. Therefore, to
effectively treat effluent streams, a monitoring system must be devised to in-
dicate types and amounts of waste materials contained in such discharges. In
a study of dungeness crab protein in Kodiak, Alaska in 1971(8) it was postu-
lated that the protein solubility of dissolved crab protein in blanch water at
the isoelectric point, in the main, was a linear function of the salt concen-
tration. However, as noted in Figure 4, such a linear function is not totally
applicable to shrimp blanch water. Although there appears to be a direct re-
lationship between protein solubility and increasing salt concentration from
specific gravity 1.025 to specific gravity 1.030, with increase in the speci-
fic gravity of the salt concentration to 1.040, the linear response becomes
negative. This is also true when the salinity is decreased from specific
gravity 1.040 to 1.030, accompanied by an increase in protein concentration.
Thus, monitoring of salt concentration in shrimp blanch water does not appear
to be a valid criterion of protein solubility; however, more detailed monitor-
ing is necessary to detail quantitative aspects of salinity and dissolved pro-
tein levels. The extreme variations that occur in salt concentration over
relatively short periods of time is also noted in Figure 4. This is due pri-
marily to inadequate methods of monitoring blanch water salt concentrations
during processing, which can give rise to an inconsistent product of less than
optimal quality.
301
-------
1.040 -
1.035
< 1.030
O
O
U.
O
LJ
Q_
CO
1.025
1.020
1.015
SPECIFIC
GRAVITY
GRAMS PROTEIN
DRY WT. / 250 ML. H20
ML. I2M HCL
ADDED/LITER H20
4.0
3.5
3.0
2.5
O
O
Z
z
UJ
l-
o
cc
Q_
(£
O
2.0
3 4
TIME (HRS.)
1.5
Figure 4. Blanch water concentration and recoverable protein (Plant A).
302
-------
Total Carbon Determination—
Shrimp Blanch Water
The effectiveness of nutrient removal from the blanch water, using procedures
of screening of meat fragments followed by chemical removal of dissolved pro-
tein and sugar, is seen in Figure 5. Data on carbon reduction are only quali-
tative representations of total reduction since no differentiation has been
made between total organic and inorganic carbon levels.
30 r
O
CM
I
z
O
CD
cr
<
o
o
i-
25
20
15
10
Figure 5. Reduction of total carbon in blanch water.
303
-------
The maximal single reduction of total carbon occurred following removal of
dissolved protein, i.e., an average of 7.3 g/gal. Since the shrimp contribute
an average of 10.1 g C/gal of blanch water, dissolved protein removal alone
accounts for a 73% removal of total carbon. The two other extraction proce-
dures, i.e., removal of suspended meat fragments and dissolved "sugar," ac-
counted for 26% and 4% reduction of total carbon, respectively. These values
add up to a total of 103.0% of the 10.1 g C/gal contributed by the shrimp to
the blanch water, suggesting that a portion of the carbon initially present
in the fresh blanch water was also removed. Based upon the initial total car-
bon value of 18.1 g C contributed by each gallon of unused blanch water, and
the final total carbon value of 17.8 g C in the final effluent, this method
of treatment of the blanch water may be an effective means of reducing levels
of biologically-active components present.
The overall potential shrimp protein recovery from a single plant, not includ-
ing the multi-tons of solid waste incorporated into meal, is diagrammed below.
POTENTIAL SHRIMP PROTEIN RECOVERY
(Based on 8-hr day)
Peeling, Separating, Blanch Raw/Cooked Debris
and Deveining Streams Stream and Shrimp Meat
I I t
900-1100 Ib/day 200-260 Ib/day 500-1000 Ib/day
Extrapolation of the above data to numerous localized plant operations over a
concentrated processing season, with often 16-20 hours processing/day, indi-
cates the significant magnitude of resource available for application.
Applications of Shrimp By-Products
Various uses for materials recovered from the shrimp processing operation are
listed in Table 5. A diversity of applications are seen, including use in
aquaculture diets, pet foods, concentrated flavor sources, additives to tex-
tured vegetable proteins for fabrication of shrimp products, and as a pigment
source in broiler diets. Proposed usages for chitin and its derivative chito-
san suggest further markets for exoskeleton waste.
An increasing amount of data is accumulating on analytical properties of
shrimp meal and its application in animal rations. Use of such meals in poul-
try diets has been well demonstrated, especially for purposes of imparting de-
sirable pigmentation to meat, skin or egg yolk. Frequently, source and type
of the meal is not documented, a factor of considerable importance in view of
the significant variability in proximate analysis of different processed
shrimp meals(4).
304
-------
TABLE 5. APPLICATION OF SHRIMP BY-PRODUCTS
Livestock Feed Ingredient Source of:
Tropical Fish/Bird Diets ~ Shrlmp Protein Concentrate
- Flavor Concentrates
Aquaculture Diets _ . , _.
- Carotenoid Pigments
Pet Food Supplement - Chitin/Chitosan
Use in Fabricated Shrimp Products
Earlier work from our laboratory(7) on analyses of nucleotides and amino acids
in shrimp blanch water from batch tanks has revealed significant amounts of
such materials. Concentrations of flavor enhancers such as IMP in blanch li-
quors are of particular interest. These studies have relevance to other fish-
eries industries using blanching procedures or where discharge streams carry
large organic loads. Furthermore, analyses and quality evaluations of shrimp
meal and in-plant streams, and development of nutritionally-sound shrimp meal-
fortified diets, are needed for production of least-cost diets for cultivation
of economically valuable aquatic animals. Analyses of shrimp waste protein,
designated SWP(9), have revealed proximate values of protein as high as 59%.
Further evaluation of such material(10) has shown that SWP has significant
nutritive value, improving protein quality by 74% when soybean protein in the
diet was replaced by 50% of SWP. Use of SWP in canned or processed pet foods,
or as an additive to textured vegetable proteins, has been projected.
There is an increasing need to develop and improve aquaculture diets, based on
workable feed formulation practices and good animal husbandry. More and more
emphasis is being placed in aquaculture on effective use of industrial by-
products or "wastes" as dietary ingredients to replace traditional feed com-
modities. Diet formulation practices must relate to current economics of ma-
rine and agricultural feedstuffs, problems of the seafood industry, and the
state of the art in processing techniques. These food/feed related considera-
tions are important in achieving economic viability in the nutrition and diet
development phases of aquaculture. It has been demonstrated in our work, as
well as in that of others, that shrimp by-products have valuable application
in fish and crustacean diets(11).
Shrimp-based flake diets developed at LSU(12) have been used in nutrition of
various fishes, especially freshwater and marine tropicals, specialty diets
to enhance pigmentation, breeding, etc., and supplementation and ultimate re-
placement of currently used live food in aquatic animal culture. The tropical
fish market is by no means insignificant. In analyses of sales of aquarium-
related products, foods of various types showed a 17.5% increase in 1973-74,
from 57 to 67 million dollars. Shrimp meals and pigment-fortified marine sub-
strates are receiving increasing attention as skin/flesh coloration agents in
salmon and trout diets(13). Shrimp protein, obtained as a by-product of a
chitin-recovery operation, has been effectively used as a pigment and protein
305
-------
source in diets for pond/pen raised salmonid fish.
Recent studies have evaluated water-soluble proteins from autolyzed shrimp
wastes as microbiological growth media. Data showed that a peptone derived
from such wastes compared favorably with commercially available peptones in
supporting growth of several microorganisms.
Considerable interest is being shown in recovery of chitin and chitosan, a
polyglucosamine substance from chitin, from shellfish processing wastes. Com-
mercial use of materials from the shrimp exoskeleton is being proposed in
paper-making, pharmaceutical, food-processing, agricultural, waste treatment
and monitoring, and adhesive industries. For instance, chitosan has been
found to be an effective coagulating agent for poultry processing wastes
wherein treatment reduced suspended solids in the composite effluents by as
much as 74-95%. Savings in water and waste treatment costs, plus the value
of coagulated solids as feed ingredients, may commend use of chitosan for pre-
treatment of poultry wastes and recovery of coagulated by-products.
Application of shrimp meat fragments is readily apparent in development of
flavor concentrates, reconstituted shrimp and for use with texture soy pro-
tein in fabricated shrimp products. Lyophilized cooked shrimp protein con-
centrate has a strong shrimp flavor and a pinkish-orange color, along with a
salty taste from the brine used. Both aspects can be adjusted via rehydra-
tion and comminution with vegetable protein extenders. A product such as this
could be used as a mock shrimp for human consumption, requiring only formation
of the shrimp-TSP mixture into a shape resembling a fantail or butterfly
shrimp. Recent studies have shown that the shelf life of minced fish can be
extended by comminution with shrimp meat. Rehydrated shrimp protein concen-
trate conceivably could be used in this process. Interest is being shown by
food flavor industries in the potential of concentrated shrimp flavors in sea-
food products. The shrimp industry as a whole is looking into processes and
products that will "extend" shrimp using procedures that combine shrimp meat
and flavor with vegetable proteins and fillers (soy or rice), forming new pro-
ducts that can be competitive with other staple proteins.
Certainly, any consideration of economic recovery of blanch discharge, as well
as materials from other plant streams, must involve specific plant character-
istics as well as plant-to-plant variation, and logistics of collections. How-
ever, the volume of valuable proteinaceous material involved and increasing
applications in food and feed industries is well documented. Further study of
the reduction of waste loads in shrimp processing is warranted, coupled with
in-depth evaluation of economics involved in use of the various recovered
shrimp by-products.
ACKNOWLEDGMENTS
These studies were supported by the Office of Sea Grant (NOAA), U.S. Depart-
ment of Commerce, through the Louisiana State University Agricultural Experi-
ment Station.
306
-------
REFERENCES
1. Meyers, S. P., and Rutledge, J. E. Economic utilization of crustacean
meals. Feedstuffs 43:16 (1971).
2. Meyers, S. P., and Rutledge, J. E. Shrimp meal—a new look at an old
product. Feedstuffs 43:31-32 (1971).
3, Meyers, S. P., and Rutledge, J. E. Utilization of economically-valuable
byproducts from the shrimp processing industry. Food—Drugs from the Sea
Proceedings, 1972 (ed. L. R. Worthen), Marine Technology Society, 75-85
(1973).
4. Meyers, S. P., Sonu, S. C., and Rutledge, J. E. Variability in proximate
analysis of different processed shrimp meals. Feedstuffs 45:34 (1973).
5. Mauldin, A. F., and Szabo, A. J. Gulf shrimp canning plant wastewater
processing. Proc. Fifth Nat. Symposium on Food Processing Wastes.
EPA-660/2-74-058, p. 199-217 (1974).
6. Mauldin, A. F., and Szabo, A. J. Shrimp Canning Waste Treatment Study.
U.S. Environmental Protection Agency Rept. No. EPA-660/2-74-061. (June
1974) 1-130.
7. Meyers, S. P., and Sonu, S. C. Nucleotides and amino acids in shrimp
blanching water. Feedstuffs 46:23 (1974).
8. Johnson, E. L., Peniston, Q. P., and Braun, F. W. Pollution Abatement and
By-Product Recovery in Shellfish and Fisheries Processing. U.S. Environ-
mental Protection Agency Rept. No. EPA 12130 F-JQ 06/71. (1971) 1-85.
9. Toma, R. B., and Meyers, S. P. Isolation and chemical evaluation of pro-
tein from shrimp cannery effluent. J. Agric. Food Chem. 23:632-635 (1975).
10. Toma, R. B., and James, W. H. Nutritional evaluation of protein from
shrimp cannery effluent (shrimp waste protein). J. Agric. Food Chem.
23:1168-1171 (1975).
11. New, M. B. A review of dietary studies with shrimp and prawns. Aquacul-
ture 9:101-144 (1976).
12. Meyers, S. P., and Brand, C. W. Experimental flake diets for fish and
Crustacea. Progressive Fish Culturist 37:67-72 (1975).
13. Meyers, S. P. Use of crustacean meals and carotenoid-fortified diets in
aquaculture. Feedstuffs 49: (1977).
307
-------
TOXICITY OF SOME CANADIAN
FRUIT AND VEGETABLE PROCESSING EFFLUENTS
by
A. Lamb*, C.W. Fulton*, and P. Mulyk*
INTRODUCTION
Fruit and vegetable processing operations are the source of a concentrated
biologically-active effluent and highly putrescible solid wastes which must
be treated and ultimately discharged to the land surface or a water body.
Stringent control measures are likely to be required in many cases to avoid
overloading the local assimilation capacity, while in other cases much less
stringent controls would be appropriate. An objective of regulatory agencies
is to establish rational effluent standards which will prevent pollution by
fruit and vegetable processing wastes, yet which will represent practical,
attainable goals. The costs of alternative abatement strategies must be
carefully evaluated as these will either be passed directly to the consumer
or, if this is prevented by competitive forces, may threaten the continued
economic operation of the facility.
In recent years there has been a growing awareness of environmental issues by
Canadians. The Federal Government responded to this public concern by putting
forward a major legislative program, the focal point of which was the
Government Organization Act resulting in the formation of a Department of
Environment in June 1971. The Environmental Protection Service (EPS) was
developed to cover the specific responsibility for environmental protection.
The role of EPS is one of problem identification and solution, being con-
cerned with the control and abatement of pollution by formulating a realistic
program of pollution control for Canada in cooperation with provincial author-
ities and industries concerned.
The approach of the Environmental Protection Service to pollution control is
to adopt a strategy of containment at source by means of the best practicable
technology. The regulations and guidelines developed to achieve this strategy
are based on control methods which are the best from both an economic and
environmental standpoint. Every effort is made to ensure the fullest parti-
cipation of provinces and industries concerned, the first steps being to
identify the problem in terms of an environmental inventory of the industry
and to conduct a "state-of-the-art" review including waste characterizations,
and process and waste treatment technologies. A Federal/Provincial/Industrial
task force assesses this information and recommends best practicable in-
process and treatment technology on which to base effluent regulations or
guidelines. These regulations or guidelines, as formulated by the
Environmental Protection Serivce, are subsequently reviewed with the provincial
pollution control agencies, the appropriate Canadian industry associations and
* Stanley Associates Engineering Ltd., Edmonton, Alberta
308
-------
international experts in the field with the objective of obtaining a
general concensus on the soundness of the regulations or guidelines.
The work described in this paper is the result of two projects carried out
by Stanley Associates Engineering Ltd. for the Canadian Federal Government;
a state-of-the-art review of waste treatment technology in the fruit and
vegetable industry, and an inventory of plants in Canada.
This work is now completed and will be published shortly by the Federal
Government.
DATA SOURCES
Prime sources of data for plants in the Canadian fruit and vegetable pro-
cessing industry were those individual plants at which site visits were
carried out during the course of the state-of-the-art review and the
responses to questionnaires distributed by Environment Canada to all known
Canadian fruit and vegetable processors during the Inventory Study.
Data have been presented in this paper in the metric system of units. A
table showing various conversion factors applicable to the fruit and
vegetable industry is given in Appendix 1.
FRUIT AND VEGETABLE PROCESSING INDUSTRY
Canadian Operations
It is recognized that the fruit and vegetable industry is complex; thus
eleven major commodities were designated in the original work.
Apples and apple sauce
Peaches, pears and apricots
Cherries and plums
Berries
Corn
Peas and beans (Blanched vegetables)
Beets and carrots
Tomatoes
Sauerkraut
Pickles and relishes
Jams and Jellies
It is considered that the above list of commodities reflects the major fruits
and vegetables processed in Canada based on a review of Canadian statistics.
It should be pointed out that this list is not exhaustive with respect to
the total processing industry in Canada nor does it include all the specific
commodities discussed in recent publications by the United States
Environmental Protection Agency.
The Canadian fruit and vegetable processing industry is a multi-million
dollar manufacturing activity. The value of shipments of manufactured goods
309
-------
in 1974 totalled approximately 865 million dollars. Three provinces,
British Columbia, Ontario and Quebec are the major processing areas con-
tributing 9%, 62%, and 15% of the total value of shipments, respectively.
Based on information available from Statistics Canada and other sources, the
weights of raw fruits and vegetables processed in Canada in 1975 were
estimated to be as follows:
Raw fruit processed 249,870 metric tonnes
Raw vegetables processed 851,580 metric tonnes
The total number of plants in Canada processing raw fruits and vegetables is
estimated to be 221. Large plants, that is those processing greater than
10,000 metric tonnes of raw material annually constitute only 10% of the
total number of plants, yet account for 51% of the total Canadian production.
In contrast, small plants, or those processing less than 2,000 metric tonnes
of raw material annually constitute 50% of the total number of plants but
account for only 7% of the total production. The remaining 40% of the plants
process between 2,000 and 10,000 metric tonnes annually and account for 42%
of the commodities produced.
Wastewater Loadings
Using data for the year 1975 obtained from questionnaires developed by
Environment Canada and distributed to all known Canadian fruit and vegetable
processing plants, estimates of total potential raw wastewater BOD,, and
suspended solids loadings from the industry were as follows:
BOD5 (kg) S.S. (kg)
Small Plants 0.82 x 10^ 0.30 x 10^
Medium Plants 5.73 x 10£ 1.87 x 10^
Large Plants 5.47 x 10 2.99 x 10
Total 12.02 x 106 5.16 x 106
Existing (1975) treatment systems are removing 76.2% of the total BOD,- and
79.8% of the total suspended solids loadings broken down as follows:
BOD5 S.S.
% Removed by Industry
operated biological
treatment systems 30.7 30.8
% Removed by municipal
systems 29.8 31.0
% Removed by land application
systems 15.7 18.0
Total Removed (%) 76.2 79.8
310
-------
EFFLUENT REGULATIONS
Effluent regulations are being established for many Canadian industries by
Environment Canada. These regulations generally specify a permissible
quantity of pollutant which can be discharged per ton of material processed
or per ton of finished product (eg. Ibs. of BOD,, or suspended solids per ton
of product). Effluent toxicity is another parameter frequently specified
in effluent regulations. The toxicity clause specifies a concentration of
diluted effluent in which rainbow trout must survive for 96 hours in a flow
through bioassay test.
The reasons for specifying a toxicity limit are twofold. First, fish have
direct value to the public both as a food source and for recreational
purposes. The Canadian Government is interested in protecting this resource.
Secondly, from the point of view of the regulatory agency, a toxicity standard
simplifies the effluent monitoring program. It would be difficult if not
impossible to perform routine testing for all toxic compounds which may be
present in a waste. Specifying a toxicity standard enables the agency to
control pollutant levels for many toxic parameters by performing the bioassay
test.
WASTEWATER TOXICITY
Wastewater may contain many materials which can cause mortality or adversely
upset biochemical functions of aquatic organisms. The bioassay test measures
acute lethal toxicity; that is, the ability of a waste to cause death of the
test organisms usually within a period of four days. Such effects as a pH
change which inhibits the transfer of soluble materials and gases or the
presence of high concentrations of suspended solids which result in physical
abrasion or clogging of the gills are frequently responsible for acute
toxicity.
It should be noted that the bioassay test gives no indication of sub-lethal
effects such as destruction of habitat or food chain organisms or inter-
ference with behavior or reproduction patterns. These could be as lethal
as acute effects in the long term.
A direct physical or chemical analytical procedure can be used to measure the
level of those physical and inorganic sources of toxicity. Receiving water
and effluent standards can then be set to protect fish and other aquatic
species. However, where complex organic compounds cause acute or sublethal
toxicity, direct analytical techniques have not been developed for water
quality control purposes.
In many wastewaters the specified compound causing toxicity is not definitely
known. Bioassay testing is therefore utilized to detect the presence and
concentration of toxic compounds in wastewater effluents. Any aquatic
species which are sensitive to pollutants could be utilized as a test
organism but fish have been chosen as the test animals for most toxicity work.
In Canada, the Fisheries Act is the legislation under which water pollution
311
-------
control regulations are promulgated and therefore the use of fish species
for bioassay testing is appropriate. Although fish may not be the most
pollution sensitive species in the adult form, they have the most direct
value to the public as a food source or sporting activity.
Bioassay Test Method
The bioassay test involves observations of fish mortality in varying concen-
trations of effluent. Rainbow trout are used as the test species because:
(1) there is an abundance of data on their response, (2) they are a common
and popular game fish, and (3) they are a particularly pollutant intolerant
species.
Large volumes of wastewater are collected over a 24 hour period, refrigerated
to 5°C and shipped to the test facility. Varying concentrations of waste and
dilution water are prepared in the range of 0 - 100%. The 0% or control
sample, consisting entirely of dilution water, must have less than 10% deaths
or the test is considered invalid.
The test conditions are defined and are carefully controlled. Temperature
is usually maintained at 15°C + 1°C. Aeration is supplied to provide a
minimum dissolved oxygen concentration of 8 mg/1. pH may or may not be
adjusted depending upon the objectives of the test. Fish size and volume of
test solution are also controlled. Fish which have been acclimated to
dilution water are then introduced into each of the vessels containing a
different concentration of waste.
Ideally, a continuous flow (once through) system is used to prevent
accumulation of waste materials generated by the fish or depletion of toxic
materials in the effluent sample. However, continuous flow bioassays require
elaborate equipment and relatively large waste samples (up to 1000 gallons
as compared to 50 to 100 gallons for static testing). Continuous flow
testing therefore tends to be much more expensive and difficult to run than
the simpler static test. As a result there is usually an attempt to develop
a correlation between flow through and static tests so that the latter can
be used for routine monitoring purposes.
Fish deaths are observed and recorded at regular intervals for a 96 hour
period. The percent mortality at 96 hours in each test vessel is then plotted
against the effluent concentration in the vessel on semi-logarithmic probabil-
ity paper and the best straight line fitted through the data as shown in
Figure 1. The parameter used to describe effluent toxicity is the LC,.- value
or the lethal concentration for 50% mortality. It is the effluent concentra-
tion at which 50% of the test organisms would die in 96 hours. It should
be noted therefore that the smaller the LCr0 value the more toxic the waste.
The toxicity levels utilized in reporting results were as follows:
Non Toxic - no mortality during the 96 hour test at any effluent concentration
including undiluted effluent.
312
-------
98 r-
D
O
I
(0
0)
t-
h
_J
<
t-
£
O
LLJ
O
a:
UJ
Q.
90
80
70
6O
5O
40
30
20
1O
LC 5O=34% by volume
I
I
1O 2O 3O 5O 7O 1OO
EFFLUENT CONCENTRATION (%) BY VOLUME
FIGURE 1. Plot of Bioassay Test Results (After Bissett; 1975)
313
-------
Partly Toxic - a number of fish die in the test but less than 50% of them.
LC5Q - 1 - 100% - half of the fish die in a 96 hour period in the concen-
tration of effluent specified.
FRUIT AND VEGETABLE PROCESSING WASTEWATERS
Raw Screened Wastewater Quality
The data presented in Table 1 was compiled as a result of site visits to
eight Canadian fruit and vegetable processing plants. Raw waste samples
could only be obtained at the five plants shown. Raw screened wastewater
from all plant samples was high in biologically degradable material as
evidenced by BOD values generally ranging between 500 and 2000 milligrams/
litre. Suspended solids concentrations typically fell between 300 and 500
milligrams/litre. :
LC,_n values were also obtained for the wastes using 96 hour static bioassay
testing. All wastes exhibited acute toxicity to rainbow trout with LC,-,.
values ranging between 12 and 36%. From inspection of these results, it
would appear that the toxicity is probably due to the high concentrations of
suspended solids and organic material. The number of samples obtained during
the study were not sufficient to state the cause of the toxicity with any
statistical reliability.
An examination of the raw waste toxicity in terms of LC was also made at
24, 48, and 72 hours as shown in Table 2. It is apparent that the 72 hour
value is in virtually all cases identical to the 96 hour static bioassay
test results. Similarly, in most cases the 48 hour LC value is also in
very close agreement with the 96 hour value. However, as there is a signifi-
cant difference between the 24 hour and 48 hour values, it would be advisable
to utilize a minimum duration of 48 hours in future testing.
Treated Effluent Quality
Treated effluent quality for the eight plants surveyed is illustrated in
Table 3. These plants demonstrated performance somewhat better than the
Canadian average with BOD and suspended solids removals generally greater
than 80%. ^
Toxicity was eliminated in effluent from all plants with the exception of two
of three corn processors.
At plant number 3 where effluent LC5Q values of 50%, 71%, 75% and 100% were
recorded, treatment consisted of an aerated lagoon with settling. Raw
screened effluent samples were not available from this plant and consequently
raw waste toxicities could not be determined. However, treatment of corn
processing wastewaters exhibiting raw wastewater toxicity of 18.0 and 17.0% at
plant numbers 5 and 8, respectively, was successful in reducing the effluent
toxicity to the point where less than 50% of the test fish died in full
strength effluent.
314
-------
TABLE 1
RAW SCREENED WASTEWATER CHARACTERISTICS
Plant
No. Commodity
CO
en
4
5
6
7
8
Tomatoes
Corn
Beets
Apples
Apples , Oranges
& Pears
(Juice Plant)
Corn, Plums
Jam, Broccoli
BODC
3
100-610
494-1400
1650-4940
748-2880
875-5300
1182-5108
SS
183-364
162-488
402-675
56 - 178
26-2510
320-580
£H
4.7-10.2
4.6-5.3
5.0
7.3 - 7.4
5.6-8.2
5.3-7.2
Total P
4.4-9.7
3.0-12.0
4.3-12.0
9.8 - 11.2
0.5-3.8
1.0-4.6
TKN LC 0
_>LJ
2.5-18.2 13.5-36.0%
5.2-14.9 18%
10.0-16.6 12.5%
0.9-16.8 N.T.-P.T.
0.8-18.5
0.5-2.5 13%-17%
NOTE: N.T. - Non-toxic
P.T. - Partly toxic
-------
TABLE 2
COMPARISON OF LC5Q FOR VARIOUS STATIC TEST DURATIONS -
RAW SCREENED WASTEWATER
Commodity 24 Hr 48 Hr 72 Hr 96 Hr
Apples - 24.0 24.0 24.0
38.0 38.0 38.0 27.0
Beets 21.0 12.8 12.5 12.5
Corn 23.0 20.0 18.0 18.0
Corn 16.0 13.0 12.5 12.5
18.0 17.0 17.0 17.0
Tomatoes 45.0 36.0 36.0 36.0
Tomatoes 42.0 42.0 13.5 13.5
316
-------
TABLE 3
TREATED EFFLUENT CHARACTERISTICS
Plant
No.
1
2
3
co
-»g
4
5
6
7
8
Commodity
Tomatoes
Pears
Corn
Tomatoes
Corn
Beets
Apples
Waste
Treatment
Process
Spray Irrigation
Spray Irrigation
& Holding Pond
Aerated Lagoon &
Sedimentation
Trickling Filter
Aerated Lagoon &
Settling
Spray Irrigation
Aerated Lagoon
Apples, Oranges & Aerated Lagoon &
Pears (Juice Plan $ Sediment at ion
Corn, Plums,
Jam & Broccoli
Oxidation Ditch
& Settling
BODc;
81-90
924-1532
415-435
5-19
366-374
148-234
158-1400
15-95
1145-1685
SS
28-48
86-89
1730-3250
6-19
26-55
21-22
31-42
15-63
240-370
.El
7.3-7.4
6.6-7.1
7.0-7.7
6.2-7.4
7.0-7.6
6.3-7.1
7.3-7.4
7.7-8.1
6.1-7.3
Total P
0.5^.0. 7
2.1-2.9
9.5-26.0
0.4-4.0
0.5-1.1
0.2-0.25
9.8-11.2
0.5-1.3
0.1-4.3
TKN
1.1-1.4
2.2-3.2
40.4-76.2
0.4-2.3
1.4-1.6
1.0-1.6
0.9-16.8
1.0-2.0
1.0-2.6
LC50
N.T.
N.T.
100%
71%
75%
P.T.
N.T.
P.T.
N.T.
N.T.
N.T.
N.T.
NOTE: N.T. - Non-toxic
P.T. - Partly toxic
-------
Examination of the chemical analyses of the toxic effluents shows that there
is no firm relationship between toxicity and measured parameters, with the
exception of suspended solids concentration. In the case of the treated
corn processing wastes which exhibited LC,-~ values ranging from greater
than 50% to 100%, there were high total Kjelkahl nitrogen concentrations
indicative of high protein content. It should be noted that wastewater
effluents from the meat processing industry exhibiting high protein levels
have been shown to exhibit high toxicity.
Conclusions
A number of conclusions can be reached regarding the toxicity of effluents
from the fruit and vegetable processing industry based on the limited amount
of data obtained from the plant surveys.
1. Raw screened wastewaters from the following commodities demonstrate
toxicity to rainbow trout, based on static 96 hour bioassays:
Tomatoes
Corn
Beets
Apples
Corn, plums, jam
Corn, plums, broccoli, jam.
2. No definite chemical cause for toxicity could be determined from
the limited testing program.
3. Corn processing wastes were the most toxic of the wastewater effluents
examined. Since corn makes up a considerable percentage of Canadian
vegetable processing, further investigation is required to determine
the causes and methods to reduce the toxicity of corn processing
effluents.
4. The high organic nitrogen content of the corn processing effluents
may have been contributory to effluent toxicity.
5. It was found that 48 hour LC,.,. values on raw screened wastewater
were in very close agreement with 96 hour static bioassay test
results. However, it is recommended that a minimum duration of 48
hours be used in future testing, as there was a significant difference
between 24 and 48 hour LC values.
6. The toxicity level which could be specified in the regulations for
fruit and vegetable processing wastewater effluents is no acute
toxicity in full strength effluent.
Only a limited number of raw and treated effluents from selected fruit and
vegetable commodities were tested. For each industrial effluent sampling was
also limited. Since acute toxicity was found to be a problem with corn wastes,
processing wastewaters from all commodities should be tested for toxicity
/
318
-------
before and after biological treatment on a statistical reliable number of
samples. As well, where toxicity persists after secondary treatment,
specific cause of toxicity should be determined.
319
-------
TABLE Al.
CONVERSION FACTORS
English Unit X
acre
acre
acre-ft
cu ft
cu in
cfm
cfs
ft
°F
gal (Imp)
gal (US)
gpd/sq ft (US)
gpm (Imp)
gpm (US)
in
Ib (mass)
Ib/cu ft
lb/1000 cu ft
Ib/ day /acre
Ib/day/acre - ft
Ib/day/cu ft
Ib/day/sq ft
lb/day/1000 sq ft
Ib/ton
mgd (Imp)
mgd (US)
sq ft
ton
ton
Multiplier =
4,046
0.405
1,234
0.028
16.39
0.2832
1.70
0.3048
0.5555 (°F-32)
0.004546
0.003785
0.0408
0.2728
0.2271
2.54
0.4546
16.02
16.02
0.112
3.68
16.02
4.880
4880
0.5
4546
3,785
0.09290
9072
0.9072
S.I. Unit
2
m
ha
m3
m3
cm3
q
m /min
vr/mLn
m
Op
m3
m
m3/day/m
m^/hr
in /hr
cm
kg
kg/m
g/m
g/day/m
g/day/m
kg/day/m
g/day/m
g/day/m2
kg/t
m3/day
m^/day
m2
kg -
t
320
-------
TABLE A2
SYMBOL DEFINITION
Symbol Definition
cu ft cubic feet
in inch
cfm cubic feet per minute
cfs cubic feet per second
gpd gallons per day
Ib pound
sq ft square foot
ton short ton
mgd million gallons per day
m metre
ha hectare
cm centimetre
kg kilogramme
t metric tonne
321
-------
REDUCTION OF WASTES FROM CUCUMBER PICKLE PROCESSING
BY USE OF THE CONTROLLED CULTURE FERMENTATION PROCESS
by
L. W. Little*, R, Harrison**, J. Davis*, J, Harris**, and S. J, Dunn**
INTRODUCTION
Several years ago Hoover (1) posed the question which must be answered in
regard to prevention of food processing wastes: "Can we change existing pro-
cesses so that less waste is produced, while maintaining or improving product
quality?" The objectives of the project described herein were addressed to
answering this question in regard to the cucumber pickle industry:
(1) to demonstrate on a commercial scale a substantial reduction
in salt usage in brining cucumbers by substitution of the
controlled culture fermentation (CCF) procedure for the
currently used natural fermentation (NF) procedure
(2) to compare progress of controlled culture fermentations and
natural fermentations under actual tankyard conditions
(3) to compare quality of brinestock and pickles produced by
CCF with those produced by NF
Currently about 50-60 per cent of the cucumber pickles which are produced in
this country are prepared from brinestock. In this case, the cucumbers are
brought in from the fields, placed in a salt brine of approximately 25°
salometer strength and allowed to ferment naturally. The bacteria and other
microorganisms indigenous to the cucumbers and attached soil ferment sugars
which leach out of the cucumbers, converting them to acid endproducts. The
main function of the brine is to inhibit growth of undesirable microorganisms
while favoring growth of lactic acid producers such as Laotobac'Lllus species.
After the active fermentation period, when the sugars have been consumed,
the pH has dropped to about 3,5, and sufficient acidity has developed, the
brined cucumbers may be stored for 3 months or more before they are used to
*Department of Environmental Sciences and Engineering, University of
North Carolina at Chapel Hill, 27514
**Department of Plant Science, North Carolina Agricultural and Technical
State University, Greensboro, North Carolina, 27411
322
-------
prepare finished pickles. During the storage period the salt concentration
is gradually increased to 45-60° salometer in order to stop bacterial acti-
vity and preserve the cucumbers.
When the cucumbers are removed for processing, the spent brine is drained
off and is usually discharged directly (via a pipeline) or indirectly (via
drainage onto the tankyard and subsequent runoff) into the plant's waste-
water treatment system. The discarded brine has a high pollution potential
(Table 1) and accounts for most of the salt in the wastewaters from pickle
manufacturing (2). Since salt is unaffected by activated sludge or aerated
lagoons, the usual methods of wastewater treatment, it passes into the
effluent and thence to a receiving stream or to a municipal wastewater treat-
ment plant. If inadequate dilution with freshwater is not available, the
salt can cause high total dissolved solids levels which render the receiv-
ing stream unfit for aquatic life, for irrigation, or for a potable water
source. It is anticipated that effluent guidelines will soon require
greatly reduced dissolved solids levels in effluents from pickle plants
which discharge to freshwater streams or to municipal wastewater treatment
plants.
TABLE 1. TANKYARD BRINE COMPARED TO TYPICAL DOMESTIC SEWAGE
Parameter Brine Sewage
TOC, mg/1
SS, mg/1
Kjeld-N, mg/1
TP, mg/1
Cl, g/1
PH
3400
330
732
87
111
3,4
124
170
31
10
<1
7,2
In addition, the high salt levels can interfere with operation of the plant's
biological treatment system, and they represent loss of salt, ar ingredient
which is cheap on a cost per unit weight basis,but which is expensive in
terms of the total amount required for pickling or in terms of cost to re-
move it from the wastewater. These and other pollution problems associated
with tankyard operations have been analyzed extensively in a previous re-
port (2).
Despite the heavy use of salt, natural fermentations have been described as
"unrestricted, heterogeneous, highly complex, and variable," often leading
to production of defective brinestock (3), Lowered brinestock quality is
especially apt to occur if pectinase-producing yeasts or coliform-type
bacteria grow in the brine,
323
-------
Because of the problems and the unpredictability of natural fermentations,
the U. S, Food Fermentation Laboratory has developed the controlled culture
fermentation (CCF) process (3). CCF brining procedure includes washing the
cucumbers before they are tanked, sanitizing the cucumbers and the tanks
with brine containing a small amount of chlorine, and inoculating the brine
with concentrated cultures of rapid-growing lactic acid bacteria, Because
most of the undesirable microorganisms have been eliminated by the washing
and sanitizing procedures, and because the heavy inoculum of the lactic acid-
bacteria tends to suppress growth of any which have survived these proce-
dures, the fermentation takes place rapidly and high levels of acidity are
quickly achieved. If freezing weather conditions are not a consideration,
the brinestock can be stored at 25° salometer (6.6%), requiring about half or
less of the salt commonly used,
The CCF process had been extensively tested in the laboratory and in rela-
tively small tankyard studies (3). However, studies on a larger scale,
accompanied by assessment of the potential for reducing pollution, had not
been conducted.
This project was initiated in 1976. Funded jointly by the U.S. Environmental
Protection Agency and the Perfect Packed Products Division of the Heinz Co,,
Henderson, N. C., the project was conducted by scientists from North Carolina
A & T State University and the University of North Carolina at Chapel Hill
with aid of Heinz personnel and the U.S. Food Fermentation Laboratory,
Three tankyard experiments were conducted during the 1976 green season. In
each case, CCF and NF tanks were set up and fermentation progress, brinestock
quality and quantity, spent brine quality and quantity^_ and finished product
quality and quantity were compared.
MATERIALS AND METHODS
Materials
Tanks. The Heinz Company provided 800-gallon wooden vats coated with fiber-
glass and fitted with stainless steel sampling channels. Vats used for CCF
were fitted with a nitrogen purging system comprised of a loop of tubing
pierced with holes (1/64-in) connected to a tank of compressed nitrogen. A
rotometer and flow-control valve were used to measure and control No flow.
CCF tanks were purged continuously during the initial phases of the fermen-
tation at a rate sufficient to keep CC^ concentrations below 20 mg/100 ml.
NF tanks were not purged. Before the tanks were covered with plastic netting
the tanks were headed by widely spaced wooden boards in which holes had been
drilled to facilitate gas exchange.
Cucumbers. Size 3 (1 1/2-2" in diam) cucumbers were used for all experi-
ments as it was felt that the large size would be the most likely to bloat
and would thus provide the most stringent test of the CCF procedure. The
green stock was obtained from that being supplied to the pickle company.
The percentage of solids to liquids was 65% cucumbers and 35% brine (wt/wt),
324
-------
Salt. The salt used was the rock salt commonly used at the plant,
CCF procedure, The CCF procedure was that suggested by Etchells et al (3).
In-tank shrinking was employed, Chlorination was achieved by addition of
calcium hypochlorite (Lo-BaxR, Olin Corporation), The only exception to the
procedure was omission of the second chlorination 10-12 hours before inocu-
lation. It was assumed that sufficient sanitizing was provided by the wash-
ing and initial chlorination and that further chlorination after the cucum-
bers had been soaking in the brine might cause production of undesirable
chlorinated organic compounds.
Cultures. Cultures of Laotobae-Lllus plantainm were obtained from Chr,
Hanson Laboratories and Miles Laboratories,
NF procedure. The NF procedure was that commonly used by the pickle com-
pany.
Routine analyses. Salt concentration was measured with a salometer, Acid-
ity was measured as lactic acid by titration with 0.1 N sodium hydroxide,
using phenolphthalein indicator. Temperature was measured with an ordinary
laboratory thermometer. C02 was estimated by a modification of the Harleco
procedure (4) .
Brinestock evaluation. Representative samples of brinestock were obtained
by passing a lucite cylinder (1 ft ID) to the bottom of the tank and remov-
ing all the brinestock by netting. Two cores were removed from each tank.
The brined cucumbers were cut longitudinally and examined for balloon,
honeycomb, and lens type bloaters (5). Texture was evaluated with a Maagde-
borg pressure tester.
Finished product evaluation, The brinestock was used to prepare hamburger
chips and whole dill pickles, packed in 5-gal plastic pails. Chips and
whole pickles were taken from selected pails, coded, and evaluated by a
panel of 6 to 8 persons, using the rating sheet devised by the U. S. Food
Fermentation Laboratory,
RESULTS AND DISCUSSION
During the project three experiments were conducted. During experiment I
two tanks were brined, one NF and one CCF; during experiment II, two NF and
two CCF; and during experiment III, one NF and two CCF, Figure 1 shows the
progress of fermentation in the tanks, as indicated by lactic acid product.
In the case of duplicate tanks, the average value is shown. The median
afternoon brine temperatures were 80.5 °F (Experiment I); 84.2 °F (Experi-
ment II); and 84.8 (Experiment III),
The initial acidity was higher in CCF tanks due to the initial addition of
acetic acid, as directed in the CCF procedure. Note that the rate of acid
production was generally more rapid in CCF tanks and that higher final
acidities were attained in these tanks. This indicates the rapid and
desirable activity of the bacteria used for the inoculum, and it also indi-
325
-------
1.3
1.2
1.1
1.0
0.9
0.8
0.7
•a '
CO -H
INJ U
0.6 -
0.5
0.4
S 0.3
>,
u
•H
•H
U
0.2 -
** 0.1 -
c*
.o
o
• Controlled Culture Fermentation
O Natural Fermentation
Figure 1.
1 2 345 678 9 10 11 12 13 14 15 16 17 18 19
Time (days)
Comparison of lactic acid production in controlled culture and natural fermentations
(three experiments).
-------
cates that the fermentation time can be shortened by use of the CCF proce-
dure.
In each case the brinestock produced in the vats was evaluated for texture
and for defects, especially evidence of bloating. The overall acceptability
and determination of usable brinestock was computed by numerical systems
developed by Fleming et al (6) and by S. D. Rubin of the Heinz Co. A summary
of the evaluation is shown in Table 2, which indicates that CCF brinestock
had slightly better texture than NF brinestock and that CCF brinestock typi-
cally had a low bloater index, It is also apparent that the quality of CCF
brinestock was much more consistent and predictable than that of NF brine-
stock, despite the fact that the NF experimental tanks received much more
attention than would a tank in the average tankyard. It can be concluded
that there would be considerably less waste of brinestock from CCF tanks.
TABLE 2, EVALUATION OF BRINESTOCK FROM NATURAL FERMENTATIONS AND
FROM CONTROLLED CULTURE FERMENTATIONS
Natural Fermentations _ . ,. _ , ,.
Examination for Freedom from «
Pressure Test Defects Bloater
//Tested Average (Ib) // Examined Estimated Yield (%) Index
40 16,7 200 82,4 13.8
40 20.0 200 97.0
40 17.0 200 96 JO
III 40 20,5 160
Average
Controlled Culture Fermentations
I 40 17.8 200 97,5 2.1
II 40 19,5 200 93.0 1,9
40 18,5 190 92,5 1.2
III 40 21,5 200 97,5 1,0
40 22.0 200 95,5 2,3
Average 19,9 95.2 1.7
-Using system devised by S, D, Rubin, Perfect Packed Products, Inc.
Fleming et al (6)
327
-------
The finished products from 6 of the experimental tanks have been evaluated by
a six-member panel made up of USDA, Perfect Packed Products, and A & T per-
sonnel. Hamburger chips and whole dills were prepared, Figure 2 indicates
that in terms of appearance, taste, and texture, the chips are very similar
in quality and compare well with those produced commercially. Figure 3 indi-
cates that appearance and texture of whole dills prepared from CCF is notably
better than those from NF, while taste ratings are similar.
Table 3 shows the volume and salt loading of the wastewaters from the two
processes. Unexpectedly, the volume of wastewater generated by the two pro-
cesses did not differ greatly, Part of this can be explained by the failure
to sufficiently desalt the NF brinestock in the first experiment. In addi-
tion, it was observed that the plant personnel tended to vary time of desalt-
ing based on the amount of salt left in the cucumbers. On the other hand, in
terms of salt to be discharged in the wastewaters there is a very striking
difference in the two types of fermentations. This difference would be great-
er if the tanks were held for longer periods and salt addition to NF tanks
continued.
In summary, the following conclusions may be drawn:
1) CCF fermentations proceed more rapidly and result in higher levels
of acidity than do natural fermentations
2) CCF produces superior brinestock quality
3) CCF results in dramatic reduction in amount of salt to be dis-
charged
4) Whole dills produced from CCF are equal in taste^.and superior in
appearance and texture to those from NF
5) Hamburger chips from the two processes are similar.
This study was conducted as part of a contract under the sponsorship of the
U. S. Environmental Protection Agency in cooperation with Perfect Packed
Products Co., Inc., Division of H. J. Heinz Co.
REFERENCES
1. Hoover, S. R. Prevention of food-processing wastes. Science IBS; 824
(1974).
2. Little, L. W., Lamb, J, C,, and Horney, L. F. Characterization and Treat-
ment of Brine Wastewaters from the Cucumber Pickle Industry. UNC Water
Resources Research Institute Report No, 99 (1976).
3, Etchells, J. L., Bell, T. A., Fleming, H, P., Kelling, R, E,, and
Thompson, R, L, Suggested procedure for the controlled fermentation of
commercially brined pickling cucumbers — the use of starter cultures and
reduction of carbon dioxide accumulation. Pickle Pak Science 3: 4 (1973).
4. Fleming, H. P,, Thompson, R. L,, and Bell. T, A. Quick method for esti-
mating CO^ in cucumber brines. Advisory statement from the U.S. Food
Fermentation Laboratory, Agricultural Research Service, USDA, Raleigh
N. C. (1974). '
328
-------
Commercial
10 _
Appearance
Average ratings:
NF
CCF
Taste
Appearance
8.2
8,3
Texture
Taste
7^2
Texture
7.6
Figure 2. Evaluation of hamburger chips prepared by natural fermentation
(including a commercial product) and by controlled culture
fermentation.
329
-------
Appearance
Average ratings:
NF
GCF
V//A
Taste
Texture
Appearance
7.0
8.2
Taste
6.3
6.1
Texture
6.9
7.8
Figure 3 >. Evaluation of whole dill pickles prepared by natural fermentation
and by controlled culture fermentation.
330
-------
TABLE 3. NATURAL FERMENTATION COMPARED TO CONTROLLED CULTURE FERMENTATION5
COMPARISON OF VOLUME OF WASTEWATER AND AMOUNT OF SALT TO BE WASTED
CO
co
Type
NF
CCF
Expt.
I
II
III
I
II
III
Spent
Vol (1)
1575
1285
1015
1350
1230
1190
1250
1180
1170
Brine
Salt (103g)
212,6
173,5
142,0
145,8
102.6
102,3
99,7
111.6
79,6
1st Process Water
Vol (1) Salt (103g)
790
200
200
1260
770
1350
1255
850
885
63,2
12,5
14,8
89.5
39,2
50,1
45,2
53.1
44.2
2nd Process Water Total
Vol (1) Salt (103g) Vol (1) Salt (103g)
2366
1040 65,0 2525
1060 72.3 2275
„•-_- 2610
Average
., — 2000
2540
2505
2030
2055
275,8
251,0
229,1
235.3
248
141.8
152,4
144.9
164.7
123.8
Average
146
-------
5. Etchells, J. L. Bell, T, A,, Fleming, H, P,, Kelling, R. E,, and Thompson,
B. L, Bloater chart. Published and distributed by Pickle Packers Inter-
national, Inc., St. Charles, 111, (1974),
6. Fleming, H. P., Thompson, R. L., Bell, T, A., and Monroe, R. J, Effect
of brine depth on physical properties of brine-stock cucumbers. In press.
332
-------
SALMON PROCESSING WASTEWATER TREATMENT
by
P. A. Bissonnette*, S. S. Lin*, and P. B. Liao*
INTRODUCTION
The Skokomish salmon processing plant was built to provide a more efficient
system for handling fish from the Indian commercial fishery on the Skokomish
River. It is located on the Olympic Peninsula near the mouth of the Skokomish
River about 45 miles southwest of Seattle, Washington (Figure 1). The
processing plant was designed in 1971, constructed in 1972, and has been
in operation since December, 1972. It consists of a fish preparation area
where salmon are hand-butchered and cleaned, smokehouses, refrigeration/
freezing capacity, and a retail outlet (Figure 2). The processing plant
is capable of processing 590 Ib/hr for yearling salmon and 2,000 Ib/hr
for large salmon.
In order to comply with regulatory requirements, the wastewater treatment
facility was constructed in conjunction with the processing plant in April,
1975, and a water quality monitoring program was initiated in September,
1975, to evaluate its performance. The treatment facility consists of an
extended aeration system and two identical aerobic polishing ponds. The
rational used for the design of the treatment facility was based on litera-
ture review and characteristics of the waste as determined by daily grab
samples. The literature review was limited since very little has been
published on the characteristics of salmon processing wastes. The EPA
effluent guidelines for salmon processing wastes were also developed based
on this same small data base. The objectives of this study were to evaluate
the EPA recommended effluent limitation guidelines with respect to the
extended aeration treatment process and to develop more reliable design
criteria.
CHARACTERISTICS AND TREATABILITY OF SALMON PROCESSING WASTES
A complete literature review for fresh-frozen processing wastes is reported
in an earlier paper(1). In general, little has been published about the
characteristics of waste generated by salmon processing and its treatability.
Possibly the most reliable data for salmon processing wastes was provided
by a seafood waste survey(2) of six plants in Alaska and one in the Northwest.
Table 1 summarizes the waste loads from all hand-butchered salmon processes
studied during, the survey.
*Kramer, Chin & Mayo, Inc., Consulting Engineers, Architects and Applied
Scientists, 1917 First Avenue, Seattle, Washington.
333
-------
SKOKO
INDIAN
RESERVATIO
1 LOCATION AND VICINITY MAPS
SKOKOMISH PROCESSING PLANT
SHCLTON, WASHINGTON
334
-------
CO
CO
cn
WASHWATER
1
FRESH FISH.
DRESSED
SORTED
BY WEIGHT
FREEZING
BRINE
SMOKING
PACKING
SHIPPING
II FRESH SALMON
21 FROZEN SALMON
31 SMOKED SALMON
FIGURE 2 PROCESS LAYOUT OF SKOKOMISH
SALMON PROCESSING PLANT
SHELTON, WASHINGTON
-------
TABLE 1. HAND-BUTCHERED
Parameter
Production, ton/hr
Process Time, hr/day
Flow, I/sec
(gal/min)
Flow Ratio, 1/kkg
(gal/ ton)
Sett. Solids, ml/1
Ratio, 1/kkg
Sol. Solids, mg/1
Ratio, kg/kgg
Sus. Solids, mg/1
Ratio, kg/kgg
5-Day BOD, mg/1
Ratio, kg/kgg
COD, mg/1
Ratio, kg/kgg
Grease/Oil, mg/1
Ratio, kg/kgg
Organic-N, mg/1
Ratio, kg/kgg
Ammonia-N, mg/1
Ratio, kg/kgg
PH
Temp., degrees C.
SALMON PROCESS
Mean
2.13
6.34
2.15
34.1
5,040
1,210
1.02
5.15
193
0.971
236
1.19
493
2.48
1,070
5.36
341
1.72
80.9
0.407
2.12
0.011
6.73
13.2
SUMMARY*
Std. Dev.
1.09
1.80
1.09
17.2
3,100
743
1.19
5.99
155
0.782
185
0.933
179
0.900
601
3.03
628
3.16
40.0
0.202
0.794
0.004
0.318
2.51
5% Min.
0.733
3.67
0.754
11.9
1,410
338
0.109
0.547
37.5
0.189
47.6
0.240
233
1.17
332
1.67
15.0
0.076
29.0
0.146
0.979
0.005
6.25
9.19
95% Max.
4.38
8.98
4.39
77.5
13,000
3,120
4.18
20.5
600
3.12
722
3.54
923
4.35
2,600
13.1
1,770
8.39
181
0.314
4.14
0.120
7.13
15.7
*From Environmental Associates, Inc., "Draft - Canned and Preserved Fish
and Seafoods Processing Industry," February 1974.
336
-------
Various investigators have reported the success of using biological treatment
for seafood processing wastes. In three studies CD(2)(3) at least 80-90
percent removal efficiency of BOD was achieved. The success of the extended
aeration process was dependent on controlled waste loadings.
The EPA effluent limitations guidelines for salmon processing plants are
presented in Table 2.
TABLE 2. RECOMMENDED EFFLUENT LIMITATIONS GUIDELINES
FOR SALMON PROCESSING PLANTS*CD
DAILY MAXIMUMMAXIMUM 30-DAY AVG.
PARAMETER kg/kkg Ib/ton kg/kkg Ib/ton
5-Day BOD
Total Suspended Solids
Grease and Oil
4.7
3.2
7.0
9.4
6.4
14.0
3.2
2.0
4.9
6.4
4.0
9.8
*0nly for West Coast hand-butchered salmon.
THE SKOKOMISH WASTEWATER TREATMENT SYSTEM
The Skokomish processing plant's extended aeration treatment system, hereafter
called the full-scale plant, was designed based on expected loadings from
projected production at 2,000 Ib/hr for large salmon and 590 Ib/hr for
yearling salmon. It was also projected by the Skokoraish tribe that the
processing plant would eventually double in capacity. Table 3 summarizes
the actual production schedule since the project's initiation. The peak
production period is July through October rather than September through
January as projected. In fact, for the past two years no fish have been
available for processing in January. The bulk of the fish processed during
February through July are yearling salmon from a private fish supplier.
The large salmon processed are generally those caught by members of the
tribe.
Based on the projected production and literature values for hand-butchered
salmon wastewater characteristics (Table 1), the extended aeration system
was designed. Table 4 summarizes the design criteria, and Figure 3 shows
a flow diagram of the facility.
The projected levels of incoming fish to process was not realized and the
decrease in operations resulted in lower flows and longer retention times
through the treatment facility than expected. A smaller scale extended
aeration system was constructed a year later (September, 1976), hereafter
called the pilot plant, to test lower retention times and higher loadings.
Table 5 summarizes the design criteria for the pilot plant, and Figure 4
shows a diagram of the facility.
337
-------
TABLE 3. SALMON PRODUCTION SCHEDULES
FROM SEPTEMBER THROUGH FEBRUARY, 1977
Month Monthly Total (Ib)
1975 September
October
November
December
1976 January
February
March
April
May
June
July
August
September
October
November
December
1977 January
February
78,382***
75,191
29,865
8,620
-0-
49,800
64,570
-0-
38,350
34,070
59,890
150,350
44,539***
94,176
22,548
28,050
-0-
-0-
Daily Average* (Ib) Size**
7,838
3,760
1,493
431
-0-
2,490
3,228
-0-
1,918
1,704
2,994
7,518
4,454
4,709
1,127
1,402
-0-
-0-
Large
Large
Large
Large
_ _ _
Small
Small
Small
Small
Small
Small
Small/Large
Large
Large
Large
Large
- _ _
*Based on 20 working days per month and one shift per day.
**Large fish approximately 10 Ibs; small fish approximately 11 oz
***Value for last half of September.
338
-------
TABLE 4. SUMMARY OF DESIGN, WASTEWATER TREATMENT FACILITIES
Waste Characteristics
Source * Process water, no sanitary waste
Minimum Flow » 0 gpd
Maximum Flow - 20,000 gpd
Peak Flow » 18 gpm
Daily Flow Variation » Continuous over approximately 18 hours
Average BOD Concentrations - 500 mg/1
Maximum BOD Loading * 80 Ib/day
Extended Aeration System
Aeration Tank:
Volume » 20,000 gallons
Detention Time at Maximum Flow » 24 hours
Maximum Oxygenation Available
Maximum Loading - 0.03 Ib BOD/cu ft/day
240 Ib 02/day
Clarifier:
Volume - 3,000 gallons
Detention Time at Maximum Flow » 2.7 hours
Surface Area - 98 sq. ft.
Overflow Rate at Maximum Flow - 202 gpd/sq. ft.
Polishing Ponds
Surface Area -0.37 acre
Loading at Maximum Flow - 21.5 Ib BOD/acre/day
Volume - 364,000 gallons
Detention Time at Maximum Flow - 18 days
339
-------
SALMON
PROCESSING
PLANT
PUMP STATION
RAW
WASTEWATER '
EXTENDED
AERATION
PACKAGE PLANT
4" PIPE
4" PIPE
POND
"^ NO. 1
6" PIPE
4" PIPE
6" PLASTIC (PERFORATED)
DRAIN PIPE
4 "PIPE
POND
NO. 2
FIGURE 3. FLOW SCHEME OF THE WASTE
TREATMENT SYSTEM
SKOKOMISH PROCESSING PLANT
SHELTON, WASHINGTON
-------
TABLE 5. DESIGN CRITERIA FOR SKOKOMISH PILOT TREATMENT PLANT
Aeration Chamber
1) Dimension:
Diameter 15 ft.
Depth 4 ft.
Total Volume - 4,500 gal. @ 3.5 ft.
Each chamber has 2,250 gallons
2) Air supply roots frame No. 22 Blower - 14 cfm @ 3 psi
3) Chamber divider, plastic sheeting siliconed in place with baffles
to equalize head.
Clarifier
Volume 135 gallons each
2
Surface area 6.25 ft each
Compressor for slug return
air lift pump 2.5 cfm @ 25 psi
Operation Conditions
Aeration Chamber -
Retention time 18 hr. * 120 hr. (if flow > 2,250 gpd dual system
will be in operation)
Clarifier -
Retent
Overflow rate 6.5 * 480 gpd/ft
Retention time 1.1 hr. * 7.9 hr.
2
341
-------
AEROBIC
POND
WIER BOX
CLARIFIER
BLOWER
' AIR LINE
1-1/2*
•p* I 1-1/2^-S^;
r~i
M COMPRESSOR
\ C^--''""""^"'-^^
f/" RETURN SLUDGE
/A'ERATION LINE -^
uu
PLASTIC SEPARATION PANEL
1 AERATION LINE
\ \ ~ " -*-
\ \
V\ \ RETURN SLUDGE
/
/ /
4' INCOMING I
SEWAGE FROM
PUMP STATION
CO
-P>
ro
CLARIFIER
AERATION CHAMBER
EFFLUENT*
I'll
COMPRESSOR
RETURN SLUDGE
«^ta
J
V
1
•hn^tB
r\ AERATION LINE
••^
O
.-, 15',,,
AlR 'LINE
.LEVELING
BAFFLE
AERATION CHAMBER
CLARIFIER
-------
WATER QUALITY MONITORING PROGRAM
Eight sampling sites were originally chosen to characterize the different
stages of the treatment process. In addition, two different treatment
schemes were planned involving the aerobic ponds. Both the sampling locations
and treatment schemes are shown in Figure 5.
Sampling Stations Location
a Incoming raw wastewater
b Aeration tank
c Return sludge and excess sludge pipe
d Effluent of extended aeration process
e Aerobic Pond No. 1
f Aerobic Pond No. 2
g Effluent of Aerobic Pond No. 1
h Effluent of Aerobic Pond No. 2
All chemical and biological sampling, preservation, and tests were analyzed
in accordance with Standard Methods**) and EPA Manual of_ Methods^). Field
tests included flow, pH, DO and temperature. Laboratory analysis included
BOD and COD (both total and soluble), settable solids, SS, VSS, TS, grease
and oil, TKN, alkalinity, turbidity, ortho-P, total-P, MLSS, and MLVSS
for the most part. Table 6 summarizes the daily sampling and analysis.
Throughout most of the study, sampling days averaged 3 days every two weeks.
Due to difficulties in obtaining fish to process during parts of the study
(see Table 3) the monitoring program was extended and is now scheduled for
completion during summer, 1977. Currently, monitoring of the pilot plant
continues.
RESULTS AND DISCUSSION
The aeration chamber of the full scale plant was seeded with 1,000 gallons
of 4,500 mg/1 returned activated sludge from a large municipal treatment
plant at the project initiation. Water meters were installed for continuous
recording of water consumption during fish processing. Wastewater from fish
processing was pumped to the treatment facility after fine screening, and
sanitary wastes were conveyed to a septic tank for treatment. Based on
pumping duration and rate, the flow entering the extended aeration system
was computed. A weir box was installed at the aerobic pond to record the
flow rate of effluent from the extended aeration system.
Processing of small yearling salmon and large salmon account for most
operations throughout the year and generated most of the wastes entering the
treatment system. Small salmon herin refers to an average weight of 11 ounces
per fish and large salmon refers to an average weight of 10 pounds per fish.
Large salmon are generally processed from August through July. During
periods, such as January, when no processing wastewater is generated, the
extended aeration system was maintained by feeding approximately 9.5 pounds
of Purina Trout Chow dissolved in 560 gallons of water into the pump station.
343
-------
•f-
RAW
WASTE
EXTENDED AERATION
S*/
RAW
WASTE
T
EXTENDED AERATION
SCHEME A
AEROBIC
POND NO. 1
SCHEME B
AEROBIC
POND NO. 2
FIGURE 5 PROCESS LAYOUTS
344
-------
TABLE 6. DAILY SAMPLING SCHEDULE
Sampling
Stations
Sampling Frequency
Tests
c
d
e
f
g
h
Grab Samples/2 hour
Eight-hour Composite
One Grab Sample Daily
Same as (a)
One Grab Sample Daily
Same as (e)
Same as (d)
Same as (d)
Flow, Temp, pH, DO
BOD, COD, SS, VSS, TS, Grease/
Oil*, TKN, Ortho-P, Total P,
Alkalinity, Turbidity
MLSS, MLVSS, DO, Alk., pH, Temp.
Same as (a) (Flow measurement
is not necessary)
DO, pH, Temp.
Same as (e)
Same as (d)
Same as (d)
*0ne grab sample for sampling day.
345
-------
Monitoring during fish food addition indicated wastewater characteristics
comparable to those during salmon processing, and the performance of the
treatment system was maintained when receiving no fresh fish processing
wastewater.
Wastewater characteristics for large and small salmon are summarized and
compared in Table 7. Approximately 20 percent of the total weight of fish
processed was wasted. The wastage was hauled away to a sanitary landfill for
disposal. Comparison of the wastewater characteristics showed that flow and
the weight of pollutants generated per ton of fish processed for small salmon
were greater than for large salmon. Conversely, pollutant concentrations
generated during small fish processing were generally less than during large
fish processing, which produced a stronger waste. As indicated by the
standard deviations, the incoming wastewater in both cases was highly
variable.
Also shown in Table 7 are the wastewater characteristics for large salmon
processing as influent to the pilot plant. To date, no small salmon proces-
sing wastewater has been treated by the pilot plant. This is scheduled
for the next phase of the study. Wastewater to the pilot plant was
even more concentrated than for previous large salmon processing, which
averaged 579 and 723 gallons of wastewater per ton of fish processed for
the pilot and full-scale plants respectively. As a result, pollutant
concentrations were greater for the pilot plant. Loadings were intermediate
between those for large and small fish at the full-scale plant.
The processing plant receives returning adult chinook, chum and coho salmon.
Generally chinook are the largest and coho the smallest in size. The 1976
adult returns contained a larger proportion of the smaller fish than for
1975. In addition, wastewater in the storage bins, which in 1975 was washed
into the septic tank, was released to the extended aeration system in 1976.
These two phenomena may account for higher pollutant loadings during 1976
for large fish processing.
Comparison of Table 7 to Table 1 indicates pollutant concentrations from the
Skokomish processing plant were generally higher than for those surveyed,
although on the average the gallons of wastewater generated per ton of fish
processed were about equal (1,210 gal/ton for those surveyed versus 1,223
gal/ton average for the Skokomish plant). Loadings were similiar for large
fish processing, but higher for small fish processing than those surveyed.
Operating conditions for both plants are summarized in Table 8. Because the
projected numbers of fish were not realized, retention times and DO concentra-
tions were high, but F/M (food-to-microorganism) ratios were within the
range of 0.05 to 0.2 recommended for the extended aeration process. (6)
Processing times were about one-third the expected. The higher flows
associated with processing small fish reduced the retention time to seven
days in the full-scale plant. The change-over to the pilot plant occurred
coincident with a switch to processing large salmon, and the consequent
reduction in flows resulted in a retention time of five days. During
the last phase of the project, wastewater from small salmon processing will
be treated in the pilot plant, and sufficient flows are expected to test
retention times of fractions of one day.
346
-------
TABLE 7. WASTEWATER CHARACTERISTICS FOR THE SKOKOHISH SALMON PROCESSING PLANT
1976
Pilot Plant
oo
Parameter
Flow, gpd
Process Tine, hr/day
Fish Processed, Ib/day
Processing Wastage, Ib/day
Gallon of Wastevater per
Ton of Fish Processed
Turbidity, JTU
PH
DO, mg/1
Temp., degrees F.
Alk. , mg/1 as CaC03
BOD (total), mg/1
BOD (total), Ib/ton
COD (total), ng/1
COD (total), Ib/ton
BOD/COD (total)
SS. mg/1
SS, Ib/ton
VSS, Bg/1
VSS, Ib/ton
VSS/SS
TS, mg/1
TS, Ib/ton
Grease and Oil, mg/1
Grease and Oil, Ib/ton
TKN, mg/1
TKN, Ib/ton
NH4 •*• N, mg/1
NH4 + N, Ib/ton
Total-P, mg/1
Total-P, Ib/ton
Ortho-P, mg/1
Ortho-P, Ib/ton
Large Salmon
Standard
Average Deviation
691
6.5
3,153
747
723
100
7.2
7.9
51.5
128
687
3.26
2,057
10.5
0.33
502
2.5
265
1.3
0.541
1,638
7.5
283
1.5
207
0.86
6.04
0.04
0.26
0.0020
0.11
0.0007
237
1.0
2,658
614
599
49
0.1
1.0
3.6
88
445
2.50
1,120
8.6
0.10
224
1.4
147
0.9
0.175
861
2.8
142
0.9
197
0.49
7.25
0.04
0.16
0.0027
0.05
0.0007
Small
Average
3,143
5.4
2,751
570
2,366
60.8
7.1
7.3
52.5
49
665
10.36
902
13.9
0.70
320
4.4
225
2.6
0.703
1,029
15.8
859
4.3
73
1.08
4.67
0.30
0.56
0.0085
0.49
0.0052
Salmon
Standard
Deviation
1,001
1.3
812
170
764
44.2
0.2
1.7
2.3
7
273
5.29
398
6.4
0.12
479
3.3
405
2.0
0.192
650
8.2
372
2.2
50
0.78
3.20
0.32
0.35
0.0065
0.23
0.0036
Large Salmon
Standard
Average Deviation
601
4.9
4,003
807
579
145
7.1
6.1
56.7
94
1,664
5.68
2,009
7.9
0.82
760
3.6
607
2.9
0.810
1,679
7.6
603
3.7
52
0.20
7.68
0.04
0.67
0.0065
0.51
0.0059
308
11.39
3,561
700
603
60.8
0.2
2.3
5.4
8
650
4.44
780
6.6
0.11
342
4.2
247
3.3
0.065
508
8.2
269
5.4
24
0.15
1.89
0.05
0.44
0.0092
0.34
0.0042
-------
00
TABLE 8. OPERATING CONDITIONS FOR THE EXTENDED AERATION FACILITIES
Full Scale Plant
Parameter
Retention Time (days)
Processing Time (hrs/day)
PH
DO (mg/1)
SVI
F/M
MLVSS
MLSS
MLVSS/MLSS
Large
Average
31
6.5
7.3
8.3
151
0.06
516
978
53%
Salmon
Standard
Deviation
8
1.0
1.0
1.4
24
0.06
174
218
Small
Average
7
5.4
7.1
5.7
129
0.05
1,015
2,240
45%
Salmon
Standard
Deviation
2
1.3
0.4
1.5
28
0.04
433
666
Pilot
Large
Average
6
4.9
7.1
2.0
153
0.21
2,011
2,498
83%
Plant
Salmon
Standard
Deviation
5
1.4
0.4
1.3
30
0.16
469
771
•
Overflow Rate (gal/process-
ing period/square foot
38
22
173
105
243
99
-------
Even though the flow rate through the aeration chamber was low, the clarifier
for the new system experienced some bulking due to undersizing. This problem
has been corrected. No samples were taken when bulking was evident.
The percentage of MLVSS to MLSS responded to organic loading. At the full-
scale plant, the volumetric loading was 2 Ib BOD 71,000 cu ft and 7 Ib BOD_/
1,000 cu ft on the average for large and small salmon respectively. This
increased to 30 Ib BOD 71,000 cu ft for the pilot plant. Both the full-scale
and pilot plants exceeded the recommended range of 10 to 25 Ib BOD /I, 000 cu
ft for extended aeration. (6)
Table 9 summarizes the extended aeration treatment efficiencies for the
processing wastewater. Comparing large and small fish at the full-scale
treatment facility, removal efficiencies for most parameters were not
appreciably different except for nutrients. Ammonia and TKN removal effi-
ciencies were twice as high for the large fish which was probably due to the
higher incoming concentrations (see Table 7) and longer retention times
causing some denitrification. Conversely, the removal efficiencies of TP are
about 50 percent lower for large fish and may have been due to over aeration
(longer retention time) causing the release of phosphate in the aeration
chamber.
Comparing large fish processed at the full-scale and pilot plants, removal
efficiencies for BOD, COD, TKN, and ammonia were similar. However, the
removal efficiencies for solids, grease/oil and phosphorus increased
dramatically. Reasons for increased removal of phosphorus are discussed
above. Improvements in removal efficiencies for solids and grease/oil are
probably due to higher incoming concentrations and shorter retention times.
During full-scale plant operations effluent turbidity was approximately
30 percent higher even though the clarifier overflow rate was lower, and a
yellowish-brown color was noticeable which disappeared with change-over to
the pilot plant. It is suspected that the longer retention times at high
DO concentrations favored filamentous growths in the full-scale plant.
Apparently, filamentous microorganisms which have poor settling character-
istics are able to utilize slowly the inert polysaccharide material produced
by the bacteria, giving the filamentous forms a source of food that is
unavailable to bacteria. (8) However, this is inconsistent with the sludge
volume indexes (SVI) which showed the pilot plant to have essentially
identical sludge settling characteristics to the full-scale plant during
large fish processing and a higher SVI during small fish processing (Table
8). This problem will be further investigated during the next phase of
the study.
Because of the low flows through the extended aeration system and the high
permeability of the soils in the aerobic ponds, no effluent has been
discharged, or analyzed, from the ponds. Therefore, the two treatment schemes
of the ponds in parallel or succession could not be tested.
Oxygen uptake for mixed liquor suspended solids (MLSS) was tested in the
field using a YSI DO probe. Only data from the full-scale plant is currently
available and is summarized as follows.
Average 1.59 mg/l/hr - Standard Deviation 0.55 mg/l/hr - Range 0.8-2.5 mg/l/hr
349
-------
TABLE 9. EXTENDED AERATION TREATMENT EFFICIENCIES FOR
SALMON PROCESSING WASTEWATER
Full Scale Plant
Parameter
BOD, Z
COD, Z
SS, Z
o VSS, Z
TS, Z
Grease and Oil, Z
TKN, Z
NH, ± N, Z
Total-P, Z
Ortho-P, Z
Large
Average
91
88
62
66
45
69
83
72
28
37
Salmon
Standard
Deviation
8
5
18
20
19
19
16
24
3
25
Small
Average
91
82
52
47
46
55
46
37
58
N/A**
Salmon
Standard
Deviation
5
14
32
36
36
20
31
29
43
N/A**
Pilot
Large
Average
93
89
93
93
65
82
89
68
43
53
Plant
Salmon
Standard
Deviation
5
8
3
3
14
17
7
13
24
26
**N/A indicated that effluent concentrations were either greater than or equal to influent,
-------
Additional data will be collected during the next phase of the study to be
used in developing design criteria.
To date, no sludge has been wasted. A sludge filterability test was conducted
to indicate sludge dewatering characteristics. One-hundred milliliters of
sludge was filtered under a pressure of 21 psi for 80 minutes, and solids
concentrations were measured at 0, 20, 40, 60, and 80 minutes. These results
are compared in Table 10. Percentage of SS increase were similar for both
large and small fish processing at the full-scale plant and slightly higher
for the pilot plant.
Table 11 compares the extended aeration effluent quality with EPA effluent
limitations. In all cases the average and maximum extended aeration effluent
loadings were below the EPA limits. The lowest loadings resulted from the
pilot plant operation. Small salmon processing effluent had the highest
loadings as would be expected from the high influent loadings.
As stated previously, this study is still underway. The final phase is
expected to be completed by mid-summer, 1977 and the final, report should be
published shortly thereafter.
CONCLUSIONS
Wastewater characteristics for salmon processing showed flow and pollutants
generated per ton of fish processed for small salmon to be greater than for
large salmon; conversely, waste strength was higher for large salmon. Incoming
wastewater flows and concentrations were highly variable in both cases.
Pollutant loadings from the Skokomish plant were similar for large fish
processing, but generally higher for small fish processing than for others
surveyed in the Northwest and Alaska.
Unexpected low flows resulted in long retention times and over aeration in the
full-scale plant. The pilot plant achieved much lower retention times,
(5 days versus 31 days), lower DO (2 mg/1 versus 8 mg/1), and higher F/M
ratios (0.2 versus 0.06).
Removal efficiencies for small fish processing were similar to those for large
salmon processing except for nutrients. Higher removal efficiencies for
nitrogen and lower removal efficiencies for phosphorus during large fish
processing were the result of longer retention times and over aeration which
may have caused some denitrification and phosphorus release in the aeration
chamber.
Pilot plant removal efficiencies were similar for BOD, COD, TKN and ammonia,
but higher for solids, grease/oil, and phosphorus. Shorter retention time and
reduced aeration is the suspected cause, but more data is needed to confirm
this result.
During long periods of processing plant shutdown, addition of fish food was
found to maintain performance of the wastewater treatment system.
The treatment system is capable of producing an effluent which meets the EPA
effluent limitations in terms of BOD, SS and grease/oil.
351
-------
TABLE 10. COMPARISON
OF SLUDGE FILTERABILITY FOR THE
SKOKOMISH
EXTENDED
AERATION FACILITY
CO
en
ro
*
Skokomish Study
Full Scale Plant
Large Fish
Small Fish
Pressure
(psi)
21
21
Initial
Concentration
mg/1
1370
2730
Time
(min)
20
20
Percentage
Increase
122
124
of SS
Range
116-126
116-135
Pilot Plant
Large fish
21
2117
20
134
123-150
-------
TABLE 11. COMPARISON OF EXTENDED AERATION EFFLUENT QUALITY WITH
EPA EFFLUENT LIMITATIONS
co
en
CO
Average Weight of
Fish Processed
(ton/day)
EPA Effluent*
Limitations
(Ib/day)
Extended Aeration
Effluent Quality
(Ib/day)
Average Range
Full Scale Plant
Large Salmon 1.58
BOD, Ib/day
SS, Ib/day
Grease/Oil, Ib/day
Small Salmon 1.38
BOD, Ib/day
SS, Ib/day
Grease/Oil, Ib/day
Pilot Plant
Large Salmon 2.00
BOD, Ib/day
SS, Ib/day
Grease/Oil, Ib/day
10.1
6.3
15.5
8.8
5.5
13.5
12.8
8.0
19.6
0.4
1.5
0.7
1.3
1.7
2.3
0.5
0.3
0.4
0.03-1.3
0.3 -3.0
0.1 -1.2
0.04-3.5
0.48-2.7
0.47-10.0
0.2-1.3
0.1-0.9
0.1-1.3
*Baaed on maximum 30-day average (see Table 2)
-------
ACKNOWLEDGEMENTS
This project was sponsored by the Environmental Protection Agency, under
Grant No. 803911, and the Skokomish Indian Tribal Council, Shelton,
Washington.
The authors wish to express their appreciation to the following individuals
whose contributions made this project possible.
Ken Dostal EPA Project Officer
Vic Martino Skokomish Tribal Council
Frank Klobertanz Kramer, Chin & Mayo, Inc.
Donna Snow Kramer, Chin & Mayo, Inc.
REFERENCES
i
1. LIN, DR. S. S. and DR. P. B. LIAO. Evaluation of an extended aeration
process for salmon processing wastewater treatment. Presented at the
PNPCA Industrial Waste Conference, Seattle, Washington, October 28,
1976.
2. ENVIRONMENTAL ASSOCIATES, INC. FOR U.S. EPA. Draft development
document for effluent limitations guidelines and standards of
performance - canned and preserved fish and seafoods processing
industry. February, 1974.
3. RIDDLE, M. J. et al. An effluent study of a fresh water fish process-
ing plant. Water Pollution Control Directorate Reprint EPT G-WP-721,
Canada, 1972.
4. STANDARD METHODS FOR THE EXAMINATION OF WATER AND WASTEWATER.
American Public Health Association, 13th Ed., 1971.
5. MANUAL OF METHODS FOR CHEMICAL ANALYSIS OF WATER AND WASTES. U.S. EPA,
Office of Technology Transfer, Washington, D.C., 1974.
6. METCALF & EDDY. Wastewater Engineering; Collection, Treatment,
Disposal. McGraw-Hill Series in Water and Environmental Engineering,
1972.
7. SEKIKAWA, Y., et al. Release of solubale ortho-phosphate in the
activated sludge process. Kurita Central Laboratories, Yokohama,
Japan.
8. McKINNEY, ROSS E. Microbiology for Sanitary Engineers. McGraw-Hill
Book Company, Inc., New York, 1962.
354
-------
FUNGAL CONVERSION OF
CARBOHYDRATE WASTES TO ANIMAL FEED
PROTEIN-VITAMIN SUPPLEMENTS
by
Brooks D. Church*, Charles M. Widraer**, and Rudolpho Espinosa***
The purpose of these studies was to establish an inexpensive, low technology,
continuous industrial method for the fungal conversion of corn wet-^nilling
wastes to an animal feed proteln-^vltamln supplement and, concomitantly, to
affect a significant BODs reduction of the industrial plant processing
wastes.
Sukhatme's (1) data suggests that while protein is not limiting in some of
the world's population provided enough calories are available to permit
efficient utilization of the protein, It Is also clear that many regions are
acutely deficient In protein but have abundant supplies of carbohydrate.
The developed countries also have abundant supplies of carbohydrate from
agricultural and industrial food processing amounting to approximately 400
million tons per year in the U,S. alone (2),
Although the studies described here were designed specifically for the bio
conversion of corn wet-milling wastes to animal feed protein, the impli-
cations of the work are broader than that however, since the process has
been extended to other waste substrates, has excellent industrial waste
pre-treatment utility, and the product might ultimately find use as a
human food.
BACKGROUND
The industrial cereal grain wastes used In all these studies were generated
from Penick and Ford, Ltd., a subsidiary of Univar Co.. Penick and Ford
is a corn wet miller producing corn dextrose syrup and starches for a variety
of industries. The corn waste liquid streams arise as a result of steeping,
washing, grinding, fractlonation, and product purification. The different
waste constituents used as fermentation substrates In these studies included
corn syrup, starches, gluten, and steep liquor separately and mixed in
various proportions. The total mixed liquor waste stream from this plant
is approximately 3 x 10 gallons per day containing 0.3 to 0,7 percent dry
solids.
*Department Biological Sciences, University of Denver, Denver, Colorado
**Penick and Ford, Ltd,, Cedar Rapids, Iowa
***Applied Research Division, Central American Research Institute for
Industry (I.CAITI), Guatemala, CA,
355
-------
On the premise that there Is and will continue to be a world protein shortage,
many university and industrial research teams have put forth a large effort
exploring the potential of single cell protein (SCP) as a novel protein
source. In the course of this work, a wide variety of other carbon-energy
sources also have been considered for use in protein production. The cost
for the carbon source represents a major fraction of the protein production
cost and may account for 40 to 50 percent (3). Utilities may represent
typically 15 to 20% mostly as cost of power for oxygen transfer and heat
removal. Although much attention has been given to hydrocarbon conversion
to protein, particularly n-alkanes and more recently methanol; we believe
waste carbohydrates of zero to a minus cost value emanating from the agro-
industrial food areas represent the most ideal substrates. Ideal substrates
for bioconversion because industry must otherwise allocate funds for their
removal and because these carbohydrates are rapidly metabolized by a variety
of high protein producing microorganisms. As a consequence of all these
SCP studies, a number of large SCP plants are presently in operation and
several more under construction (4). Therefore, we are no longer talking
about a protein source of the future, but rather a protein source in use
today (5).
Fungi used in bioconversion of carbohydrate waste substrates are the vegeta-
tive mycelium usually of species of Pencillium, Aspergillus, Trichoderma,
and Gliocladium. The fungi used in these and other studies grow over a wide
range of pH (2.5 to 5.0), and are rich in the B-group of vitamins and contain
40 to 55 percent protein. They contain low nucleic acid levels of approxi-
mately 4 to 6 percent and have a protein conversion efficiency from a given
quantity of carbohydrate of about SO percent which is far superior to the
4, 5, 15, and 20 percent efficiencies for protein production by beef,
poultry, milk and pork respectively. Other advantages the fungi have
compared to bacteria, algae, and yeast are: simplicity of process ferm-
entation, utilization of zero to minus value waste substrates for growth and
nutrition, flexibility of the fungal biomass to maintain its dominant
cultural position in the face of other microbial contaminants and process
malfunctions, its generally favorable process economics for protein pro-
duction, and lack of toxicity in numerous animal feeding trials (6,7,8,9).
Therefore, the use of fungal microorganisms to relieve the global protein
shortage can be accomplished rather safely and economically. The major
problems remaining are probably psychological and adequate funding for
conducting large scale feeding trials to meet quality standards.
Industrial waste treatment studies employing a variety of fungal strains
have been conducted over the past 5 to 8 years (5,6,7, and 12). In general,
the results have demonstrated an 80 to 95 percent reduction of the plant
processing BOD5 waste. Short retention times of 6 to 8 hours have been
experienced where the waste substrates were chiefly sugars (glucose, lactose,
and sucrose) and longer times of 10 to 15 hours with starches. A very
short retention time of 4.5 to 5 hours is realized by the Finnish workers
(7) using waste streams from the pulp and paper industries. They anticipate
a protein production of approximately 10-15 thousand tons a year from their
"Pekilo" process. These are all continuous fermentation processes and
several have been in operation for over four years.
356
-------
METHODS
Laboratory Studies
Screening and selection of the most desirable fungal strains were based on
two criteria: a high fungal protein content and the fungal ability to
rapidly metabolize the raw waste substrates. In the case of corn wet
milling wastes, described here, two fungi best met these criteria:
Trichoderma viride, 1-23 and Gliocladium deliquescens, 1-31. The two
fungal strains were maintained on potato-dextrose agar slants. At time of
experimental start-up, a slant was used to inoculate 50 ml. sterile
potato dextrose broth. After 24 hours, 5 ml. of the potato dextrose fungal
culture was used to inoculate the 50 ml. sterile corn waste medium (2nd
transfer). After another 24 hours, 50 ml. from the 2nd transfer culture was
used to inoculate 1000 ml. of non-sterile corn waste medium. All culture
transfers were made in shake flasks and incubated at room temperature on
a mechanical shaker. The final or 3rd transfer culture (1000 ml.) was
added to the 15 liter fermentor which contained 5 1 of waste substrate and
10 1 of water. The design and operation of the fermentation process was
such as to maintain the fungal biomass in the vegetative mycelial state at
all times. Certain deviations from this fungal morphocytogenetic state,
e.q. sporulation, indicated plant process malfunction. The altered fungal
forms often constituted valuable indicators of fermentation malfunction
which called for specific equipment and nutrient adjustment. Morphogenetic
changes in the fungal biomass were determined by frequent rapid phase micro-
scopic observations of the effluent material.
The waste carbohydrate substrates used in the laboratory studies were
composed of 66 percent corn starch, 33 percent corn dextrose, and 1 percent
heavy corn steep liquor (51% solids). This same waste substrate ratio was
not always maintained in the pilot and full scale processes. When substrate
alterations occurred in the plant waste effluent stream (influent to the
waste treatment process), adjustments to restore optimal nutrient conditions
were quickly made via nutrient additions or deletions usually in the form
of corn steep liquor, urea, and inorganic phosphate. Chemical analyses
were made on frequent grab and composite samples from both the influent
waste substrate and the effluent fermentation liquor. No sterilization of
the waste substrates was ever conducted in the laboratory, pilot, or full
scale fermentations after the second stage of inoculum build-up.
Fermentation design and equipment used in the laboratory studies, were
described previously (5) and are shown here in Figure 1. This apparatus
evolved from the author's original prototype in 1968 (11) to what now
appears in Figure 1. It has proved to be extremely flexible for use both
as a batch and as a continuous type fermentor. Later fermentor modifica-
tions include a battery of 4 fermentor units which can be operated individ-
ually, jointly, or in series. The design is such that only minimal amounts
of fungal solids adhere to the internal surfaces. Sparged air introduced
at the fermentor bottom, serves the dual prupose of providing oxygen as
well as continuous.mixing of the fermentation materials. The influent
substrates drip onto the top surface of the mixed fermentation liquor
357
-------
PH METER
co
en
oo
CONTINUOUS
HEAD
TOR
FEED
STIRRER
FEED .
CONTROL
\FEED
RETURN
FEED
FEED
RESEVOIR
PH
SWITCH
SAMPLE
PORT
ACID
SOLENOID
ACID
V
ROTOMETER
ffl GAUGE
0b<>-3>—i
L>
ACID
PUMP
PRESSURE
VALVE
0
FERMENTOR
COMPRESSOR
EFFLUENT
Figure 1. Laboratory Continuous Fermentation Syst'etn.
-------
thus allowing removal of the influent line as an internal fungal adhering
surface which was present in our earlier fermentors. At the same time the
effluent port was relocated to the bottom of the fermentor for continuous
effluent discharge or for intermittent sample taking. This modification
avoided any chance of "short circuiting" undigested waste substrates as
was the case where effluent takeoff was located at the top surface level of
the fermentor (11). In addition, the bottom effluent port allowed instal-
lation of a flexible effluent conduit which served, by simply raising or
lowering the conduit, to vary the fermentor volume and thus adjust the dilu-
tion rate and retention time without adjusting the feed flow rate. The most
recent modification to the system, not shown here, was to admit the air up
through the bottom port, This releocation of the air line removed the last
remaining permanent internal fungal adhering tube surface. Many of the
15 liter laboratory fermentor features were incorporated into the 50,000gal.
pilot and the 3 -x lOgal.full scale plants. Fermentor process control
parameters e.q. pH, temperature, flow rate of waste substrate influent,
oxygen, and added nutrient were recorded continuously onto standard recorder
charts.
Pilot Plant Process
Individual processing steps of the continuous pilot fungal treatment process
are shown in Figure 2. The total mixed plant waste stream was split to
supply the fraction required for the pilot operation and flowed into a
polyethylene lined concrete rectangular (74 ft. x 14 ft. x 10 ft) 190 m3
tank at a flow rate of approximately 35 gallon/min. A minimal volume of
hydrochloric acid was added to maintain the proper pH of 4.2 to 4.5.
Nutrients such as heavy corn steep liquor (51% solids diluted 3-fold to
facilitate pumping), urea, and sodium dihydrogen phosphate were added as
required to maintain the desired nitrogen and phosphate concentrations.
Air was supplied by two 40 H.P.,.800 ft. /min. Roots-Connerville blowers.
Air was blown into six-inch pipe headers which were connected through two-
inch pipe leads to 80 one-inch rubber hoses. The hoses dropped into the
fermentation tank and were connected to 80 Dravo flutter valve aerators
located approximately one-foot above the tank bottom. These aerators were
spaced in such a manner that one aerator supplied air to approximately a
four ft2 bottom area and up through a depth of six to eight feet. The
dilution rate was controlled by adjusting either the influent waste feed
flow rate or (as was most frequently the case) adjusting the tank volume
by raising or lowering the pilot effluent flexible conduit - see Figure 2.
Various procedures were used to harvest the fungal solids. The most
successful was to continuously exit the fermentor effluent containing the
fungal floes into a 2000 gallon cone clarifier. Proper positioning of the
fermentor effluent line in the clarifier allowed discharge and rapid settling
of the fungal floes to the clarifier bottom while lighter fungal fragments
and undigested waste particulates floated out at the clarifier top. Proper
positioning of the fermentor discharge (X) and the clarifier baffles (B)
for maximum floe collection efficiency are shown in Figure 2. The total
residence time in the clarifier was approximately 1 hour and a solids
(fungal) concentration from the clarifier underflow was 3 to 4%. The
359
-------
CO
en
o
ACID NUTRIENT
IP04= 1 I CSL 1
^ ^ k^ S METER METER
PLANT L
WASTEWATER NO.
AIR
J pH DO. ^FLEXIBLE CONDUIT
^ 1 /
^ /
f^ lf-^ ,,-, rf>__/%i i<~* _j~* *% ^~^^»^*>^ > __ n~ n ii*i ,j~»-rj
U u 3
AEROBIC TANK 1
V X J
CONE \ /1B
CLARIFIER\ /
VIBRATING h-J— I
SCREEN LJ
TREATED WASTEWATER
i
t l( )>-r-^
•>^ s <*&
VACUUM "FILTER
SOLIDS
RECOVER'
Figure 2> FLOW DIAGRAM OF FUNGAL PILOT PROCESS.
-------
vibratory screen, shown in Figure 2, was found to be an unnecessary harvesting
step in subsequent studies and the clarifier underflow was discharged
directly onto the Amatec continuous vacuum belt filter. A fungal cake of
3/8 to 1/2 inch thickness and 20 to 22% solids was continuously discharged
from the vacuum screen as shown in Figure 3.
Full-Scale Process
4,
A full-scale, 3 x lOgal.fungal treatment plant was constructed following the
pilot studies at Penick and Ford. The fermentation facility consists of
two L.5x1 Og^l.circular, 40 ft. high by 80 ft. diameter, coal tar resin
lined steel plate tanks. Air is supplied by Worthington positive pressure,
two stage, intermittent cooling blowers. The air is blown through header
pipes into the bottom of each tank. From there, the air is routed through
the tank bottom pipe distribution system before escaping up through 16 six
foot verticle Kenick spiral aerators per tank. Since the two fermentation
tanks are run in "series", the first (north) tank is always full and the
waste water flows from a pipe near the top of the first tank into the second
(south) tank. By varying the volume of the second (south) tank, one can
vary the total residence time of the system.
Two alterations were introduced into the fungal treatment process following
construction and operation of the full-scale plant. One change was to allow
the total treated plant effluent to discharge into a lime neutralizing tank
before entering the cone clarifiers. This alteration of the pH from 4.5
to 7.0 in the harvest system allowed the fungal solids to settle faster in
the cone clarifier than experienced in the pilot system and provided a pH
7.0 effluent discharge from the clarifier overflow to the city waste treat-
ment plant. The other alteration introduced into the system was to replace
the vacuum belt filter following the clarifier step with a DeLaval Basket
Centrifuge. This fungal harvesting change resulted in a more economical,
energy efficient, and higher dewatered solids product. Thus, the 20 to
22% dewatered solids from the vacuum filter was raised to 32% fungal solids
by this equipment change.
Drying of the 32% fungal solids was studied under a variety of conditions
and will be briefly described under RESULTS.
Start-up of all continuous fermentations in the laboratory, pilot, and
full scale processes was carried out according to the inoculation—dilution
technique described to the author by the late Dr, H. Orin Halvorson. In
accordance with this procedure, 1/3 the total fermentor volume was filled
with raw waste. The remaining 2/3 volume was filled with water. Inoculation
of the fermentor consisted of a 15 to 20% actively growing vegetative fungal
mycelium added with respect to the waste substrate volume not to the total
fermentor volume. Employing start-up procedure, continuous fermentations
were initiated at zero time. The inoculation-dilution relationship used in
these studies is shown in Figure 4 where the "theoretical" build-up of the
waste substrate, assuming perfect and instantaneous mixing and providing no
digestion of the waste substrate occurs, is determined by the integrated
formula shown in Figure 4. The "actual" waste substrate (COD) reduction
361
-------
Figure 3. Continuous Belt Vacuum Filter
Showing Discharge of Fungal Cake
362
-------
10,0001
9000
m
/
8000
7000
I
I
w
/
Thtortticol
COO
-In coming Fod COD
I
C 6000|
v.
e»
E
I
o
5000
4000
9
t>
o
o
o
II
S 30001
e
O
2000
1000
at t = 0, COD = 3330mg/l =C0
Cw - Feed COD = 9200 mg/l
V - Fermentor Volume = 10,500ml
Q = Fttd Rote = lOml/min
.29%
V— Actual
\ COD
.88%
94%
6
Days
Figure 4. START-UP OF CONTINUOUS FERMENTATION OF CORN WET-MILLING WASTE
IN A LABORATORY FERMENTER BY GLIOCLADIUM DELIQUESCENS
363
-------
shown in Figure 4 represents the effect of fungal growth and metabolism on
degradation of the corn wastes and IB indicated In terms of percent reduction
of the "actual." COD. Although only the first seven days data are plotted in
Figure 4, the continuous system operated for 14 days before termination
and showed a percent waste reduction of 90 to 95 with both T\ viride and
G.L deliquescens.
Analytical procedures used for determination of the effectiveness of fungal
fermentation of the waste substrates were chiefly those described in Methods
for Chemical Analysis of Water and Wastes (13). Influent and Effluent
"grab" samples were taken at 6 hr, intervals. 24 hr. composite samples
made-up from the 6-hr, "grab" samples were also analyzed. All samples
were filtered through Whatman #4 filter paper under slight negative pressure
and the filtrates and retentates dried on the tared filters and in tared
dishes overnight at 95°C. Weights from the tared filters were recorded as
fungal mass and from the tarred dishes as total solids for all effluent samples.
Influent samples were not filtered prior to drying except in instances where
suspended solids were measured. Retentate samples were also heated at
650°C for 2 hrs. for determination of volatile solida and ash. In the pilot
and full scale processes samples were never less than 1 gallon each.
During start-up and periods of operational breakdown samples were taken
more frequently than at the six hour intervals.
A routine procedure used in all these studies for evaluation of process well
being or malfunction was that of phase microscopy. Microscopic changes in
fungal morphology reflected gross changes in the fermentation process and
deviation from the steady-state. The method was simple, reliable, rapid,
and required minimal operator training. The method was enthusiastically
endorsed by the chemical engineers! Optimal fermentation conditions resulted
in a phase microscopic picture of fungal morphology that revealed a homo-
geneous cytoplasm, long tapering mycelial branches devoid of septa, and
somewhat swollen hyphae which occasionally contained an early developing
arthospore. Lysing and sporulating mycelium, on the other hand, indicated
nutrient starvation or a too low dilution rate. The microscopic technique
could also be used to verify a well-mixed fermentation system, or conversely,
pick out "dead" areas (poor nutrient mixing, oxygen starvation, etc.)
by employing a grid or patterned microscopic slide sampling examination of
the fermentation tank. Other information obtained from this technique included
estimates of the level of contaminating microorganisms and undigested waste
substrate particulates.
The techniques and methodologies used in these studies to evaluate the
fungal drying, and animal feeding quality of the fungal protein will be
included with the experimental procedures and results in the following
sections.
364
-------
RESULTS
Development of Continuous Fungal Growth and Waste Substrate Utilization
a. Laboratory Studies
Laboratory and pilot experiments were conducted under batch culture conditions
in the fermentation apparatus described under METHODS and the data were used
to determine fungal growth rates and substrate (waste) utilization optimums
with _T. yiride and (3. deliquescens. Attempts were made to define the
C:N:P (carbon to nitrogen to phosphate) ratio of the influent substrate
medium in the hope that such a ratio would indicate the optimal rate of waste
substrate utilization and the highest fungal mass yield. Such studies con-
ducted in batch culture pointed up the need for additional nitrogen and
phosphate above what was supplied by the raw waste substrate. Thus, it was
determined that a carbon (carbohydrate + organic acid) to nitrogen (protein
and ammonia nitrogen from corn steep, urea, and ammonium sulfate) to
phosphate (derived from the waste ash + added sodium dihydrogen phosphate)
ratio was approximately 50:10:1, This ratio is shown in the influent waste
column of Table 3 which also contains the data from a continuous culture
laboratory experiment where steady state balanced growth was maintained.
Initially, experiments were undertaken to generalize the given relations of
fungal growth mathematically in batch culture. Maximum growth rates
(M max) and COD reductions were measured in several media to evaluate fungal
growth and protein formation. Fungal growth at 26°C in an excellent
synthetic salts-urea-dextrose medium (Mandel's medium) was compared to a
corn wet-milling waste medium both of which contained excess nitrogen and
phosphate. One external physical condition, pH, was studied in the two
media. The results are shown in Table 1.
365
-------
TABLE 1. EFFECT OF PH AND MEDIA
ON THE GROWTH RATE OF FUNGI
Fungus
Trichoderma
viride
Gliocladium
deliquescens
pH
3.5
4.5
6.0
3.5
4.5
6.0
Handel's Salts
Urea + Dextrose
td
1+RS hrs.
4.2
3.6
5.1
8.4
8.1
9.1
y
hr-1
0.164
0.193
0.137
0.083
0.086
0.076
*Corn-Wet-Milling
1% Corn Steep
td
1+RS hrs.
11.1
8.6
10.3
9.6
10.0
10.4
M
hr-1
0.062
0.080
0.067
0.072
0.069
0.066
*Mixed corn starch, 66%; corn syrup, 33%; and heavy corn steep liquor
(51% solids), 1%. Sodium dihydrogen phosphate added, 100mg/l.
td = time to double fungal mass
U = specific growth rate, slope of exponential growth rate line.
Cultures grown at 26 C with excess oxygen.
The importance of the growth rate as a criterion of the state of the culture
as a whole, is well understood and is expressed here in the mathematical
relations derived by Monod (14). The data used here were derived alge-
braically by analysis of the fungal growth of the cultures in batch process
during the period of logarithmic growth. Growth results in the two media
at three pH levels are shown in Tabel 1. The fungal mass average doubling
times (td) were determined by calculation from the straight line part of
the growth curve in these media. The data were obtained from periodic
samples of the culture in which dry weights of the fungal mass (g/1), were
plotted vs. time (hrs.). The specific growth rates (y) were calculated and
also shown in Table 1. y is the slope of the exponential growth and a
meaningful indication of the change in the culture. Since any restriction
in the internal (carbon, nitrogen, phosphorous) or external (pH, temperature,
oxygen) environment is reflected in the specific growth rate of the fungus
operational deficiencies in a pilot plant situation can be detected quickly
In spite of the chemical differences in the two media and the pH range,
G. deliquescens showed less variability in doubling times and specific
growth rates than did T. viride. In all cases the optimal PH appeared to
be approximately 4.5 and Handel's medium superior to the corn waste. Since
there were no restrictions imposed on the cultures in these growth media
the specific growth rates should be considered maximum (y max).
Use of a limiting nutrient, in our case the phosphate concentration, allows
366
-------
use of Monod (14) equations where growth increase with respect to time is
proportional to the fungal mass present at that time, and we can calculate
the specific growth rate (y) of the fungal mass. The purpose of the kinetic
study in the batch culture system was to find both the maximum fungal mass
growth rate and the maximum waste substrate utilization in order to evaluate
the residence time and dilution rate for further design of a continuous
waste treatment process. Using the data plotted in Figure 5, the specific
rates within the linear or log phase of growth are defined as:
A) Specific growth rate of fungus is y
i ^ dlnM
and v « M* dt = dt at T*
where t • time
T* - 26 hour point
M « fungal mass
M* = fungal mass at T* - l.Og/1
S - substrate as COD, BODs, or TOC
B) Specific rate of substrate utilization is y'
!_ -dS dlnS
and y' = M* dt » dt at T*
- InSl
then yf - M* - t2 - ti
and assuming a first order reaction behavior
In 2 0.693
td or td ' = y or y ' = y or y ' = hours
where td * doubling time of fungal mass
td'=* % life time of substrate
The (3. deliquescens experiment shown in Figure 5 demonstrated termination of
growth and substrate utilization by the concentration of the growth limiting
nutrient, phosphate. In this experiment, following an initial lag period,
the substrate beg<.n to disappear and the rate of substrate utilization and
fungal synthesis were calculated between the 20 and 30 hour time periods.
During this time period, microscopic examination showed the fungal mycelium
to have numerous, homogeneous growing hyphal tips and no evidence of spore-
ulation. Calculations of the specific growth rate (y) and substrate h
life are shown below:
Fungal Mass:
Where T* = 26 hr. and M* •
dlnM 7.30 - 7.91
y = dt = 10 10 = 0.061 hrs-1
0.693
td ** 0.061 = 11.4 hours
367
-------
o -o
FUNGAL MASS • •
400
300
200
100
10
20 30 40 50
TIME (HOURS)
FIGURE 5. BATCH DIGESTION OF CORN WET-MILLING
WASTE BY GLIOCLADiUM DELIQUESCENS.
368
-------
For Substrate Utilization:
COD
I /-dlnS\ 1 A.7 - 3.0)
' - M*V dt > 1 V. 10 / = 0.
17 hrs-1
td
4.07 hours
TOG ± /dins ] 1 /0.55
U1 = M* (^ dt / = 1 \ 10
0.693
td' = 0.055 - 12.6 hours
0.055 hrs-1
Calculation of the waste substrate half-life can be misleading depending
on the carbon parameter one chooses for evaluation. As shown in Figure 5
and Table 2, the initial COD value is 3 times the TOC and the rate of
utilization of COD to TOC is 3 times faster. It would appear, therefore,
that £. deliquescens incorporates carbon into cell mass with a conversion
efficiency of approximately 33% if the COD is used and 100% if the TOC is
used. 100% conversion is obviously incorrect since it would not allow
for carbon oxidation for energy purposes required to synthesize cell mass.
On the basis of doubling times (td and td1), the COD and carbon measurements
would indicate that 37% of the carbon in converted to cell mass and 63%
to energy and heat. C02 evolution measurements during the course of this
experiment would have resolved the dichotomy. It would appear, from other
data not shown here, that the COD test better reflects the carbon fate
than does TOC. On the basis of COD, the data in Figure 5 and data from
other experiments summarized in Table 2 indicate that carbon is incorporated
into cell mass more efficiently by ^T. viride (61%) than by G. deliquescens
(36%).
TABLE 2. KINETICS OF GROWTH AND
SUBSTRATE UTILIZATION
Growth
Fungal
Strain
Specific
Rate
V
hr-1
Doubling
Time
td
hours
Substrate Utilization
Carbon
COD
TOC
Specific
Rate
V '
hr-1
h Life of
Substrate
td1
hours
T. viride
G. deliquescens
0.067
0.061
10.4
11.4
COD
COD
TOC
0.109
0.170
0.055
6.35
4.07
12.60
369
-------
It will be necessary for the investigator to preestablish which parameter
to use for carbon balance studies and to include a COz measurement before
the true % life of substrate utilization can be determined. This would be
critical to establishing optimal continuous fermentation and overall process
economics in the waste systems.
Continuous fermentation studies were initiated in the same apparatus used for
the previously described batch studies. The dilution rates of the contin-
uous systems were the same as the fungal growth rates found earlier in the
batch studies. Since (dM/dt = pM) for rate of the fungal mass growth and
since the rate of loss of fungal mass through continuous overflow also
equals (dM/dt) and can be stated, dM/dt = f/v - DM, where the flow rate
(f) is measured in culture volumes (V) per hour, then the expression
(f/VM) is called the dilution rate (D), thus:
yM - DM and y » D
which states that the fungal growth rate equals the fertnentor dilution
rate in a stabilized continuous system. The concentration of phosphate
was the limiting substrate nutrient to maintain the steady state; the waste
medium, pH 4.5, excess oxygen, and temperature were all set at the same
level as used in the batch studies.
The continuous fermentation was set-up as a batch system for the first 48
hours with I_. viride and fungal mass adjustment was believed to have occurred,
the system was switched to continuous. The residence time was set at 10.4
hours or a dilution rate of 0.067 hr-1 in accordance with the batch studies
(Table 2). As indicated in Figure 6, operational difficulties plagued the
continuous system from day 4 through 7. Resolution of these pH, feed line,
air pump, etc. problems, resulted in system adjusted to a steady state at
day 10 through 16. Nitrogen determinations made during the day 10 to 16
interval showed the system was being controlled by a limiting concentration
of nitrogen - phosphate was decreased but not limiting. Thus, nitrogen in
the form of urea was added at days 16, 19, and 23. Apparently both nitrogen
additions at days 16 and 19 relieved the nitrogen restriction and allowed
the controlling influence of limiting phosphate concentration to maintain
the steady state which continued for another two weeks before termination.
During the two week period following fungal adjustment to the second urea
addition, a number of measurements showed 90 to 92% removal of the influent
COD and fungal mass yield of 56%. These results would indicate that 36%
of the carbon was evolved as C02.
A similar continuous experiment to the one with T?. viride was conducted
with G. deliquescens where all conditions established to be optimal in
batch studies were employed in the continuous fermentation. Thus, the
dilution rate was set at the specific growth rate (y) found previously
to be 0.061 hr^1 or a residence time of 11.4 hours. Again a deficiency for
nitrogen was corrected by adding urea and the fungal mass density was
increased. Both these experiments with J. viride and G. deliquescens
demonstrated that substrate nitrogen enrichment increased fungal mass density
while not affecting the dilution rate or specific growth rate of the fungi.
Data from the study with G, deliquescens taken from samples analyzed over
a four day period following tungal adjustment after urea nitrogen addition,
370
-------
CO
COO FUNGAL MASS
9 r4.5
Figure 6. »
CONTINUOUS FERMENTATION OF CORN WET-MILLING WASTE
BY TRICHODERMA VIRIOE
10
COD •-
FUNGAL MASS
OPERATIONAL
PROBLEMS
BATCH -*^- CONTINUOUS
15
20
TIME (days)
-------
are shown in Table 3.
TABLE 3. CONTINUOUS DIGESTION OF
CORN WASTE BY GLIOCLADIUM DELIQUESCENS
Test
Condition
Total Solids
Total Volatiles
Total Ash
COD
BODS
Carbohydrate
Organic Acid
Protein -N
Ammonia -N
*Phosphate
Fungal Mass
Influent
Waste
mg/1
9640
6010
3630
6200
5600
4436
580
1030
88
17
68
Effluent
mg/1
3520
520
3000
545
128
310
50
42
2.7
1.0
3500
Percent
Reduction
%
63
92
18
91
98
93
92
96
97
91
"""
* This is only phosphate added to the raw waste medium, total phosphate
is not shown.
The conversion efficiency of carbon to fungal mass can be calculated from
a number of parameters shown in Table 3. Again, with this experiment as
was observed earlier in the J_. viride experiment, much depends on which
carbon parameter one chooses as to what the true values are for substrate
utilization or carbon to fungal mass conversion efficiency. Subtracting
the residual (Cr) carbon in the overflow effluent from the original influent
carbon (C) and dividing the fungal mass by the subtracted vali e shows that
conversion efficiencies for volatile solids, COD, and BODs are all 62 to
64%. While that from the combined carbohydrate -I- organic acids is 75%!
Since three of the four*parameters are in close agreement, one is persuaded
to accept any one of the three. Also, because all three include all the
organic constituents of the medium additional strength is given to these
parameters. On the same bases we believe the carbohydrate + organic acid
parameter might be excluded. A definitive decision cannot be made, however,
until more experiments are carried out and particularly experiments where
total C02 evolution is also measured.
b. Pilot Plant Studies
Corn wastes generated at the Penick and Ford corn wet-milling plant were
the same as those used in the laboratory studies. Whereas the concentration
of corn starch, gluten, and syrup wastes could be carefully controlled in the
372
-------
laboratory, it was much more difficult to do this in the pilot work due to
unforseen operational changes in the company processes. The operation
of the 50,000 gallon pilot facility was initiated by the inoculation-
dilution procedure described earlier and continued for 5 months. Results
of the first six and one-half weeks are shown in Figure 7. COD (mg/1) is
plotted vs. time in weeks and also the fungal mass (mg/1) is shown. Gaps
in the influent feed indicate the hours during which the system was switched
from continuous to batch operation during weekend plant shut-downs. The
dilution rate was 0.043 hr^1 corresponding to a residence time of 16.0
hours. Although this rate was somewhat slower than that achieved in the
laboratory studies, it was probably a result of fluctuating nitrogen and
phosphate concentrations. Either nitrogen or phosphate could control the
continuous fermentation as shown in the laboratory studies in Figure 6
by their limiting concentration effect. Since the raw waste influent concen-
tration was undergoing radical, almost daily change, it would appear that
there were very low nitrogen and phosphate levels at times when the influent
COD concentration became extremely high - for example, at the 3.0 and 5
to 6 week periods. A closer control of the nitrogen and phosphate concen-
trations would undoubtedly have improved the yield and allowed for an
increased dilution rate.
One particularly interesting feature of this treatment process shown in
Figure 7 was the manner in which raw waste influent concentration extremes
were reduced to a rather low, constant effluent level. The reason for this
is obviously the rapid response of an increased fungi mass when the substrate
level increased. The picture in Figure 7 is one of fungal mass density
adjustment to variable substrate concentrations at a constant dilution rate.
Although the fungal mass yield based on conversion efficiency of the substrate
was never near that obtained in the laboratory, the yield was quite variable
and ranged between 25 and 50%. On several occasions when the nitrogen
level was raised either by adding more corn steep or urea, an immediate
improvement in fungal yield was observed. This result was similar to the
seen in Figure 6 for the laboratory study.
Often the waste substrate concentration increase was not balanced with regard
to starch and the syrup dextrose. When the imbalance was due to high
dextrose concentration, excessive foaming occurred. Low fungal yields and
an increased fungal growth rate resulted in rapid depletion of the limiting
phosphate. Such an occurrence (recognized early by an increasing foam build-
up) could be controlled by increasing the dilution rate from 0.043 to 0.067
kr"1, When the foam subsided, the dilution rate was returned to the lower
value again,
Later experimental work with the pilot system demonstrated that where a carbon
to nitrpgen to phosphate ratio of 50;10;1 could be maintained in the face of
changing carbon (yav waste influent) concentrations, a reliable steady
state condition prevailed.
Drying the. 22% fungal solids from the vacuum filter or the 32% solids from the
373
-------
10. 000
9000
8000
7000
6000
X
Influent
X Effluent
Fungal Mass
2000
1500
0)
0)
nl
1000 §
(*<
500
0
WEEKS
Figure 7. COD Reduction of Corn Wet-Milling Waste by
Glioclad-itm deliquesoena
-------
centrifuge was carried out by several procedures. No washing of the fungal
solids was done prior to drying. The conditions used for drying the fungal
mycelium and the protein content after each drying procedure are shown in
The Flow Diagram, Figure 8. The protein content of each dried sample was
determined from amino acid analysis. Although not carried out in these
studies, we believe premtxing the vacuum filter or centrifuged solids cake
with other animal feed ingredients prior to drying would improve the protein
content of the biomass through protection. Under these conditions it would
appear that almost any drying procedure where the temperature was held at
70 c would not unduly compromise the protein. The higher drying temperatures
appeared to reduce the level of the essential amino acids; lysine, trypto-
phane, and methionine more than the others. Since methionine is already
deficient in this protein for nutritional quality of most animals, the
drying procedure used becomes a step of critical significance.
The amino acid levels of the J. viride and G. deliquescens protein has
been reported before (5, 10, 11) and both fungi demonstrated levels of the
essential amino acids, an leucine-iso-leucine ratio, low proline, and over
all amino acid balance comparable to soy protein. In addition to an excel-
lent amino acid balance, the fungi used in these studies had excellent
vitamin levels as shown in Table 4. The fungal vitamin level was obtained
from lyophilized pilot plant material. Vitamin analyses of some other food
materials are also shown for comparative purposes (15, 16).
TABLE 4. FUNGAL VITAMIN LEVELS
COMPARED TO OTHER FOOD MATERIALS
(VALUES EXPRESSED AS mg/kg. DRY WEIGHT)
Vitamin
Fungal Beef
Beef
Liver
Yeast Milk
Dry
Whole
Cereal
Thiamine 15
Riboflavine 160
Nicotinic Acid 575
Pantothenate 167
Pyridoxine 38
Cobalimine 10
Biotin 4
Folic Acid 18
Choline
D-Ergosterol 20
1-3
2
40-100
7-21
1-4
5-10
16
75-275
30-60
5
8
2-20
30-60
200-500
30-200
40-50
8
1-5
21
2000
3.4
15
7.3
20
1-3
4
0.3
0.02
862
1-7
1-2
10-30
5-20
3-6
2000
It was observed by Champagnat et. al. (15) that drying a product (in his
case, yeast) at low pH could lead to vitamin losses, particularly the
375
-------
Figure 8.
Flow Diagram of Fungal Treatment
of Com Wet Milling Waste
Pilot Plant Effluent Solids
0.16% Solids
Cone clarifier underflo
or
Vibrating Filter
#120 mesh screen
3.0% Solids
CO
-•vl
en
Vacuum Drum Filter
20-22* Solids
«*--"•
Fresh
-
227. Solids
-
•i'
48% Protein
*PF-0 —
" ^
Lyophiliced
-
947. Solids
1.5 Ibs.
•I
42% Protein
PF-1^ PF-2
»•*• /
Forced Air
115°C
95% Solids
7.5 Ibs.
J,
337. Protein
PF-3
I
Forced Air
80°C
93% Solids
5.5 Ibs.
\,
38.6% Protein
PF-4 PF-5
\ /
Vacuum
80°C
94% Solids
9.5 Ibs.
4-
38.5% Protein
PF-6
^
Nk
Drum Dried
o
75°C
96% Solids
5.8 Ibs.
Or
40% Protein
Patterson-Kelley
826C
96% Solids
3.0 Ibs.
•If
36.2% Protein
* » All proteins based on amino acid analysis (GLC)
t - Ibs. -= pounds on hand in storage at D.R.I.
x - Code designations - PF-1, PF-2, PF-3, etc., designate Penick and Ford fungal material harvested and
dried, 4/72
-------
water soluble vitamins. Our material was dried at pR 4.5, therefore some
of the vitamin levels shown in label 4 could be minimal values. Never-
theless, the fungal vitamin levels shown here are equal or higher than
those contained in the other food materials.
Animal Feeding Evaluations of the Safety and Quality of the Fungal Product
According to PAG Guidelines the physical, chemical, and animal feeding
properit±es of the industrial product should be essentially the same as
those of the experimentally tested material. To be truly significant,
the studies should be conducted on the fungal product as made on a produc-
tion scale rather than on laboratory batches. Also particular attention
must be directed to the composition of the fungal growth medium from the
viewpoint of the possible presence of chemical components regarded as
hazardous to health. In answer to these requirements and according to the
studies performed and reported in this paper, the source materials which
form the substrates for the growth of the nutritive fungal material are
the same basic substrates which are utilized for the various food products
manufactured by this company. In addition, the fungal product used in
these animal feeding studies was the material harvested and dried from
both the pilot and full-scale industrial treatment facilities reported
herin.
Gross chemical component analyses of the fungal product are shown in Table
5. Although these analyses are not definitive to show the possibility of
the presence of toxic contaminants derived from fungal metabolism or from the
source substrate materials (e.g. pesticides, lubricants, binders, heat
processing reactants, etc) it was assumed that the presence of any haz-
ardous feed materials in the fungal product would be detected in the course
of sensitive animal feeding trials.
TABLE 5. GROSS CHEMICAL COMPONENTS
OF PILOT PLANT HARVESTED FUNGAL PRODUCT
Chemical
Component
Protein
Carbohydrate
Nucleic Acid
Lipids
Ash
Moisture
Vitamins
Percent of
Dry Weight
45
30
5
4
6
10
0.1
The data shown in Tables 4 and 5 together with that on amino acids (5, 10,
377
-------
11) were used to formulate the animal diets for the feeding trials described
hereafter.
Animal feeding trials were conducted with weanling rates, chicks, mice,
nursing piglets, and gilts. Only the details of feeding trials in rats
and chicks will be reported here.
a. Weanling Rat Feeding Trials
Three types of feeding trials were conducted in weanling rats employing
fungal biomass as the protein supplement in the animal diets. These
were: 1) Growth response over a 3 week period
2) Protein Efficiency Ratio (PER)
3) Nitrogen Balance tests to determine the biological value (BV),
true protein digestibility (TPD), and net protein utilization
(NPU).
Three diets were formulated for these trials: one, a 3% egg protein
adjustment diet; two, a standard control casein diet; and three, a test
fungal diet in which the casein was replaced by fungal protein. After
a two week adjustment period the weanling rats were placed on the standard
and test diets. During the experimental three week feeding period In which
feed and water were fed ad libitum, animal weights, feed intakes, urine,
and fecal samples were recorded and collected for each rat each day. The
protein levels were set at 23% of the diet and rats were fed fungal protein
from each drying condition utilized in Figure 8. Growth rates of rats
fed the lyophilized fungal and casein control diets are shown in Figure 9.
The curves represent cumulative percent weight gains for the rats fed these
diets. The slope of weight gain was determined for each rat and the average
of the slopes (see Insert in Figure 9) for the casein and fungal protein
fed animals showed no significant difference. Growth of rats fed the other
dried fungal diets were somewhat less favorable ranging from 74 to 94
percent of the growth demonstrated In Figure 9-
Protein efficiency ratio (PER) evaluations do not distinguish between
utilization of protein for maintenance and that needed for growth, however
it is probably the most commonly used procedure. PER involves the average
net gain in body weight per unit weight of protein consumed. The PER's
of each of the dried fungal materials were compared to the casein control.
The data shown In Table 6 demonstrates the loss in protein efficiency
due to the particular drying condition Imposed on the fungal biomass prior
to diet formulation and feeding.
378
-------
CO
V)
I-
e
in
u
z
u
170
160
150
140
130
120
110
100
90
80
70
60
50
40
30
20
10
0
Figure 9.
RAT GROWTH CURVES
STANDARD CASEIN DIET x
FUNGAL TEST DIET -(PF-I)«
SLOPE OF WEIGHT GAIN / RAT
STANDARD
AVE-
7.0
6.9
8.0
7.5
7.0
7.6
7.3
TEST
5.9
7.2
7.4
8.0
7.5
7.1
AVE- 7.2
23456
8 9
K> II 12
DAYS
13
14 15 16 17 18 19 2O 21 22
-------
TABLE 6. PRpTEIN EFFICIENCY RATIOS
OTCANLING
Animal Np.of
Group Rats/
No. Test
Dried Fungal
Material
Diet
Ave .Wt . *Ave .Protein
Gain Intake
Cms . Cms .
PER
1
2
3
4
5
6
7
8
6
6
6
6
6
6
6
6
Casein Std. 142.0
PF-1 (lyo) 141.3
PF-2 (Air-115C) 130.0
PF-3 ( " -80C) 141.0
PF-4 (Vac.SOC) 125.4
PF-5 (Drum Dry) 140.0
PF-6 (Pat.Kelley) 120.5
PF-1 (lyo) 139.9
(no vitamin supplement in Group 8)
60.0
62.2
77.1
69.0
72.0
65.0
70.6
62.2
2.37
2.28
69
04
74
15
71
2.25
*Total feed consumed times x 23% protein
Group 8 rats fed unheated (lyophilized fungus) without the vitamin supple-
ment used in all the other groups, showed a PER which demonstrated that the
fungal material could supply the vitamin requirement as well as the protein
needed for rat growth.
Nitrogen balance procedures, originally used for studies in man (17),
have recently been widely used for rat evaluations of novel protein sources
(18, 19). By these procedures, one can differentiate between the degree
of digestibility of the nitrogen source and the proportion of nitrogen
retained for storage or anabolism. In other words, nitrogen balance
techniques permit differentiation between the proportion of dietary
nitrogen absorbed, whcih is a function of the susceptibility of the protein
source to the action of the gastrointestional proteolytic enzymes; and
the proportion retained in the body, which is a function not only of the
essential amino acid composition and content, but of the tfctal protein
content and the caloric density of the diet.
Digestibility and net protein utilization (NPU) were based on nitrogen
analyses of the animal feeds, urine, and fecal samples collected separately
each day during the rat feeding experiments described above. These are
apparent digestibility and NPU values, since no endogenous or metabolic
nitrogen was measured. Thus, on the basis of diets containing 23% protein
the apparent digestibility for rats fed the standard casein and the funeal'
test diets were 97% and 94% respectively. These values were determined from
the standard nutritional formula of:/IN - FN _^ where (IN) is the
IN
-xlOO
intake nitrogen from the feed and (FN) is the fecal nitrogen. NPU values
were determined according to the nutritional formula: /jN - (FN + UN)
\ IN
380
x 100
-------
showed a net protein utilization of 75% for standard casein diet and 50%
for the fungal (PF-1) diet. The 50% NPU of the fungal diet was undoubtedly
a low value since no corrections were made for endogenous, metabolic or
that approximately 20% of the nitrogen of the fungal mycelium was non-
protein nitrogen.
Because of the importance of nitrogen balance data on protein feed quality
and because 50% NPU was at variance with the growth response curves, PER,
and digestibility data for the fungal protein? a second nitrogen (NPU)
study was undertaken in which endogenous and metabolic nitrogen were also
measured. Digestibility (TPD) is measured in terms of the ratio of the
absorbed nitrogen to the total nitrogen (i.e., the difference between
the ingested and intestinally excreted nitrogen, the latter corrected for
"metabolic-N"). Nitrogen retention (BV) is calculated from the ratio of
the retained nitrogen to the absorbed nitrogen (i.e., the difference between
absorbed N and that eliminated in the urine, corrected for so-called
"endogenous-N"). The procedures used in these measurements are according
to the techniques of Njaa (20).
The results of these nitrogen reevaluation studies are shown in Table 7.
TABLE 7. NITROGEN BALANCE DATA
Animal No.ef Ave. Ave. Ave. Ave. Endog. Metab. BVl. TPD2' NPU3.
Group Diet Rats Wt. N-in Urine Fecal N N
No. Fed gm. (IN) (UN) (FN) (EN) (MN) (%) (%) (%)
1 Casein 7 250 1.08 0.382 0.090 0.144 0.087 77 97 75
2 PF-1 8 220 1.20 0.451 0.240 0.139 0.092 74 90 67
3 PF-3 7 247 1.35 0.449 0.349 0.134 0.079 76 80 61
4 PF-5 7 240 1.28 0.507 0.363 0.148 0.083 73 86 63
5 PF-1 6 160 1.21 0.460 0.235 0.136 0.093 73 88 64
(no vitamin supplement
1. BV (biological value) = IN - (UN - EN) * IN
2. TPD (true protein digestibility) - IN - (FN - MN) * IN
3. NPU (net protein utilization) = Product of BV and
TPD or IN - (FN + UN) corrected for m and m.
The nitrogen balance data; BV, TPD, and NPU shown in Table 7 compares the
four fungal feeds on these bases to the casein control and to each other.
Although differences can be seen in the four dried products from the same
381
-------
harvest. The lyophlllzed material was the best although the other two
were not all that much poorer. Unlike the NPU results obtained earlier
where endogenous and metabolic nitrogen were not considered,these NPU
Values showed the fungal protein to be comparable to the casein control
diet when the corrections were included.
All of the animals on the fungal diets appeared healthy, with clean fur,
and bright eyes. No animals had thin coats, bloody noses, or showed any
signs of baldness. Several animals fed the PF-4 fungal material to 70%
of their diet had tarry black stools. Autopsies showed no difference in
average organ weights between the casein and fungal fed rats. Only in the
case of those rats fed the PF-2 (air dried - 115°C) and PF-4 (vacuum dried-
80°C) diets, was there some evidence of fatty livers and excess internal
fat. These diets, however, were not considered "preferred" diets.
b. Chick Growth Feeding
Fungal feeding experiments were conducted using one week old birds tt
evaluate the nutritional quality and the toxicity of the fungal biomass
ao a total protein replacement for casein and soybean protein in chick
diets. Feed and water were proveded ad libitum. The Cornish Red -
New Hampshire White cross birds were randomized, wing banded, weighed, and
separated into groups of 5 chicks each. Each group had two replications
Forty chicks, age one-day, were placed on a pre-experimental soybean-
casein diet for one week before being grouped and placed on the fungal
experimental diet for a six-week growth trial. Thus, the groups of birds
were separated and Group 1 fed soybean diet, Group 2 fed the casein diet,
and Groups 3 through 8 fed the various fungal (PF-1, PF-3, PF-5, etc.)
diets. The protein content of all diets was 22% and DL-methlonine was
added to the soybean and fungal diets.
As shown in Figure 10, chicks fed the three experimental fungal diets
showed no significant difference in weight gain when compared with chicks
fed the two control diets during the six-week experimental period. Indeed,
the total weight gain of chicks fed the fungal (PF-1) diet was greater than
weight gains from the control diets. This PF-1 growth was not significant how-
ever, at the 0.05 probability level. Since there was no significant
difference in feed consumption or conversion between the experimental and
control groups, it was concluded that all the experimental fungal diets
were satisfactory for chick growth. Mortality did not occur in either the
control or experimental chick groups.
Feed efficiencies, determined by dividing the weight gained by the feed
consumed, ranged from 45% for the control to 50% for the test fungal mat-
erial. Feed conversion, or the inverse of feed efficiency and often used
in place of feed efficiency, showed the fungal diets to have a slightly
higher value than the controls - but the difference was not significant
(P4.0.05). 6 ^
382
-------
600
500
w 400
oc
o
z
300
(S
UJ
UJ
200
100
50
0
WEEK
OLD
PF-
CONTROLS • •
FUNGAL x x
12345
WEEKS ON EXPERIMENTAL DIETS
Figure 10.
6
CHICK GROWTH CURVES OF AVERAGE WEIGHT GAIN PER GROUP
383
-------
c. Other Feeding Studies
As stated earlier, multi-generation fungal feeding studies were conducted
in mice as well as piglet and gilt hog feeding studies. The mice feeding
trials were done according to the Protein Advisory Group (PAG) Guidelines
#6 on pre-clinical testing. Extended mouse feeding studies through 5
generations indicated that the fungal biomass (PF-1) is a nutritious
source of high quality protein and that it is safe and wholesome when used
as the sole source protein in these diets. Less than adequate maternal
dietary protein has been associated with long-term detrimental effects
in progeny more than one generation removed from the deficit (20). No
such effects were observed with the fungal diet in this study.
Baby gilts were fed, starting at two weeks of age, through weaning (6
weeks) on creep diets. Only four baby gilts were fed each diet. All
baby gilts were taken from the same dam litter. Following the pre-weaning
feeding trial the same animals were continued on a post-weaning diet for an
additional 25 days. The fungal protein was not the sole source of protein
in any of these diets and consisted of 55 to 58% of the total dietary
protein. Baby gilts, in the basis of weight gain and weight gain to feed
intake ratio, showed similar responses compared to the control diets.
The post-weaned pigs (gilts) however, showed a significant increase in
body weight from the fungal diet compared to those gilts fed only the control
diet. These studies showed the fungal biomass to be clearly superior to
soymeal for growing pigs beyond the weaning stage.
Economic Considerations
Justification for the commercialization of an "agro-industrial" waste
bioconversion and utilization process for the production of high grade
protein livestock feed supplements, has significant potential for improving
human nutrition in developing nations, adding to the economic development
of the rural sectors in those nations, and sparing the protein reserves of
the developed nations. The objective of this analysis was to determine
the aggregate economic feasibility of commercial fungal waste conversion
for animal feed supplements. More specifically, the objectives centered
upon: (1) an estimation of the animal feed supplement market necessary to
economically justify a commercial "agro-waste" installation, (2) an estima-
tion of the potential size of an animal feed supplement market, and (3)
an estimation of economic return on investment. Engineering cost data
were obtained from the studies reported heren. All Investment and opera-
tional cost estimates were verified through industrial sources. Plant
specifications established in this report were used as the basis of analysis.
Feasibility determination was comprised of standardized methods, including
an analysis of break-even (that market required to cover all annual Invest-
ment and operating costs) and pay-back (a measure of return on capital
investment).
Based upon cost estimates in terms of 1976 U.S. dollars provided previously
(5, 10), investment in "agro-waste" installations would be economically
feasible for the corn wet-milling and grain processing industry, A
384
-------
feasibility study centering upon "agro-waste" processing of industrial
fruit and vegetable camery wastes would likely precipitate potential mar-
kets and installations.
Summarized data from these studies showed a profit margin of 1.25 cents
per pound of BOD5 waste based on the credit accrued from sale of the har-
vested fungal solids and equated to the prevailing soymeal price of $116/
ton, was determined for the process described in this paper. Since the
price of soymeal today is twice this quotation and the capital and invest-
ment costs are not twice the costs used here - the profit margin is better
today than when these estimates were made in 1976. Also, the fungal biomass
can replace the vitamin requirement as well as protein in animal feeds, and
therefore should be more valuable a feed supplement than soymeal.
More updated estimates than those reported (5, 10) will be available
in the near future based on the more efficient plant processing studies
carried out in this report.
CONCLUSIONS AND DISCUSSION
Lest anyone has the impression that fungal fermentation is relatively
recent or something of a novel curiosity, we refer the reader to the list
of fungal food products reviewed by Hesseltine (21) where in at least one
case, that of shoyu, the fermentation is over 1000 years old. Hesseltine
believes that two major developments will be made in (1) the use of fungal
enzymes themselves to produce desirable fermented products and (2) use of
fungi on mixed substrates, i.e. combinations of cereal grains and legumes
so that the final fermentation product will have a more desirable nutri-
tional balance and the substrate mix will totally provide the carbon and
nitrogen required for fermentation.
Laboratory studies, both batch and continuous, were carried out in our
own developed apparatus which was designed to avoid the pitfalls inherent
in commercial type equipment when used for fungal studies. Details of the
fermentor will be published elsewhere.
A fungal inoculum level (10 - 20%), a low pH 4.0 to 4.5, selection of a
rapid growing fungal strain, and the inoculation - dilution start-up
procedure were found to be required for maintenance of a near pure culture
condition. Although the waste substrates were heated during plant corn
processing, they were never sterile at the time of fungal treatment. Under
optimal fermentation conditions, the contaminant level rarely exceeded 200
microorganisms per ml. of the fermentation medium. In instances where
contaminants were present ,e.q. lactobacilli, acetobacter, yeasts, an
occasional protozoan, only fungal strains predominated. Fungal con-
taminants which constituted the most challenge during periods of equipment
malfunction were found to be strains of Geotrichium, Penicillium,
Aspergillus, and Mucor.
One of the more valuable observations made during these studies was in the
area of balanced growth and steady-state kinetics achieved where nitrogen
supplementation was in slight excess . Phosphate proved to be an excellent
385
-------
limiting nutrient for a balanced systemI Although this was established
in the laboratory studies, it was more difficult under the industrial
conditions. Significant to the work, however, was In the pilot observations
where substrate concentration and substrate compositional changes occurred
regularly and such changes were rapidly accommodated by a rapidly changing
fungal biomass density. A better balanced pilot and full-scale fermentation
can be achieved than shown here by exercising more control over the nitrogen
and phosphate levels.
Dilution rate changes were made at times of severe temperature shift
(these were out-door systems) and, on occasion, recycling (not described
in this paper) of the fungal mass was found necessary to maintain
a near steady—state.
Animal feeding trials were exceedingly successful and will be described
in more detail in a forthcoming publication. Other investigators working
with fungal fermentation of other type waste substrates (7, 8, 12, 15,
21, 22) have also reported excellent animal feeding trials. Based on
healthy animal growth together with the lack of any toxicologic develop-
ments, it appears that extensive utilization of fungal protein - vitamin
feed supplements could aid in the alleviation of protein deficiency in
world population groups as well as provide valuable Industrial waste
pretreatment.
REFERENCES
1. Sukhatme, P. V. Recent trends in world food availability and their
implications, in Chavez, A., Bourges, H., and Baata, S. (ed.).
"Foods for the expanding world", Proc. 9th Internet. Cong. Nutr.,
Mexico, 1972, Vol. 3., S. Karger, New York (1975).
2. Humphrey, A. E. Product outlook and technical feasibility of SCP,
in Tannebaum, S. R. and Wang, D. I. C. (ed.). "Single Cell Protein-
II", M. I. T. Press, Chapter 1., 1-23 (1975).
3. Wang, D. I. C. Chem. Eng., _26 (17), 99 (1968).
4. Lipinsky, E. S. and Litchfield, S. H. Food Technol. 28 (5): 16,
(1974).
5. Church, B. D. Application of Fungi Imperfectl for treatment of cereal
processing wastes. Final Report to Agricultural Research Service,
Norhtern Utilization Research and Development Division, Contract
Number 12-14-100-11022 (71), United States Department of Agriculture
Peoria, Illinois, (1977).
6. Imrie, F. K. E. and Vlitos, A. J. Production of Fungal protein from
carob. In Single Cell Protein-II, M. I. T. Press, 223, (1975).
386
-------
7. Romantsehuk, H. • The Pekilo process; protein from spent sulfite liquor.
In Single-Cell Protein-II, The M. I. T. Press, 344, (1975).
8. Duthie, I. F. Animal feeding trials with a microfungal protein. In
Single-Cell Protein-II, The M. I. T. Press, 505, (1975).
Q Khor, G. L., Reade, A. E., and Gregory, K. F. Nutritive value of
thennotolerant fungi grown on cassava, J. Inat. Can. Sci. Technol.
Aliment., 9 (3): 139, (1976).
10. Church, B. D., Nash, H. A., and Brosz, W. Use of Fungi Imperfecti in
waste control. Water Pollution Control Research Series, 12060 EHT,
07/70, (1970).
12. Rolz, Carlos. Use of Cane and Coffee Processing by-products as raicro-
bial protein substrates. In Single-Cell Protein-II, The M, I. T,
Press, 273, (1975).
13. Methods for Chemical Analysis of Water and Wastes. Environ. Protect.
Ag., Water Quality Laboratory, 16020 ... 07/71, Cincinnati, Ohio,
U.S. Gov. Printing Office: 1971 0-427-263, (1971).
14. Monod, J. Annual Review of Microbiology, _3, 371-394, (1949).
15. Champagnat, A., Vertet, C., Laine', B., and Filosa, J. Nature, 197;
13, (1963).
16. Kamazawa, M. Production of yeast from n-paraffins. Single-Cell
Proteinll, Tannebaura and Wang ed., The M. I. T. Press, Cambridge,
MA., 438, (1975).
17. Thomas, K. Biological value of nitrogeneous substances in different
foods. Arch. Anat. Physiol., 219, (1909).
18. Allison, J. B. The efficiency of utilization of dietary proteins.
In Protein and Amino Acid Nutrition, (A. A. Albanese, Ed) Academic
Press, N.Y., (1959).
19. Mitchell, H. H., Hamilton, T. S., Beadier, J. R. and Simpson, F.
The importance of commercial processing for the protein value of food
products. J. Nutr., ^9, 13, (1945).
20. Chow, B. F., Comments on prenatal malnutrition. In: Malnutrition,
Learning, and Behavior. Schrimshaw and Gordon eds., M. I. T. Press,
Cambridge, MA. (1968).
21. Reade, A. E. and Gregory, K. F. High-temperature production of protein-
enriched feed from carsava by fungi. Appl. Mlcrobial., 30; (6),
897, (1975).
22. Smith, R. H. Fungal enriched barley grain for pigs. Pig Farming,
May, 68-73, (1972).
387
-------
ACKNOWLEDGEMENTS
Certain aspects of the waste treatment process were initiated with
the support of the Environmental Protection Agency under Grant Numbers
12060 EHT, 07/70 and 12060 EDZ, 08/71. Development of the laboratory
continuous fermentor animal feeding trials, and economic analyses were
carried out with the support of the Agricultural Research Service,
U.S.D.A., Northern Utilization and Development at Peoria, IL under
Contract Number 12-14-100-11022(71). Pilot and full-scale facilities
were financed by Penick and Ford, Ltd., Cedar Rapids, IA.
388
-------
WATER REUSE OF WASTEWATER
FROM A POULTRY PROCESSING PLANT
by
J.B.Andelman* and J.D.Clise**
INTRODUCTION
A wastewater treatment and reclamation system has been constructed at the
Sterling Processing Company in Oakland, Maryland. This plant slaughters and
processes approximately 50,000 birds per day in an eight hour operation,
utilizing approximately 350,000 gallons per day of treated well-water. Be-
cause of the lack of additional water of acceptable quality, the wastewater
reclamation system was constructed with the intention of mixing the renovated
water on a 50/50 basis with well water, ultimately increasing the quantity of
water needed for an increase in production capability.
The objectives of the study reported here are to determine 1) the ability and
reliability of the water reclamation system to deliver water, mixed with the
well water source, that is safe for use in processing poultry; 2) if the1 pro-
cessed poultry have any constituents harmful to human health as a result of
exposure to this mixture of renovated and well water; and 3) to recommend
monitoring procedures and parameters needed to insure the safety of the system
and the protection of human health.
This project is important as a demonstration of the technical and economic
feasibility in the food industry of moving towards the national goal of limit-
ing discharges into navigable waters. Water re-use is an important strategy
in achieving this goal. However, in such re-use of water involving human con-
sumption or exposure, it is mandatory that the health of the consumer be pro-
tected. If this project can successfully demonstrate that the renovation
process can deliver a safe and potable water, and that the poultry processed
with the water similarly pose no threat to human health, this will constitute
an important step in our overall effort towards reducing emissions to the nat-
ion 's waterways.
APPROACH
An initial study of the feasibility of reclaiming poultry processing wastewater
for reuse at the Sterling plant was reported by Clise (1). The reclamation
*Graduate School of Public Health, University of Pittsburgh, Pittsburgh,
Pennsylvania
**Maryland State Department of Health and Mental Hygiene, Baltimore,
Maryland
389
-------
system consisted of aerated lagoons, followed by microstraining, floccula-
tion and sedimentation, and filtration, with two stages of chlorination.
The schematic of the system is shown in Figure 1. The initial study showed
that the reclaimed water met U.S. Public Health Service 1962 Drinking Water
Standards for chemical, microbiological and physical constituents without
actual recycle through the poultry processing plant. Nevertheless, there was
concern that with actual reuse there was the possibility that unmeasured con-
stituents, such as pathogenic microorganisms, heavy metals, pesticides and
toxic organic chemicals, might build up in recycle and be absorbed by the car-
casses in the processing plant.
An additional project was proposed by the Maryland State Department of Health
and Mental Hygiene and funded by the E.P.A. , the purpose of which was to modi-
fy and optimize the reclamation system, to determine the capability and reli-
ability of the system for delivering water satisfactory for processing poultry,
and to evaluate the exposure of the processed carcasses to constituents that
could be harmful to human health. The role of the Graduate School of Public
Health, University of Pittsburgh, was to design, organize, and supervise the
sampling and analytical part of the study, as well as evaluate the results
from the points of view of both the quality of the renovated water and the pro-
cessed poultry possibly affected by it. Three phases were planned in this
study, the first two of which have been completed. Phase 1 involved the opera-
tion of the reclamation plant with a new sand filter, and measurement of those
characteristics pertinent to optimizing the process. Phase 2 involved a study
of a wide range of physical, chemical and microbiological constituents^ both
at various points in the reclamation system, as well as in processed carcasses
exposed to renovated water, but without actual recycle through the plant.
Phase 3 was to involve recycle of the renovated water into the processing
plant by mixing on an average 50/50 basis with well water, the mixture then
to undergo additional full-scale conventional treatment. The carcasses were
again to be measured, as was the renovated water, and comparisons made Of the
levels of contaminants with those in normal plant operation utilizing well
water only in Phase 2.
Prior to proceeding to Phase 3, an evaluation was made of the Phase 2 results
by a committee consisting of representatives of the E.P.A., the Maryland
Health Department, the processing plant, the U.S. Department of Agriculture,
and the Graduate School of Public Health. The level of contaminants in the
treated wastewater and processed carcasses were considered as to their possi-
ble health significance. A similar evaluation was to be made following Phase
3 so as to determine the safety of proceeding to continuous operation with
reclaimed wastewater. Finally, recommendations were to be made as to the need
for continuous monitoring of the system.
This paper reports primarily the principal results of the Phase 2 study and
some additional measurements obtained subsequently. Phase 3, the actual re-
cycle of the renovated water into the processing plant for a trial three-month
period, has not been instituted, even though the above Committee, constituted
for the purpose, recommended that there was no significant risk in so doing.
Some of the background and reasons for this delay will be discussed subsequent-
ly-
390
-------
FIGURE 1. WATER SAMPLING POINTS AND CONNECTIONS
FOR PHASE 2 AND PHASE 3 STUDIES
WASTEWATER TREATMENT AND
WATER RECLAIMING FACILITIES
SAMPLE IDENTIFICATION
TO RIVER
COLLECTION
BASIN
CHLORINE
CONTACT
CHAMBER
ROTARY
SCREENS
CONNECTED
IN PHASE 2
ONLY
A
C
D
SCREENED RAW WASTEWATER
SECOND LAGOON EFFLUENT
MICROSTRAINED EFFLUENT
X SED. BASIN EFFLUENT
E RENOVATED WASTEWATER
Y UNTREATED WELL WATER
Z TREATED WELL WATER
7 TREATED MIXTURE (50/50) OF
M WELL AND RENOVATED WATER
FLOCCULATION-
SEDIMENTATION BASIN
SAND
FILTER
CHLORINATOR
PRESSURE CONNECTED IN
STORAGE TANK PHASE 3 ONLY
POULTRY
PROCESSING
PLANT
NORMAL
TREATMENT
Z
1
J»
t>>Z
/ 4,
MIXING
BASIN
M
WELL
WATER
391
-------
NATURE OF STUDY
Following the optimization of the operation of the water renovation system
(Phase 1), the two-month study of Phase 2 was performed. This involved
seven separate days of sampling, approximately once a week in the Spring
of 1976. Two preliminary sampling trips were conducted to optimize samp-
ling at the site. On each of these days various water and wastewater sam-
ples were collected. In addition, twenty-five carcasses were collected
for analysis from the plant chiller system. Finally, a small "experimental"
chiller was set up to simulate the plant chiller. Twenty-five carcasses
were taken from the plant, prior to exposure to the plant chiller, and placed
in the experimental chiller in such a way and for such a period as to simu- ,
late the plant chiller. However, the experimental chiller was filled with
renovated water. The purpose was to analyze and compare the possible build-
up of contaminants in carcasses exposed to the plant chiller using normal,
treated well-water, versus those exposed to the experimental chiller using
renovated water. It must be emphasized that at no time during Phase 2 was
renovated water ever used in the processing plant.
Figure 1 shows the schematic of the wastewater renovation system, the con-
nections for Phases 2 and 3, and the possible sample points to be utilized
in the study. The actual Phase 2 sampling points and the number of samples
taken on a typical sampling day are shown in Table 1.
Not all of the sampling points were subject to all the analyses. The types
of analyses that were to be performed are shown in Tables 2 and 3. Table 2
lists the Category I analyses. As shown there, Categories la, b, and c were
to be performed only on water samples, while Id was to be done on carcass
samples as well. In fact, some la and Ib analyses were performed on carcas-
ses.
*
Table 3 shows the Category II and III analyses. As noted there, all of these
analyses were to be done on both water and carcass samples. All were per-
formed, with the exception of the halogenated methanes of Category III.
These were done subsequent to Phase 2, and will be reported here.
Although analyses of viruses were not contemplated originally in this study,
the decision was made to attempt to measure an avian virus and use it as an
indicator or sentinel of the behavior of other viruses in the water renova-
tion system. Since it was reported that the chicken flocks were routinely
inoculated with attenuated Newcastle Disease virus (NDV), it was decided to
develop the methodology and sample the water and carcasses for it. The re-
sults of this investigation will be reported. In addition, some laboratory
die-off studies were performed using lagoon water spiked with NDV.
Finally, and subsequent to Phase 2, total organic carbon measurements (TOC)
were performed on the renovated water, the normally treated well water, and
that taken at other selected sampling points. In addition, a few samples
of the treated well and renovated water were analyzed by gas-chromatography-
mass spectrometry (GC-MS) for some specific organics, other than those shown
in Table 3.
392
-------
Table 1. PHASE 2 SUMMARY OF TYPES OF SAMPLES
TYPICALLY TAKEN FOR ANALYSIS EACH WEEK
Water Samples
Location
A - Raw Waste
E - Renovated Water
Z - Treated Well Water
PC - Plant Chiller
EC - Experimental Chiller
Maximum
Number per day
2
2
2
2
2
Birds (Carcass Samples)
PB - Plant Chiller Birds
EB - Exp. Chiller Birds
5*
5*
*50 carcasses were taken for analysis each sampling day, 25 PB and 25 EB.
In each case washings from 5 carcasses were composited or otherwise com-
bined to become a single carcass sample. Hence, 25 carcasses reduced to
5 samples.
Note - Occasionally water samples were taken at other sample points shown
in Figure 1 or within the lagoons.
393
-------
Table 2. WATER RE-USE PROJECT
Types of Analyses to be Performed
Category I Analyses
la - These analyses relate primarily to waste treatment efficiency
and none will be performed on carcasses
BODc Suspended solids
Grease Total solids
Organic nitrogen Alkalinity
Ammonia nitrogen
Ib and c - These analyses relate to waste treatment efficiency and
potable water quality; none will be performed on carcasses
Ib Ic
Turbidity CCE (carbon chloroform
extract)
Color
Total Dissolved solids
Residual chlorine (total)
Residual chlorine (free)
Id - These analyses relate to waste treatment efficiency and water
quality. They will be measured for water and carcass samples.
Total plate count Fecal coliform
Total coliform
394
-------
Table 3. WATER RE-USE PROJECT
Types of Analyses to be Performed
Category II Analyses
These analyses will be performed on water and
carcass samples
Salmonella (enumeration)
Drug residual
Category III Analyses
These chemical analyses will be performed on water and
carcass samples
Arsenic
Barium
Cadmium
Calcium
Chloride
Chromium
Copper
Cyanide
Fluoride
Halogenated methanes
Hardness
Iron
Lead
Magnesium
Manganese
Mercury
MBAS
Nitrate
Potassium
Selenium
Silver
Sodium
Sulfate
Zinc
Pesticides
395
-------
METHODOLOGY
Except where noted below, all analytical methodology is consistent with
that specified by the E.P.A. or Standard Methods for the Examination of
Water and Wastewater (14th Edition). All necessary precautions for trans-
porting the samples, such as refrigeration or the addition of chemical
preservatives, were taken.
The experimental chiller using renovated (E) water was operated on a batch
basis. That is, it was filled with renovated water and plant-made ice
was added to bring the temperature initially to about 55°F. The 25 car-
casses were then lowered into the bath within the drum and rotation begun.
Additional ice to maintain 55°F was added. Approximately 15 minutes later
more ice was added to reduce the bath temperature to about 34°F, and chil-
ling continued for another 10 minutes.
The carcasses were removed by handling with clean plastic gloves, and both
plant and experimental chiller carcasses were treated in the same fashion.
They were placed, after draining, in either clean or pre-sterilized (by
autoclaving) plastic bags and carried to the laboratory trailer. 1500 ml
of either distilled or distilled and sterile water was added to each such
bag, which was then shaken for one minute and the water contents poured for
analysis.
Salmonellae were isolated from the water and poultry samples using membrane
filtration techniques modified from the recent literature. Quantitation of
salmonellae was attempted employing 15 dilution tubes to determine the most
probable number (MPN) index of organisms. Subsequent to the Phase 2 study
our research on media selective for salmonellae developed to the point of
being able to quantify by this technique. In the Phase 2 results, however,
only their qualitative presence or absence are reported.
A disc assay method modified from the procedure of Huber et al. (2) was em-
ployed to detect residual drugs in the water and carcass samples. The
first method used in this laboratory employed Bacillus subtilis (Difco spore
suspension containing approximately 1015 spores/ml) as the test organism
and seed agar (antibiotic medium No. 1) as the test substrate. This proced-
ure was further modified according to the method of Read et al. (3) in order
to detect bacitracin and to enhance the sensitivity of detecting sulfa drugs
and antibiotics in the water and carcass samples. Bacillus cereus and Muel-
ler-Hinton agar were thus employed as the test organism and the test sub-
strate, respectively.
Primary chick embryo cell cultures were used with a plaque assay to detect
the presence of avian viruses in the water and carcass samples, which were
first filtered through 0.45 micron pore-size membrane filters. Following
inoculation of the cell monolayers with the filtered samples and various in-
cubation procedures, plaques were counted and the plaque-forming units per
ml (PFU/ml) determined. In order to detect the possible presence of such
viruses in the carcasses, tissue extracts were prepared from spleens, lungs
and livers, processed and analyzed for the presence of avian viruses, such as
NDV, capable of hemagglutinating chicken red blood cells.
396
-------
RESULTS
Most of the individual analytical results will not be presented. Rather,
for the purpose of this paper and wherever possible, results will be sum-
marized or presented in statistical fashion.
Microbiological and Drug Residual
Neither salmonellae nor any other Enterobacteriaceae were ever isolated in
the chlorinated renovated (E) or well water (Z). The carcasses exposed to
and the water from the experimental chiller, using the renovated water, were
similarly negative in this respect. Salmonellae was isolated in raw waste-
water (A). However, the aerated lagoon and renovation system was complete-
ly efficient in disinfecting this bacterium.
All samples tested for drug residual using both Bacillus subtilis and
Bacillus cereus were negative. Thus, this test has not detected any anti-
biotic drug in either the water or carcass samples.
Similarly, avian viruses, including NDV, have not been detected either in
the water or carcass samples, nor in tissue extracts from chickens processed
at the Sterling plant. Only preliminary results of the NDV laboratory die-
off experiments using lagoon water can be reported at this time. These sam-
ples, spiked with NDV, were studied for up to seven days at two temperatures,
in the light and dark. There was no apparent effect of light. At 25°C die-
off was complete (indicated by the absence of plaque-forming units in the
assay) after 3 days. At 7° C, however, the die-off was very slow, less than
a factor of 10 in one week. These experiments are continuing, particularly
to elucidate the possible role of particulate matter in the lagoons in virus
survival.
The renovated water (E) was analyzed for bacteria on eight days during and
just prior to the Phase 2 period, 19 samples being taken. On six of the
eight days the total and fecal coliform results were less than 2 organisms
per 100 ml. On April 5 and 12 four E samples were analyzed. The reported
total coliform ranged from 15 to > 240 organisms per 100 ml, and the fecal
coliform from 9 to > 240. However, there is a strong likelihood that
these E sample results were confused with those of EC, the experimental
chiller water. The results for the latter for those two weeks were report-
ed as negative for coliform and fecal coliform, a highly unlikely result
compared to all other EC and PC coliform analyses, which were >240.
The total bacterial plate count analyses on the samples described above
indicated a maximum value of 25 organisms per ml for the Z samples and 15
for E, except for the four samples which are believed to be incorrect. For
these latter four, the values ranged from 1 to 2,000. A reasonable maximum
total plate count for municipal water is 500 per ml.
In view of the high probability that the inadequate bacterial quality of the
four E samples was due to sample reporting error, and that no E samples have
yielded any enteric pathogens, it is concluded that the renovated water is
397
-------
of excellent bacterial quality. It should also be noted as well that in
actual recycle the E water will receive additional treatment, including
chlorination, after being mixed 50/50 with the untreated well water.
Inorganic Chemical Water Quality
A summary of the results for several macro constituents and water quality
characteristics for Z and E is presented in Table 4. Other than pH, alkali-
nity, and nitrate, the mean concentrations of all these parameters were high-
er in the renovated water, E, than in the treated well water, Z. However,
these concentrations in E are not hazardous, and only one maximum value shown,
252 mg/1 for sulfate, is at the criterion level, which has only been a second-
ary standard. This sulfate buildup is not of concern, but could be reduced,
since it is primarily due to the alum coagulant.
It should be noted that some of the differences in the mean concentrations
of chemicals in Z and E are not statistically different when the variability
of the results are considered. Thus, color cannot be considered to be statis-
tically different in the Z and E water. However, there is unquestionably a
build-up of dissolved macro constituents in the renovated water. This is not
and, at the levels encountered, does not pose a health hazard.
Attention should be called, however, to the low alkalinity in E, and the low
pH values. On some days the pH was as low as 3.3 due to the large doses of
chlorine. The E water will be mixed with untreated Z in Phase 3, which will
mitigate this problem. However, to avoid corrosion in the system, soda ash
or caustic soda has been added to raise the pH and alkalinity when required.
Table 5 presents a comparison for Z and E for certain measurements related
to waste treatment parameters. The total solids in E are obviously higher
than in Z, but this is due to the dissolved solids already discussed. The
organic nitrogen values for E reflect its higher organic content.
Of some interest and possible concern among the constituents reported in
Table 5 is the relatively high average concentration of ammonia nitrogen in
the renovated water. It probably results from biological denitrification re-
actions in the lagoons, although some concentrations measured in the raw waste
are comparably high. The principal concern is that the ammonia reacts with
the chlorine disinfectant to form chloramines, which are less effective dis-
infecting agents. However, because of this very reaction, unless the samples
are analyzed immediately, it is unlikely that with the practice of breakpoint
chlorination the ammonia should have been detected. Subsequent to Phase 2
the ammonia concentrations in E decreased considerably. It should be empha-
sized that the excellent bacterial quality of E water indicates that disin-
fection has not been affected. Also the presence of these concentrations of
ammonia is not known to be a health hazard. Finally, it is noteworthy that
ammonia is sometimes added in municipal water treatment plants in order to
react with chlorine and form longer-lived chloramines.
A summary of the results for the trace constituents in Z and E waters is
shown in Table 6. It should be emphasized that the concentration units here
are micrograms per liter. All of the maximum concentrations were well below
398
-------
Table 4. PHASE 2 - WATER QUALITY CHARACTERISTICS AND MACRO CONSTITUENT CONCENTRATIONS*
(in rag/1, except where noted)
Treated 1962 PHS
Well Water (Z) Criterion
Turbidity (JTU)
Color (units)
PH
Alkalinity
Chlorine Res.
Surfactants (MBAS)
co Na+
UD
K+
„ -H-
Ca
Mg~
cr
3
S0~"
HCO^est)
Ion Total
Dissolved Solids
N
6
6
12
12
-
14
14
16
16
16
12
8
8
-
12
X
0.9
3.0
7.0
107
-
<0.01
7.4
2.9
40.1
2.4
11.6
4.0
9.6
130
208
143
Value
£ Max
0.5 1.7 5
1.6 5 15
0.2 7.2
15 130
-
0.5
3.4 12.5
3.6 8.7
4.8 49.4
0.3 2.9
1.0 13.4 250
0.6 4.9 45
3,6 13.5 250
- - -
33.8 194 500
N
35
36
30
15
52
13
14
15
16
15
12
8
8
-
16
Renovated Water
X
1.6
3.7
5.8
35.3
1.7
0.04
30.2
14.7
53.7
2.95
88.5
3.5
150
43
393
389
£
1.7
3.3
1.4
37
1.3
0.02
12.5
2.3
5.8
0.3
16.8
1.5
85.6
-
66
(E)
Max
2.7
10
6.9
140
6.0
0.08
47.7
17.4
64.9
3.4
121
6.2
252
-
492
^Occasionally extreme values (beyond 2cr) discarded
-------
Table 5
PHASE 2 - CERTAIN WATER CHARACTERISTICS RELATED PRIMARILY TO WASTE
TREATMENT EFFICIENCY (in mg/1)
Treated Well Water (Z)
Renovated Water (E)
Total solids
Suspended solids
BOD5
Grease
Organic-N
Ammonia-N
Dissolved Oxygen
Table 6 . PHASZ 2
Treated Wall
N X
Cu 16 41.4
F 14 58
Fe 16 19.2
Mh 16 1.8
Pb 16 21.2
Zn 12 25.2
N X
12 165
12 10.4
12 5.3
12 5.2
8 0,013
8 0.017
12 8.1
_a Max N X _a
33 234 13 418 72
8.1 26 14 16.8 19.7
4.3 18 14 3.4 2.2
6.1 23 13 5.1 4.8
0.008 0.030 10 1.7 1.5
0.013 0.040 10 19.0 4.4
0.6 9.4 14 9.3 1.4
Max
501
70
7.0
18.2
5.0
23.0
11.7
- TRACE CHEMICAL CONCENTRATIONS* (ug/1)
Water (Z)
a Max.
11.2 57
33 13
7.8 31
1.0 3
15.7 50
3.8 33
PHS 1962
Criterion Renovated Water (E)
Value
N X a
1000 14 38.8 12.9
1000 14 151 54
300 14 57.1 24.3
50 14 2.6 1.0
50 15 23.8 13.9
5000 12 27.2 6.9
Max
56
230
98
4
50
38
Occasionally extreme values (beyond 2o) were discarded
400
-------
the PHS criterion values, except for lead in both E and Z samples. These
were not significantly different for the two waters, and the high concen-
trations ^in each case were just about at the criterion value. Only in the
case of iron and fluoride were the concentrations significantly higher for
E compared to Z. However, it is judged that for none of these six trace
elements were there any hazards in the renovated water.
Although not shown in Table 6, all the measurements for silver, arsenic,
cadmium, chromium, and selenium were negative (below the sensitivity levels).
About half of the E samples were positive for cyanide at concentrations up
to 12 yg/1, but well below the health criterion value of 200 yg/1. It is
difficult to imagine any oxidizable cyanide being present in E because of
the large quantities of added chlorine. However, the analysis was for total
cyanide, so that the measurement may have detected such harmless combined
complexes as those involving iron, often used as an anti-caking agent, such
as in road salt. Several water samples were positive for mercury. These
occurred in three of the eight weeks, two of which were analyzed on the same
day. It is likely that in the latter cases there was contamination or analy-
tical error, perhaps as a result of the mercury preservative added to the ni-
trate sampling bottle. On two subsequent weeks when this possibility was re-
moved, only one out of six E samples were positive (0.6 yg/1), and no Z sam-
ples. In contrast, during the previous two weeks E, Z and all other samples
were positive and much higher. For the samples of the first four weeks none
of the E or Z samples were at concentrations higher than 0.2 yg/1, the level
of sensitivity of the method. In view of the fact that the highest E sample
was 1.7 yg/1, even though it was probably an erroneous reading due to con-
tamination, and the fact that the criterion value is 2 yg/1, it may be con-
cluded that there is no observed health hazard from mercury in the renovated
water.
Carcass Chemical Analyses
The results of the chemical analyses of the washings from the carcass samples
PB, from the plant chiller, and EB, from the experimental chiller, will be
summarized here, but not tabulated. Most of the chemical species analyzed in
water, as shown in Tables 4, 5, and 6, were also analyzed in the carcass wash-
ings. Among the macro constituents only the mean nitrate concentrations were
statistically significantly higher at 0.38 mg/1 in the EB samples compared to
0.06 in PB. The reason for this difference is not known, since the Z and E
waters are not significantly different in nitrate concentrations, as shown in
Table 4. It is very unlikely that this poses any kind of a health hazard.
Taking the mean value of 0.38 mg/1 of nitrate in the EB samples, and assuming
that this all came from the carcasses, since 1.5 liters of distilled water was
used to elute the carcasses, this would be equal to 1.5x0.38, or 0.57 mg nitrate
per carcass. This is not a quantity of concern in the perspective, for example,
of the current primary drinking water regulation, which permits an average daily
adult ingestion of about 90 mg nitrate.
The mean values of all the waste characteristic and trace element measurements
were the same, within statistical confidence, for the PB and EB samples in
Phase 2, with the exception of ammonia nitrogen. The mean EB value was 1.1
mg/1, while that of PB was 0.09. This result is most probably attributed to
401
-------
the high ammonia content of the renovated water, but cannot be considered
to be a health hazard.
Organic Water Quality
A variety of measurements of gross organic parameters and specific organic
chemicals have been performed during and subsequent to Phase 2. These in-
clude 8005, organic nitrogen, and CCE (carbon chloroform extract), as Cate-
gory I analyses, shown in Table 2; halogenated methanes, MBAS (surfactants),
and pesticides, as Category III analyses, Table 3;total organic carbon (TOC);
and specific organics extracted with methylene chloride and identified by gas
chromatography-mass spectrometry (GC-MS).
A comparison of the mean results for the surfactants, shown in Table 4, in-
dicates significantly higher concentrations in the renovated water, E, com-
pared to the treated well water, Z. However, both the mean concentration
of 0.04 mg/1 and the maximum of 0.08 are well below the 0.5 mg/1 1962 PHS
criterion value.
The BOD5 summary values are shown in Table 5, indicating a somewhat lower
mean value for the E compared with the Z water, although the difference is
not statistically significant. In contrast, the average for the organic nitro-
gen, defined as the organically bound nitrogen in the negative-three oxida-
tion state, is higher in the E water at 1.7 mg/1 than that of Z, 0.013. It
also shows considerable variation. It most probably consists of protein mat-
erial and its breakdown products, such as amino acids, from the chickens.
Analyses were performed for nine pesticides, samples being collected on three
separate days in a three-week period. These analyses were discontinued when
all samples from the second and third days were negative. The results are
summarized in Table 7. No positive values were obtained for the E or Z sam-
ples, only for A, the raw waste, and PB and PC, the carcasses and chiller
water in the plant. The total number of E samples analyzed was nine.
A summary of the total organic carbon (TOC) analyses of samples taken on five
successive weeks in November and December, 1976 is shown in Table 8. Each in-
dividual TOC value is a result of at least two measurements on the same sample,
the TOC being the difference between the total carbon and the inorganic carbon
analyses. The total number of samples for Z, E, X, and C shown in Table 8
were 10, 8, 6 and 4, respectively. Two samples each were measured at each of
the other sampling points. These results show that there was a considerable
reduction in TOC through the lagoon and then the renovation system. Although
the mean value for the renovated water, E, was about 5 mg/1 higher than that
for Z, the treated well water, they each showed considerable variability, as
indicated by their respective ranges and standard deviations.
Several carbon chloroform extract (CCE) measurements were made, both during
Phase 2 and at other times, the results being shown in Table 9. These measure-
ments were done by the newer miniaturized two-day sampling technique (4) for
which a standard had been proposed of 0.7 mg/1 compared to the 0.2 value'for
the older technique in the 1962 PHS standards. In fact, it was shown in a com-
parison of the two techniques that the newer method measures about 6.7 times
402
-------
Table 7. Pesticides Analyzed and Found (ug/1)
Pesticide
Chlordane
Endrin
Heptachlor
Hept. Epoxide
Lindane
Methoxychlor
Toxaphene
2,4-D
2,4,5-T
Sensitivity
Limit
0.2-1
0.1
0.06
0.1-0.2
0.06
0.5-1
3-6
0.05-1
0.5-1
Criterion*
Value
3**
0.2
0.1**
0.1**
4
100
5
100
10
Positive
Values
-
0.1 (A, PB)
0.09-0.4 (A, PB, PC)
-
0.04-0.14 (A, PB, PC)
-
-
-
-
*E.P.A. National Interim Primary Drinking Water Regs., except for:
**Proposed as above, but not adopted
Table 8. Total Organic Carbon (TOC) Analyses of a Few
Samples in Fall 1976 at the Sterling Plant
Concentrations - mg/1
^ Sample Point Mean Stand. Dev. Range
Z - treated well water 14.5 7.0 6-26
Effluents from;
L! - lagoon 1 50.5 2.2 49 - 52
C* - lagoon 2 35.0 1.0 34 - 36
C - lagoon 2 (chlorinated) 38.8 4.9 35 - 45
D - microstrainer 39.8 7.4 35-45
X - sedimentation basin 22.4 3.2 19 - 29
E - renovated water 20.0 5.1 15 - 31
403
-------
Table 9. Carbon Chloroform Extract Concentrations (CCE)
at the Sterling Renovation System
1976 Week
Sampled
Untreated Well
3/8
Renovated Water
3/15a
3/29
4/19
5/17
6/14
7/12
7/26
11/15
Renovated Water - Mean
Standard Dev.
CCE - mg/1
0.58
0.96
1.2
1.5
0.18
0.36
0.22
0.58
0.61
0.52
0.68
0.45
a - stored water
404
-------
as much CCE as the older one (4). The results in Table 9 indicate consid-
erable variability in the CCE values in the renovated water, E, but these
values overlap with and on the average are not very different from the
single measurement of the untreated well water.
On four successive weeks in November and December, 1976 volatile halogenated
methanes were measured, essentially using the technique of Bellar and
Lichtenberg (5). The only organic found by this technique in the renovated
water, E, was chloroform, the maximum concentration being 3 micrograms per
liter. In some of the treated well water samples, Z, chloroform was measured
at concentrations of less than one microgram per liter, and one showed traces
of carbon tetrachloride and dibromochloromethane. For reference, the mean
chloroform concentration in finished U.S. public water supplies was reported
to be 21 micrograms per liter in the National Organics Reconnaissance Sur-
vey (6) .
Finally, samples of E and Z water collected on two separate days subsequent
to Phase 2 were extracted by methylene chloride, concentrated by evaporation
and analyzed by GC-MS for specific organics. Three general types of compounds
were identified by this technique. At the moment, the analyses have not yet
been confirmed with standards. In the renovated water, E, the following nor-
mal fatty acids were identified: C-10, -11, -12, -14, -15, -16, and -18; in
the treated well water, C-12, -14, -16, and -18. None of these contain chlor-
ine atoms. In the renovated water dioctyl and dibutyl phtalate were identi-
fied, but only the former in the treated well water. Both of these compounds
are used as plasticizers and have been widely found in a variety of waters,
including potable municipal supplies (7).
The most unusual set of compounds that was found in all of the E and Z sam-
ples, analyzed by GC-MS, consists of five organics that appear to be halo-
genated or hydrohalogenated derivatives of cyclohexene. They are: 3-chloro-
cyclohexene, 2-chlorocyclohexanol, 1,2-dichlorocyclohexane, l-bromo-2-chloro-
cyclohexane, and 2-bromocyclohexanol. Only one U.S. company manufactures
cyclohexene, the presumptive parent compound, in any significant quantity, and
it has very specialized uses. In view of this and the fact that such uses are
unlikely in the vicinity of the Sterling plant, it is also highly unlikely that
it is a contaminant either of the ground water source or from the poultry
plant operation. Recent evidence developed elsewhere seems to indicate that
it, or the halogenated derivatives, are probably contaminants of the chlorine
used for disinfection. Some of these compounds have recently been found in
two municipal water supplies, probably from the chlorine used in the treat-
ment process. This situation is being investigated.
DISCUSSION
The focus of this discussion will be whether the quality of the renovated
water, E, is sufficient to justify its reuse in processing poultry at the Ster-
ling plant, without risking the health of the consumers. It should be empha-
sized in this discussion that, prior to actual use in the plant, the reno-
vated water will be mixed 50/50 with the untreated well water, then receive
the normal water treatment that is currently utilized for it. This consists
405
-------
of prechlorination in the mixing basin, alum-lime flocculation, with final
pH adjustment to precipitate iron, settling, and filtration through two
sand filters. Additional chlorine is introduced into the main service line
leading to the processing plant for residual control.
The principal operation in the processing plant that exposes the carcas-
ses to water is the chiller, which precedes the icing and packaging of the
processed meat. Each carcass can take up as much as 10 per cent of its
weight in water in this operation. The chiller is a continuous, counter-
current system which itself rapidly accumulates contaminants in the course
of a day's operation (8). The evaluation of the renovated water, E, studied
in Phase 2, should be done in the perspective of the further treatment it
will receive in actual reuse, as well as the intended use of that water.
In that use the most important water exposure of the poultry being processed
in the chiller operation.
The microbiological quality of the renovated water, E, studied in Phase 2
is excellent. No pathogenic bacteria were detected in that water. Aside
from some confusion, probably from mislabeling of samples, all the coliform
and fecal coliform concentrations were below the level of detection. The
total plate counts were also low. Avian viruses were also not detected in
the renovated water. In view of the approximately two-week's detention time
in the aerated lagoons and the nature of the disinfection processes subsequent
to them, which involve two stages of chlorination, this excellent microbiolo-
gical quality is to be expected. With actual recycle into the plant, this
high quality and the additional treatment, including disinfection, would in-
sure, with a high degree of certainty, that there would be no danger from
pathogenic organisms in the reuse of this renovated water.
The inorganic water quality of the renovated water studied in Phase 2 is also
quite good. As would be expected, several of the major cations and anions
are significantly higher in the renovated water than the normally treated
well water. Such build-ups are common, since typical waste and water treat-
ment processes are not designed for, nor capable of removing significant
quantities of these materials. The concentrations of these constituents in
the renovated water, shown in Table 4, do not constitute a health hazard in
water reuse.
The trace elements and waste parameters shown in Tables 5 and 6 are also not
at levels that are of concern, were the renovated water actually reused as
contemplated. The trace elements are not significantly different for the
renovated and normally treated well water, with the exception of cyanide,
fluoride, and iron. All of these are, nevertheless well below criterion lev-
els in the renovated water. One can justifiably conclude, therefore, that,
as with the microorganisms, the inorganic constituents of the renovated water
do not pose a risk to human health in the actual planned recycle system. This
conclusion is reinforced by the comparison of the analyses of the carcasses
processed normally in the plant chiller and those exposed to renovated water
in the "experimental" chiller. Among all the macro and trace inorganic con-
stituents, only ammonia and nitrate were higher in the carcasses processed
with renovated water, but not at concentrations that would constitute a
health hazard.
406
-------
The organic chemical quality of the renovated water at the Sterling plant
remains perhaps the greatest area of possible concern, as it does for the
municipal water supply systems of the U.S., principally because of the re-
cent advances in our ability to identify and quantify trace organic chemi-
cals at very low concentrations. Several specific organic chemicals have
been identified and some quantified in this study. As noted previously,
pesticides were not found in either the renovated or treated well water.
Surfactants in the former were well below criterion levels. Several fatty
acids were found in the renovated water, but also in the treated well water.
In any event, these constitute no human hazard. The maximum concentration
of the only halogenated methane found in the renovated water, chloroform,
was three micrograms per liter, well below the approximate median value of
20 found in the E.P.A. National Organic Reconnaissance Survey of U.S. pub-
lic water supplies (6). Two phthalates were found in the renovated water,
and one of these in the treated well water. As noted perviously, both of
these, widely used as plasticizers, have been found in potable U.S. munici-
pal water supplies, as well as many natural waters.
The unusual halogenated and hydrohalogenated derivatives of cyclohexene,
found in both the renovated and treated well water, in all probability are
contaminants of the chlorine used in disinfection. They have been found in
finished municipal water supplies and are unrelated to the renovation sys-
tem. That is, it is quite likely that changing to a different source of
chlorine will eliminate their presence.
It is a reasonable judgment therefore, that the specifically identified
organic chemicals arising from this waste renovation process do not con-
stitute a human hazard were this water to be used as contemplated in full
recycle. One might nevertheless raise the question of the possible health
hazard from organics not yet identified. It is unlikely that, in terms
of the reuse of this water for processing poultry, such organics would be
hazardous, since they arise primarily and originally from the poultry wastes
and are likely to be only natural materials and their degradation products.
A possible question concerns the reaction of chlorine with these materials
to form hazardous by-products. This concern is difficult to address. It
should be pointed out, however, that chlorination is practiced in certain
food processing, and has probably not been associated with any ill effects
in humans.
The measured gross organic parameters, namely BOD, CCE, TOG, and organic
nitrogen, are also of interest as indicators of specific organic constituents,
The higher organic nitrogen in the renovated water most probably reflects
the proteinaceous material and its breakdown products from the poultry, but
only those constituents that are not readily biodegradable, since the BOD
values for the renovated and treated well waters were quite comparable.
The mean BOD value of 3.4 mg/1 for the renovated water is not untypical of
many raw surface waters that are used for municipal water supplies, such
as Minneapolis and St. Cloud (9).
The mean TOC values for the renovated and treated well water, 20 and 14
mg/1, respectively, are perhaps somewhat high, but not unusually so. The
statewide average for the major Minnesota rivers is 20 mg/1, with several
407
-------
large municipal supplies utilizing them as raw water supplies (9). Similar-
ly, most of the larger rivers there have concentrations of 15 to 30 mg/1.
In a study of 80 municipal water supply systems of the U.S., non-volatile
TOG concentrations as high as 19 mg/1 were measured in the raw water and 12
mg/1 in the finished water (6). About 98 per cent of the latter were less
than 5 mg/1. This indicates that, in terms of a finished water supply, the
renovated water TOG values are undoubtedly high. However, after mixing 50/50
with the raw well water and then full scale treatment, the final mixture
should not be significantly different in TOG than the currently used treated
well water.
The CCE measurements of the finished water show considerable variability.
However, time variations, seasonal or otherwise, are also common in finished
municipal water supplies (4). Similarly, concentrations of 1.0 mg/1 of CCE
are not uncommon in finished water supplies, and concentrations considerably
higher than this have been found in rivers. The average concentration of
0.7 mg/1 for the renovated water system had been recently proposed as a criter-
ion value for public water supplies.
One can reasonably conclude that the organic water content of both the treat-
ed well water and the renovated water is somewhat high. Nevertheless, no
specific organics in the renovated water have been found at high enough con-
centrations for the renovated water to be considered a significant risk to
human health. At the same time, the most likely non-hazardous nature of any
incremental organic material not yet identified in the renovated water is
such as to reinforce that judgment.
An evaluation of many of these results was performed immediately following
the completion of Phase 2 by a committee constituted to do so and make a
recommendation about proceeding to Phase 3, a three-month trial period of
full recycling through the plant and water reuse. That committee did not at
the time have access to some measurements of organic water quality which were
done subsequent to Phase 2, including most of the CCE analyses, and all of
the TOC and specific organics, other than pesticides. The committee recom-
mended that there was no significant risk in proceeding to Phase 3. We con-
clude that the additional data on organic chemical quality should not modify
that judgment.
Nevertheless, the final decision to proceed to Phase 3 was not and has not
been made as of this writing. The delay arose because of the requirement
of the Department of Agriculture that the water to be reused in the plant
be designated as potable. There are differing opinions as to whether it
could be so regarded. The concerns center around two areas. First, is the
chemical and microbiological quality of the renovated water sufficient to
meet criteria of potability? In terms of meeting constituent limits speci-
fied in drinking water standards or regulations, the answer is yes. the
gross organic load is high, but not much more so than the normally treated
well water. Nevertheless, a certification of potability has been made by
the legally authorized agency, the State of Maryland.
408
-------
The second area of concern related to potability is the nature of the raw
water source. A long-standing concept, as stated in the 1962 Public Health
Service Drinking Water Standards (10), is the following:
"The water supply should be obtained from the most
desirable source which is feasible, and effort should
be made to prevent or control pollution of the source.
If the source is not adequately protected by natural
means, the supply shall be adequately protected by
treatment."
In terms of the intended goal of this water renovation system, namely the
augmentation of the limited non-community well water source for the Ster-
ling plant, other possible available sources should be considered, using
the above concept. The local community, Oakland, will not and cannot
provide additional water to the Sterling plant. The only other possible
source is the Little Youghiogheny River, often not more than a small creek,
polluted immediately upstream by raw, municipal sewage from Oakland. It
is thus apparent that in terms of the concept of most desirable feasible
source, the renovation system meets that criterion. It must also be empha-
sized that the renovated water, studied in Phase 2, will receive the addi-
tional full-scale, normal water treatment during the Phase 3 trial period
of reuse. In this sense it may also be regarded as a raw water source, and
certainly a most desirable one.
Nevertheless, one can and perhaps should put aside this legal question of
potability and consider the following question. If this renovated water
were to be recycled into the plant as described and used as intended, would
there be any significant, discernible risk to the consumers of the chickens
processed there? It is our considered judgment after weighing all the re-
sults of this study and that which preceded it, that the Phase 3 trial period
of full recycle and reuse should proceed, and that the public health will not
be jeopardized in so doing. In view of the urgent need to conserve our water
resources, limit waste discharges, and improve water quality, the nation will
have to proceed to selective reuse of wastewater. Such a project as this is
a useful step in that direction.
ACKNOWLEDGEMENTS
This study is a joint effort by the Maryland State Department of Health and
Mental Hygiene (MHD), the Sterling Processing Company, and the Graduate School
of Public Health, University of Pittsburgh (GSPH). The study is being funded
by the E.P.A. under two grants, one to GSPH and the other to MHD. The authors
gratefully acknowledge this support and the technical guidance and encourage-
ment of the two E.P.A. project officers, Herbert Pahren and Jack Witherow.
We are also pleased to acknowledge the enthusiastic cooperation and partici-
pation of Mr. Gilman Sylvester, the manager of the Sterling Processing Corpora-
tion, and his staff; also, Dan McGrail who has been operating the water reno-
vation system, and Edward S. Hopkins, who supervises its operation.
Many analyses were performed by the Cumberland laboratory of MHD, and several
by the Analytical Services Laboratory of the NUS Corporation in Pittsburgh.
409
-------
Drs. John Armstrong and Robert Yee, faculty members of the Microbiology
Department of GSPH, have been particularly helpful in advising on the
microbiological analyses, as has Dr. Iain Campbell of the Life Sciences
Division of the University of Pittsburgh on the trace organic analyses.
Several staff members at GSPH were fully involved in this project, most
notably Teresa Lester, who had the principal responsibility for all as-
pects of the microbiological analyses.
REFERENCES
1. Clise, J.D., Poultry Processing Wastewater Treatment and Reuse,
Environmental Protection Technology Series, EPA-660/2-74-060,
U.S. Environmental Protection Agency, Washington, D.C., March, 1974.
2. Huber, W.G., Carlson, M.B., and M.H. Lepper, "Penicillin and Anti-
microbial Residues in Domestic Animals at Slaughter," J. Amer. Vet.
Med. Assoc., 154, 1590-1595 (1969).
3. Read, R.B.,Bradshaw, J.G., Swartzentruber, A.A., and A.R. Brazis,
"Detection of Sulfa Drugs and Antibiotics in Milk," Applied Micro.,
_21, 806-808 (1971) .
4. Buelow, R.W., Carswell, J.K., and J.M. Symons, "An Improved Method
for Determining Organics by Activated Carbon Absorption and Solvent
Extraction - Part 1, " J. Amer. Water Works Assoc.. 65, 57-72 (1973).
5. Bellar, T. and J.J. Lichtenberg, "Determining Volatile Organics at
Microgram-per-litre Levels by Gas Chromatography," J. Amer. Water
Works Assoc., 66. 739-744 (1974).
6. Symons, J.M., Bellar, T.A., Carswell, J.K., DeMarco, J., Kropp, K.L.,
Robeck, G.G., Seeger, D.R., Slocum, C.J., Smith, B.L., and A.A. Stevens,
"National Organics Reconnaissance Survey for Halogenated Organics,"
J. Amer. Water Works Assoc., 67, 634-647 (1975).
7. Shackelford, W.M. and L.H. Keith, Frequency of Organic Compounds
Identified in Water, Environmental Monitoring Series, EPA-600/4-76-062,
U.S. Environmental Protection Agency, Washington, D.C., December 1976.
8. Hamza, A., Saad, S., and J. Witherow, "Water Reuse in Poultry Pro-
cessing: Case Study in Egypt," THESE PROCEEDINGS.
9. Maier, W.J. and H.L. McConnell, "Carbon Measurements in Water Quality
Monitoring," J. Water Pollution Control Feder., 46, 623-633 (1974).
10. U.S. Public Health Service, Drinking Water Standards. 1962, U.S.
Department of Health, Education and Welfare, Washington, D.C.
410
-------
WATER REUSE IN POULTRY PROCESSING:
CASE STUDY IN EGYPT
by
Ahmed Hamza*, Samia Saad*, and Jack Witherow**
The past decade has been marked by unprecedented global activity directed
toward man's concern for environmental quality. In many nations, this
concern has been translated into initiation of intensive research programs
and adoption of strict measures to protect and enhance the quality of the
environment.
In Egypt, population and industrial growth have made increasing demands on
the scarce water resources, and at the same time people are demanding up-
grading of water quality for public health, aesthetic and recreational pur-
poses. The emergence of these paradoxical demands has presented a problem
in providing water to meet the rising demands and obtaining the desired
water quality. In addition to this problem, the poultry industry faces a
unique situation, since new production farms and processing plants will be
located in remote areas as a part of an integral plan to revive the Egyptian
deserts. The stringency of water supply for the new plants coupled with the
compelling need for prodigious quantities of water have led to the recogni-
tion of used water as a vital part of the water cycle in poultry processing.
Despite the considerable research efforts which have been devoted to the problem
of water renovation and reuse in poultry processing (1,3), little has been
done to implement the accumulated information in long term field studies.
Water recycling in poultry processing is restricted as it confronts public
health regulations and established traditions.
Recognizing that applied research is needed in this vital area, Alexandria
University was awarded a U.S. Environmental Protection Agency grant to study
application of water recycling in poultry processing. Field studies were
initiated at Alexandria Poultry Processing Plant (APPP), during April 1976.
Research efforts in the first phase of the study were directed toward evalua-
tion of water and poultry characteristics during processing, and documenta-
tion of necessary information for selection of potential sources of water for
reuse. The second phase will focus on application of a multiple water reuse
system, assessment of its technical and economic aspects and most important
its impact on sanitation of the process and product.
*Assistant Professor of Sanitary Engineering, Higher Institute of Public
Health, Alexandria University, Egypt.
**U.S. Environmental Protection Agency, Food and Wood Products Branch,
Corvallis, OR.
411
-------
UNIT OPERATIONS AT APPP
A detailed discussion of the operations at APPP will not be attempted since
it is almost identical to those in the U.S. and has been discussed in consi-
derable detail by previous authors (1, 4). However, a brief description
of several unit operations at APPP is necessary for understanding the avail-
able water sources for reuse.
A flow diagram of APPP and water usage by various operations is shown in
Figure 1 and Table 1, respectively. Table 1 is based on 69 sets of observa-
tions from May to September 1976.
Killing and Bleeding
The live birds are stunned by low voltage current, and slaughtered by cutting
the jugular vein and allowing the blood to drain in a restricted area.
A sizable waste load is associated with this area due to the loss of blood
to the sewer system.
Scalding and Defeathering
Birds are subscaled at 54-60°C, while adjusting the daily overflow to 5 liters
per bird, (I/bird). The scalder effluent is highly contaminated with bacteria,
blood and other residues. The scalded birds are defeathered in three conse-
cutive picking machines, and the feathers are flumed to the rendering area
using the effluents from the scalder and pickers in addition to fresh water.
Water used in this area accounts for 28.3% of the total water usage.
Evisceration
The carcass cavity is opened for removal of entrails. Giblets are cleaned
and wrapped for bulk selling. After opening and washing the gizzard,
they are transferred manually to the wrapping area without fTurning to save
water and to eliminate the problem of treating the fTurning effluent. The
inedible viscera are dropped into a trough which is flushed by water from the
hand washing faucets located along the evisceration line.
Hashing and Chilling
The eviscerated carcasses are washed with fresh water in a horizontal immer-
sion tank. The daily water usage averages 2 I/bird. Chilling is accomplished
in two consecutive tanks at a combined daily overflow rate of 2 I/bird which
is equal to USDA requirements. Effluents from both operations are discharged
to the offal flume.
Other Water Uses
The screened feathers and offal are cooked at high pressure and temperature
to produce poultry feed. The steam condenser has once-through water usage
which averages 95 cu. meter of fresh water per day. In addition, about
50 cu. meter of fresh water are used daily to cool the ammonia compressors
which operate continuously to refrigerate the storage coolers and produce
ice used in the chilling operation.
412
-------
ANALYTICAL PROCEDURES
Chemical analyses were performed in accordance with the "EPA Methods" (5).
Samples for trace metal analyses were concentrated according to a method
described by Price (6), and measurements by atomic absorption techniques
were in accordance with "Standard Methods" (7).
Water samples for bacteriological analyses were collected in 100 ml sterile
glass containers, while carcass samples were kept in specially prepared
sterile polyethylene bags. To each bag 0.2 liter distilled water was poured
followed by shaking for 5 minutes. The washing was poured on a sterile funnel
and filtered through sterile No. 1 filter paper. Bacterial counts were
made according to the procedures of "Standard Methods" (7).
CHARACTERISTICS OF POULTRY AND PROCESS EFFLUENTS
The waste loads of all operations presented in Table 2 and Figure 2 were cal-
culated from average flow, concentration and production data. From each
wastewater sample point (Figure 1), six sets of composite samples were
collected during July and twelve sets of grab samples were collected between
April and August. The average values for the twelve grab samples were
generally higher than the average value for the six composite samples. A
summary of the data from the grab samples is presented in Table 3 to charac-
terize the wastewater effluents. The results indicate that the compressor-
cooling water is virtually clean, except for trace contamination by ammonia
leaked from the compression cylinders. This water is suitable for reuse in
the scalding or defeathering operations. The COD for most effluents was
highly correlated to BODs, Total Residue (TR), and Total Volatile Residue
(TVR), which implies the possible use of COD as an indicator of pollution in
poultry processing.
Discharge of the washing and chilling effluents to the offal flume results
in noticeable dilution of the concentrated effluent from the evisceration
process. The mean concentrations of COD and Total Residue are less in
chiller II than in chiller I or in the washer. This offers an opportunity
for reuse of water from chiller II in chiller I and the washer.
Trace metal concentrations in various effluents are shown in Table 4. The
relatively high iron concentration in the killing, scalding and eviscerating
effluents is attributed to the blood contamination and the possible corrosion of
fittings and pipelines at elevated scalding temperatures. Trace metal concen-
trations in the compressor cooling water are nearly identical to those in
the fresh water. In general, concentrations of trace metals in various effluents
are similar to those of fresh water, which indicates the near absence of pollu-
tion by trace metals during poultry processing. Magnesium and iron concentra-
tions exceed the desired levels for drinking water in the U.S.
Antimicrobial residues in the effluents from the scalder, evisceration
trough, washing and chilling tanks were tested according to the method
suggested by Huber (8). The tested samples were either free or contained
antimicrobial residues less than the sensitivity limit of 25 yg/1 of chloram-
phenicol. In addition, four sets of wastewater samples were found free of
parasites, except for the bleeding and killing effluent in which Ascarldia
was detected twice.
-------
240
200
160
$120
t$ 80
40
•
•
•
s
I
\
r
^
[ COD ] [ BOD ]
|^_
I
%
I
--
' ..,., r
PMUKM
lT>
W **J W S3 O O O W *fl M S3" O ' O
h-'(I» >•*• "CO H* H*- 3 (B 0) H- CO H- H'
(D rt CO 31" M I-1 CL 3rt)ft»3 H-fl>fl>3fl>m
31-tHOOI^i-tCO S^^OO^H
00 • (B 00 •
Hv (BMMH Hi (DMM
fl> M (D t-h W (BI-'fDHi M
HlCHlHl- tD HlCHlHl
HiaHlMfD^rtlHI HlSHlMfOd)
H1 (0 H1 • HlHlHl H1 fl) M • H> Hi
• • Hi Hi M • • Hi Hi
I^MB
••MM
i
7////////M//A
^
1
[ Total Residue ]
•Wtt
1^
lla
l^$
W////////M/.
^
1
!ON
",:.
o w *^ tfl s3 o o o
O I-1 fl> < 01 3*" 3*- O
3 (D Bl H- 01 H- H- 3
CL (V ' ft -XO 3* h-> t-> O.
(B O« 3" O H- )— » t— '
Figure 2. WASTE LOADS OF POULTRY PROCESSING OPERATIONS
-------
TABLE 4. TRACE METALS IN GRAB SAMPLES OF WASTEWATER FROM APPP
(ug/D
Sample
_CP_
Pb
_F_e_
R
Final I
Effluent S.D.
R
Bleeding X
Effluent S.D.
R
Scalding X
Effluent S.D.
R
Feather X
Flume S.D.
Evisceration X"
Effluent S.D.
R
Offal X
Flume S.D.
Washing X~
Effluent S.D.
R
Chiller I X
Effluent S.D.
R
Chiller II X
Effluent S.D.
Condenser X"
Effluent S.D.
R
Compressor X
Cooling S.D.
U.S.
Drinking
Water
Standards
*Values exceed
R = Range
2-16
5.8
4.7
1-18
7.0
5.7
1-18
7.1
5.3
2-15
7.4
4.2
1-17
7. 8
5.9
1-17
7.8
5.8
1-17
6.3
6.3
2-19
8.3
5.2
3-18
9.1
4.7
2-14
8.3
3.8
2-18
7.9
4.7
--
drinking
X = Mean
2-17
5.6
5.1
1-13
4.9
4.0
1-12
6.0
4.2
0-12
5.4
4.9
1-8
4.4
3.1
1-14
5.3
4.9
1-12
5.8
4.0
1-8
3.8
2.8
1-6
3.4
1.9
1-9
3.5
3.1
1-15
4.9
4.4
50
34-117
61.8
39.0
25-200
68.9
60.5
21-170
57.4
48.1
22-190
64.3
54.3
31-150
64.5
37. a
28-130
67.5
34.4
28-90
62.0
25.8
21-110
66.4
27.4
30-120
68.4
34.1
26-140
61.5
44.7
32-190
70.1
53.1
—
water standards
S.D. = Standard
1-6
3.0
1.9
1-4
2.6
1.1
1-5
2.6
1.4
2-5
3.4
1.2
1-4
3.1
1.3
1-5
3.0
1.5
1-8
3.5
2.6
1-9
3.6
2.7
1-5
3.1
1.6
1-9
3.7
2.9
1-12
3.1
3.7
—
Deviation
1-12
3.6
3.7
1-12
2.8
4.1
1-14
3.5
4.6
1-12
3.0
3.7
1-14
2.9
4.5
1-22
4.1
7.3
1-12
4.0
4.6
1-10
3.3
3.9
1-9
3.5
3.3
1-8
2.3
2.8
1-14
2.8
4.6
50
n = 8
19-180
72.3*
51.7
21-210
61.4*
66.2
26-210
62.5
62.9
13-180
53.3*
55.0
14-130
53.5*
36.1
13-140
49.5
40.8
12-72
26.7
22.5
10-160
44.0
52.2
10-220
61.4*
86.3
9-140
50.3
58.7
10-190
63.5*
66.3
50
23-60
36.1
11.3
20-40
27.1
7.2
21-40
29.4
7.8
15-40
28.3
7.6
20-40
30.6
6.8
20-40
28.0
8.1
20-40
29.7
6.4
21-180
51.9
53.7
18-32
27.1
6.0
13-100
38.7
35.2
9-40
25.4
12.8
1000
1 30-800
485 *
220
180-630
470 *
165
201-810
524
222
215-640
427 *
132
210-730
466. *
165
310-530
419
68
190-400
311 *
95
90-500
279
142
210-650
374 *
147
240-890
454 *
255
120-600
362 *
155
300
2-3
2.1
0.4
1-4
3.0
1.1
1-6
2.7
1.6
1-4
2.9
0.1
1-5
2.7
1.4
1-4
2.3
1.0
1-2
1.8
0.5
1-3
2.4
0.8
1-3
2.0
0.8
1-5
2.6
1.8
1-3
2.0
0.2
10
415
-------
Bacterial identification and counts of the various effluents are shown in
Table 5 and the counts are illustrated on Figure 3. The highest counts
were found in the killing and eviscerating effluents. Less contamination
was observed in the scalding and feather flume effluents. This may be
due to a reduction of viable-cells from scalding or by dilution with fresh
water.
The decrease in bacterial counts in the offal flume in comparison with the
eviscerating effluent is due to the dilution effect of the washing and
chilling effluents. The most prevalent bacteria in the effluents were
E- coli, Pseudomonas, and Proteus. Salmonella, Para colon, Shigella, and
Staphlococci were identified in several samples. Coolingwater from the com-
pressors was generally free from bacterial contamination.
The results presented in Table 6 and Figure 4 show the identified bacteria
and/or counts associated with the carcass samples. Heavy contamination
was encountered after both scalding and evisceration. Washing decreased
the bacterial counts on the carcasses by 3 to 4 orders of magnitude and
chilling decreased the bacterial counts by 2 additional orders of magnitude.
The bacterial counts increased again during the final weighing and packaging
operations. This striking recontamination indicates the need to modify
the process by eliminating or minimizing the human contact during the
weighing and packaging operations.
The bacteria associated with the carcasses were similar in their order of
prevalence to those found in the wastewater effluents.
ACCUMULATION OF CONTAMINANTS IN THE PROCESS WATER
A study of buildup of contaminants in the scalding, washing, and chilling
tanks was undertaken to provide information concerning the quality of water
which comes in direct contact with the carcasses during processing.
The usual operation involves initial fillings of the tanks every morning
and discharging the used water and refilling with fresh water once or twice
during the processing period. Refilling frequencies depend on the operator's
judgement for the need to renew the process water.
During this study no refilling was done. However, overflow from the tanks
was kept constant according to the rates previously mentioned. Water samples
were taken from the tanks after one hour of operation, and every hour after
that for a total period of six hours. Samples were not collected during the
lunch hour. This sampling scheme was repeated four times in May.
Accumulation of total aerobic and coliform counts in the process water is illus-
trated in Figures 5 and 6, respectively. Bacteria contamination of the scalding
water was greater than in the washing and chilling waters. During the first
processing period (8-11 a.m.) bacterial counts in all tanks increased, and
continued to increase during the second processing period (12-3 p.m.).
The results of this study support the value of continual removal of water
contaminants in these "immersion processes".
416
-------
TABLE 5. BACTERIA IDENTIFICATION AND COUNTS OF WASTEWATER SAMPLES
en
c c
*"" !Tj
I'm
fO 4-*
Final Effluent
Scalding Effluent
Feather Flume
Eviscerating Trough
Offal Flume
Wash Tank
Chiller I
Chiller II
Condenser Water
rrtmnvacerty* Uafar
f Observations
o
A
O
19
18
21
21
21
20
21
21
14
id
.*°
o
E
3 O
0} -0
•*-> :3
o at
£- v>
O- O.
10 57
16 n
23 5
33 33
39 19
33 28
33 57
14 62
28
id
(Percent of Positive Results)
«£ O2 •!-
O
CO to *r~ (O *r"* O
t— •— ^= r— x: o
i— i — CX 1 — Q. «3 O
aj a) >, , r— o
C C +J C +J — —
O •!- O O
E J= E «J E «
f— O. •— i. i— s-
IQ ^) O.
fc. i— -r-
o. o on oo
23 5 19
32 66 77
9,5 9.5 19
5 9.5
19 5 14
28 14 9.5
9.5 9.5 5
19 5-14
7
•r-
0
O
O (J fO
a. o i—
-------
10
12
10
11
io
io
io
10
io5
io
io
io
10
Total Aerobic Count
Coliform Count
[Mean Values!
I
(D »*1 (D W (D C/3 hh hid (DM
TO
3
rt
Hil-1 H, O
Hirt> hh (B
HfD
c D
33
e &
CD H-
33
rtOQ
C
9
fD
HI <
CO
n
ro
h!
HO
M-h
CH,
gfu
fDH-1
n
EF*
H-
L_j
n^
h-
0>
H
M
O
3"
i_j
r^
h-
n>
h(
M
ro o
Hi O
hh 3
c ro
ro 3
3 to
rt ro
i^
€ o
(a O
rt g
t-t ro
n
CO
co
o
Figure 3. BACTERIOLOGICAL COUNTS OF WASTEWATER
418
-------
TABLE 6. BACTERIA IDENTIFICATION AND COUNTS OF PROCESSED CARCASSES
en
c c
i- O
§"•0
IO +>
After Scalding
During Eviscer.
After Eviscer.
After Washing
After Chiller I
After Chiller II
During Weighing
After Packaging
o. of Observations
z:
18
18
12
12
12
18
12
18
scherichia
Oil
tif ^j
72
88.9
83.3
100
100
88.9
100
77.8
3
01
O
Q.
16.6
38.9
25
8.3
33.3
44.4
33.3
27.8
seudomonas
Q.
11.1
50.0
58.3
16.7
41.7
38.9
75.0
66.7
(Percent of Positive Results)
«C oa •!-
o
IO IO •!- IO ••- O ••-
•— i— j= i— .c o u
i— i— CL r— Q. IO O U
O) O >> <1J >t •— O O IO
C C 4-> C +•> i— •(-> O -i-
§"r- O O QJ O. O r—
s: EIOEQ en > IOIOIOIO X: 4-> 'r- O
t/)4-> C/^CLOOQ. (/>
-------
10
13
10
12
10
11
10
10
10
to o
03 in°
Cfl 10
O
S-l
n) -,
10
10
10"
10
Total Aerobic Count
Coliform Count
[ Mean Values ]
w >
n HI
(X> ft
H n>
Cb h{
H-
cw
(D >
P- I-!
CO H-
n o
(D TO
rt
0)
CU Ml
0) rt
S4 fD
H- H
3
00
o >
tf I-h
n >
(D
i-t
SI O
fl) C
H- l-t
TO H-
D4 0
H- (K
TO
3"
TO
rt
(D
h(
Figure 4. BACTERIOLOGICAL COUNTS OF CARCASSES
420
-------
4x10
10
10"
o
u
10
3x10"
7.7 x 10
1.7 x 10
[ Median Values ]
n = 4
SCALDER
WASHINC
TANK
CHILLER
I I f
123
3 56 7
Hours after startup
Figure 5. Accumulation of Total Aerobic Count in Process Water.
421
-------
io5 r
10£
10'
10
2.3 x 10'
[ Median Values ]
n = 4
SCALDER WASHING CHILLER
TANK
CHILLER
i i i I i
I I
I 2 3
23 56 1
6 ' 1 * 3 D 6 ' 1 A 3 3 6 '
Hours after start up
Figure 6. ACCUMULATION OF COLIFORM COUNT IN PROCESS WATER
422
-------
To evaluate the effect of process water on the bacterial quality of the
carcasses, the accumulation study was repeated 7 times during Nov./Dec.
Figure 7 shows parallel deterioration of the bacterial quality of the
carcasses as a direct result of the increased contamination of the process
water.
DISCUSSION
Egyptian health regulations agree in principal with those of the USDA (9),
which requires the use of potable water in all operations involving direct
contact between water and the carcass or giblets. However, in the common
water immersion and water transport operations used in poultry processing
the potable quality is lost when the first carcass or giblet enters the
tank or flume. The addition of fresh water does not regain potable quality
but does cause considerable energy losses in maintaining water temperatures
in the scalder and chillers.
The available option to save process water and conserve energy is water
renovation and reuse. Although used water may not be potable, it can be
treated to a better quality than the process water which is presently used
in contact with the carcasses. Contrary to the popular belief, use of
renovated water can result in improving the quality of the process water (10)
and most likely the poultry products.
Present governmental regulations specify quality and quantities of freshwater
in poultry processing to maintain processing methods which experience has
shown satisfactory to obtain a product safe for consumption. The purpose
of this research project is to demonstrate environmentally desirable modifi-
cations of water use in several unit operations which will maintain or improve
the water quality in these operations and produce a product equally safe
for consumption. Most of the data gathered in phase one has been to establish
a baseline of water and product quality in each unit operation. This data
is to be compared with similar data collected in phase two to document expected
improvements in water and product quality.
MULTIPLE USE SYSTEM
The results of the first phase of the U.S. EPA demonstration project add
support to the premise of water reuse in poultry processing. The multiple
use system shown in Figure 1, has been initiated with the second phase of
this project which began in March 1977.
The system involves use of fresh water in the final chiller and recycling
its overflow after grease separation and bacterial reduction to the chiller
I at a rate of 20-25 1/min. To supplement the need of the chiller I fresh
water will be added at a rate of 12-15 1/min. Counter current flow of all
the water from the final chiller to the chiller I cannot be applied because
of insufficient chilling capacity at the plant. The overflow of chiller I
will be reused with an equivalent flow of fresh water as a continuous feed
to the washing tank. Effluent of the washing tank will be reused in the
feather flume to replace the fresh water used in this operation. Water
used for cooling the compressors will be reused as a continuous feed to
the scalding tank, which will reduce both the fresh water and energy requirements
423
-------
10'
10°
10'
§
o
10"
10
ID-
SCALDER
, »T.C. of process water
&• ®T.C. of carcasses
o—oC.C. of process water
o- «C.C. of carcasses
[ Median Values ]
n - 7
[Counts/Carcass = Counts/ml x 200]
CHILLER II
\ \ \ \ \ \ \
!2 345 67 X23 456 712 345 6 7 123 456 7
Hours since startup
Figure 7. ACCUMULATION OF BACTERIA IN PROCESS WATER AND CARCASSES
424
-------
of the scalding operation. This multiple system is flexible, since the
ratio of fresh water/used water can be adjusted to insure the quality of
the water used in each unit operation.
High pressure spray washers will replace the existing faucets of the evis-
ceration trough to reduce water usage in this area. The trough will be modi-
fied to facilitate movement of the inedible viscera to the evisceration flume
without flushing with water.
Multiple water use together with modifying the. evisceration operation will
save 40% of the current process water usage.
In the final phase of the study, the once through condensing water will be
renovated and recycled continuously to the rendering cookers which will save
an additional 25% of the water usage. Renovation is needed to remove
grease, residues and volatile matter. Reduction of bacteria counts in the
renovated water is not essential for this particular operation.
SUMMARY
The results obtained from this study have provided the following:
1. Manual handling during weighing and packaging results in noticeable
recontamination of the processed carcasses. Elimination of human
contact in this operation is required to upgrade the quality of the
product.
2. Water used for cooling the compressors can be reused in the scalder
without treatment as it is virtually free from contamination.
3. A study of accumulation of contaminants in,the immersion processes
(scalding, washing, and chilling) revealed a direct relationship between
the bacterial counts in the water and on the carcasses. Continual
renovation of water to retard the rate of buildup of contaminants is
needed. Use of renovated water can enhance the quality of the process
water in these operations, besides saving part of the energy required
for cooling and heating processes.
4. Counter current water use through the chillers and washer is planned
after applying renovation measures for grease and bacterial reduction.
5. A comprehensive evaluation of the water use system will be performed
during the second phase of the study.
ACKNOWLEDGMENTS
This research is supported by Grant No. PL 3.542-2 from the EPA. Grateful
acknowledgment is extended to Dr. Ali Mourad, Faculty of Medicine, Alex-
andria University,for his help in the bacterial analyses.
425
-------
REFERENCES
1. Berry L. S., et al. Sand Filtration and Activated Carbon Treatment of
Poultry Process Water. Jour. Water Poll. Control Fed., 48:2394 (1976).
2. Hamza A., et al. Conservation of Water in Food Processing by Use of
Low Volume High Pressure Sprays. The Third National Conference on
Complete Water Reuse, AIChE and EPA, Cincinnati, OH. June 1976.
3. Clise, J. D. Poultry Processing Wastewater Treatment and Reuse.
Envir. Prot. Tech. Series, EPA-660/2-74-060, Washington, DC. 1974.
4. Carawan, R. E., et al. Water and Waste Management in Poultry Pro-
cessing. Env. Prot. Tech. Series, EPA-660/2-74-031, Washington, DC.
1974.
5. Manual of Methods for Chemical Analysis of Water and Wastewater.
Tech. Transfer, EPA-625/6-74-003. 1974.
6. Price, W. J. Effluent and Water Treatment Jour., April 1967.
7. Standard Methods of Examination of Water and Wastewater. 13th Ed.
Amer. Pub. Health Assn., New York, NY. 1971.
8. Huber, W. G., et al. Penicillin and Antimicrobial Residues in Domestic
Animals. Jour. Amer. Vet. Med. Assoc., 154:1590. 1969.
9. Regulations Governing the Inspection of Poultry and Poultry Products.
USDA, Consumer Marketing Service, Poultry Division. 1968.
10. Rogers, C. J. Recycling of Water in Poultry Processing Plants.
Technical Report on Project S-800930, EPA, IERL, Cincinnati, OH.
1976.
426
-------
THE TREATMENT AND DISPOSAL OF WASTEWATER
FROM DAIRY PROCESSING PLANTS
by
James A. Moore* and Boyd M. Buxton**
The concern over environmental quality has brought about the National Pollu-
tion Discharge Elimination System (NPDES), which has set limitations on the
type and quantities of pollutants that may be discharged from the dairy pro-
cessing and other industries. These increased standards have forced policy
makers within the industry to review their present waste-handling systems and
consider alternatives in order to comply with the above standards.
The three alternatives which are open to the processing plants are upgrade or
install a private treatment-disposal system; connect to, or continue to dis-
charge to, a municipal sewage treatment plant, most of which are reassessing.
the cost of treatment; or go out of business. Any of the above choices will
result in an increase in the cost of the product to the ultimate consumer.
The objective of this project was to evaluate the cost of dairy products to
the consumer as influenced by the new water pollution standards. To meet this
objective and obtain this data, the effluent of some typical processing plants
and several commonly-used treatment-disposal systems were monitored. This
paper will report on the activities and results of this monitoring.
An inventory was conducted of the wastewater handling methods used by pro-
cessing plants in Minnesota. It was found that only 271 dairy processing
plants were in operation in November of 1975 compared with 563 in 1965. Most
of the reduction had taken place with the smaller plants and milk-receiving
stations. These transfer or receiving stations numbered 98 and their sole job
was to collect and cool milk for shipment to the processing plant. Bottling
plants number 39 and the remaining 134 plants represent some form of milk pro-
cessing plant.
A break down of the type of waste-handling system was obtained from a mail
survey in which 121 of the 271 plants responded. Over 68 percent of the
plants discharge to a municipal sewage-treatment plant. Just over 16 percent
of the plants utilize private waste-treatment systems, with septic tanks being
the most common. Nearly 9 percent of the plants discharge to a water course
directly without treatment and the remaining 6-plus percent remain unknown.
*Assistant Professor, Department of Agricultural Engineering, University of
Minnesota, St. Paul, Minnesota.
**Agricultural Economist, Economic Research Service, USDA, and Professor,
University of Minnesota, St. Paul, Minnesota.
Approved for publication as Paper No.10002 in the Scientific Journal Series,
Minnesota Agricultural Experiment Station.
427
-------
Most of the 16 percent, or 44 plants, using private waste treatment systems
were visited and evaluated as to their suitability for serving as typical, of
the industry. Six different types of plants were selected as typical plants
with good treatment-disposal systems. These included ridge and furrow, single
stabilization pond, two-cell stabilization pond, trickling filter, package
aeration plant and two-stage aerated lagoon with chemical treatment. While
on these field visits we also saw some of the plants that discharge without
treatment. All of the operations discharging directly without treatment were
small plants and transfer stations outletting into a natural swampy or low
area. While not promoting this practice, it was noted that there appeared to
be no negative effect of the surrounding natural environment.
Ridge and Furrow
The ridge and furrow system was not sampled as there is no system effluent.
The several operations which were visited were functioning well and the man-
agement was well pleased in all cases. When properly designed, the management
requirements are quite minimal. While the necessary size is a function of
water use and soil_type, a recommendation of 5,000 gallons of wastewater per
acre-day (4.67 1/m -day) can be used to design a system (1). This is the
winter rate (half the summer rate) which is recommended for areas which re-
ceive severe winter weather.
Single Stage Lagoon
Some sampling was done of the single-stage lagoon system. However, the dis-
charge to surrounding pasture was never required, so no output data was avail-
able for this system.
Small Transfer Station With Package Aeration Plant
The data gathered in September from the small transfer station (or receiving
station) with the package aeration plant are shown in Table 1. Sampling was
also done in May and March but not included because of space limitations.*
The total volatile solids (TVS) values in the influent (.flow from the plant
into the treatment system) are rather constant and reflect only the milk and
wastewater from the can washer and the bulk tank. The values from the package
plant (effluent) represent the discharge point and have been diluted by the
addition (after the treatment plant) of the cooling water. This accounts for
the low values shown for both TVS and chemical oxygen demand (COD).
The COD values are quite variable flowing from the plant and values jump by
factors of 3 during the day. The data from Table 1 together with the flow
data shown in Figure 1 are needed to get the complete picture of the mass load
flowing into and out of the plant and treatment unit. The high COD values
seen in the late evening represent a carry-over of the wash-down period be-
cause of the very low wastewater flow (almost zero) coming from other than the
cooling water. The total system efficiency for TVS and COD were 27% and 67%,
respectively.
*A11 the data can be obtained upon request from the author.
428
-------
TABLE 1. SMALL TRANSFER STATION WITH A PACKAGE AERATION PLANT -
WASTEWATER ANALYSIS
INFLUENT
23-24 Sept. 75
Time
6:00 a.m.
7:00
8:00
9:00
10:00
11:00
12:00
1:00 p.m.
2:00
3:00
4:00
5:00
6:00
7:00
8:00
9:00
10:00
11:00
12:00
1:00 a.m.
2:00
3:00
4:00
5:00 a.m.
AVE.
Time
2:00 p.m.
3:00
4:00
5:00
6:00
7:00
8:00
9:00
10:00 p.m.
AVE.
TS
mg/1
2540
2240
3460
3600
4060
3400
3940
3840
5790
4760
4040
6240
6520
4680
4060
6680
4240
3700
2860
3820
2960
440
4060
3200
4130
TS
mg/1
1540
1770
1720
1230
1190
1150
820
850
1280
TVS
mg/1
1240
900
1840
2280
2340
1820
2240
2180
3720
2900
2680
4060
4300
2920
2440
2900
2620
2140
1500
2320
1560
2640
2340
1740
2400
TVS
mg/1
700
790
850
370
380
450
80
70
460
COD
mg/1
1617
3823
1260
3206
3289
2704
2633
2041
5178
3548
5836
7594
7229
7347
8689
6668
7947
7586
3979
4286
3438
3634
1825
989
4431
EFFLUENT
COD
mg/1
2080
2590
2881
1610
2100
828
848
322
262
1502
PH
Units
7
7.5
7
7
7
8
7
7
7
6
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
PH
Units
7
7
7
7
7
6.5
6.5
7
7
6.9
Ortho
Phos.
mg/1
216
270
225
258
348
258
371
239
218
290
284
258
318
273
251
246
219
315
294
- 229
194
186
208
202
257
23
Ortho
Phos.
mg/1
70.5
94.0
125.5
83.1
83.6
92.9
125.0
128.9
137.0
104
BOD
mg/1
2420
1183
2121
Sept. 75
BOD
mg/1
792
891
1286
898
891
808
919
352
231
785
429
-------
This unic appeared to be underdesigned for the peak loads it was asked to
handle during the working hours of the plant. The small aeration chamber and
settling tank did not allow the system to benefit from the cyclic loading from
the plant, and solids were carried through when the wash-down process would
occur. This accounts for the low TVS treatment efficiency.
I5O-
100-
UJ
50-
WATER FLOW
SMALL RECEIVING STATION
3/22-3/23 1976
A - 0 *
MID 2 4 6 8 10 NOON 2 4 6 8 10 MID
Figure 1. Water-flow rate to a small receiving station
in March.
Transfer Station With Trickling Filter Treatment System
A larger transfer station was also sampled. This one was using a settling
basin and trickling filter for treatment before discharging to a stream. The
processing plant effluent and treatment system effluent was sampled in Sep-
tember, February and April. The February data are shown in Table 2.
The total solids (TS) data was less steady in this plant than the small trans-
fer station with the TVS and COD values showing wide ranges of concentrations.
The highest values were recorded when the first milk arrived in the morning
and another peak occurred again around 4 p.m., which reflected the plant and
equipment cleanup period. This is shown in Figure 2.
430
-------
TABLE 2.
RECEIVING STATION WITH TRICKLING FILTER TREATMENT
SYSTEM - WASTEWATER ANALYSIS
INFLUENT
4-5 Feb. 76
Time
6:30 a.m.
7:30
8:30
9:30
10:30
11:30
12:30 p.m.
1:30
2:30
3:30
4:30
5:30
6:30
7:30
8:30
9:30
10:30
11:30
12:30 a.m.
1:30
2:30
3:30
4:30
5:30 a.m.
AVE.
TS
mg/1
1840
2220
3320
2800
1800
1630
2010
1860
1600
2900
2370
3130
2860
2750
2470
2010
2010
2130
2120
2000
1710
2030
2540
3260
2307
TVS
mg/1
790
1370
1120
1540
840
830
830
560
850
1650
1120
870
900
840
660
720
690
660
580
630
580
790
1330
1180
914
COD
mg/1
783
1361
1380
3949
1110
757
818
463
988
2374
1517
1002
1034
875
823
636
783
784
776
675
506
623
1040
1280
1097
pH
Units
8.4
6.8
7.1
6.8
6.6
6.6
6.5
6.4
6.5
6.3
6.7
10.4
10.0
9.0
8.8
7.8
7.3
7.8
8.2
7.5
7.1
7.5
8.6
8.3
7.6
Ortho
Phos.
mg/1
105
69
78
65
933
1064
1026
546
277
233
156
59
82
135
111
114
120
119
92
73
129
124
103
137
248
BOD
mg/1
802
927
439
EFFLUENT
4 Feb. 76
Time
8:30 a.m.
9:30
10:30
11:30
2:30 p.m.
TS
mg/1
880
850
950
960
800
TVS
mg/1
260
380
330
430
370
COD
mg/1
89
84
121
146
78
PH
Units
6.8
6.7
6.7
6.7
6.7
Ortho
Phos.
mg/1
66.3
70.2
193.4
170.3
104.5
BOD
mg/1
101
AVE.
888
354
104
6.7
121
431
-------
4000
3000
2000
1000
0
RECEIVING STATION EFFLUENT
2/4-2/5 1976
COD
I
MID
2 4
8 10 NOON 2
TIME IN HRS
6
8
10 MID
Figure 2. Twenty-four hourly values of TVS and COD for
Receiving Station.
The high ortho-phospate values recorded about noon were not found on the other
two sampling periods. The other data were much more consistant and lower,
with the average values being 35 and 18 mg/1. It was suspected that the pri-
mary source of ortho phosphate resulted from hand washing the bulk tanks of
the trucks used to bring the milk to the plant. This is one activity that
can be managed to help reduce the loading to any treatment-disposal system.
Another parameter worthy of note on Table 2 are the pH values. High pH values
were observed about the time of the late-afternoon washup. This load could
be detrimental to a biological treatment system and may require dampening,
diluting or changing the cleanup chemical.
Coupling this information with the data on water use shown in Figure 3 may
give some clue as to the activities in the plant. This plant has a high per-
cent of its flow going to cool the milk. The wash-down water, while quite
variable, was dampened with a rather large base flow of cooling water. The
spikes were due to tank truck and equipment wash down.
432
-------
250-
2OO-
z 150
5
o:
LU
^ IOO -
50-
, 70 -
0 -L Q\LU*
WATER FLOW
RECEIVING STATION
4/19 8 4/20 1976
MID 2 46
10 NOON 2
TIME IN MRS
10 MID
Figure 3. Water-flow rates to a receiving station in
April.
Small Butter Plant With Two-Stage Lagoon Treatment System
A small, one-man butter processing plant utilizing a two-stage lagoon system
proved to be the best waste-treatment system. Typical of processing plants,
the water use was less variable than the preceding transfer stations due to
the rather uniform water use of the various processing component. For a com-
plete breakdown of the water use in the dairy industry, the reader is directed
to Water and Wastewater Management in Dairy Processing, by Carawan, Jones and
Hansen (2).
The flow for the small butter plant is shown in Figure 4 and shows a signifi-
cant percentage of water used for cooling. The greatest water use was during
the working day and was rather uneven. This uneven flow cycle could cause
problems to some treatment systems. However, the lagoon system was quite cap-
able of receiving these daily slug loadings.
433
-------
200-
150-
100-
50-
50
40
30
o 20
10
Q -L 0 I/'/. I I it/ i ii I X
WATER FLOW
SMALL BUTTER PLANT
3/18 a 3/19 1976
HI I I J I /"I
MID 2 46
8 10 NOON 2 4 6 8 10 MID
TIME IN HRS
Figure 4. Water-use rate for a small butter plant.
The pumping station from the plant up to the first lagoon was sampled on a
24-hour basis, four times during the year and the September data are shown in
Table 3. While the flow data was rather uniform during the day, the measured
parameters, with the exception of pH, change quite drastically. The ortho-
phosphate values are quite variable. There is some correlation between the
high values recorded of ortho phosphate and the TS and COD values, but the
pattern was not identical. These September ortho-phosphate values were
higher, with an average of 15.3, and more varied than the other three sam-
plings, which averaged 3.4, 11.1 and 5.5 mg/1 for each of the other days.
Biochemical oxygen demand (BOD) drops about 98%, from 516 to less than 9 mg/1,
after the plant stops operation. The TS and TVS dip to 25% of their peak
values while COD values drop to almost 10% during the 8:30-to-9:30 a.m. sam-
pling periods in the morning. These changes point out the hazards of grab-
sampling wastewater stream flows. The peaks could be even higher as the
hourly sampling represents only a partial picture of the real wastewater con-
centration versus time.
434
-------
TABLE 3.
SMALL BUTTER PLANT WITH TWO-CELL STABILIZATION POND
TREATMENT SYSTEM - WASTEWATER ANALYSIS
INFLUENT
(SAMPLED AT PUMPING STATION)
16-17 Sept. 75
Time
6:30 a.m.
7:30
8:30
9:30
10:30
11:30
12:30 p.m.
1:30
2:30
3:30
4:30
5:30
6:30
7:30
8:30
9:30
10:30
11:30
12:30 a.m.
1:30
2:30
3:30
4:30
5:30 a.m.
AVE.
TS
mg/1
920
7510
2180
500
920
570
1260
320
1110
400
500
420
410
360
340
340
400
300
400
380
360
410
340
340
870
TVS
mg/1
720
6900
1940
380
200
140
480
80
520
130
60
50
140
10
120
120
110
150
140
100
110
180
120
170
540
COD
mg/1
1243
932
3236
349
340
311
89
86
1280
101
133
8
8
5
10
0
0
0
4
4
0
0
0
0
339
pH
Units
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
Ortho
Phos. BOD
mg/1 mg/1
.7
127.2
37.1
2.5
103.0
9.0
41.5
3.6
10.4
.8 516
2.4
17.3
1.4
1.2
1.3
1.2 8.8
.9
1.0
.8
.7
.7
.7
.7
.6
15.3
EFFLUENT
16 Sept. 75
Inflow at
First Lagoon
Flow Between
Lagoons
Outflow
52
20
82 8.1
19
18
45
38
5
3
100
63
7.9
7.8
18
7
64
5
435
-------
The COD values vary more in the September data than in the other three sam-
pling dates. To indicate not only the variability of the sample strength but
also the variability of operation within the day, Figures 5 and 6 are in-
cluded. These show the low values recorded during the shut-down period and
also some spikes of high concentrations occurring throughout the day. In
comparing the two graphs, notice that the peak values came at quite different
times of the day. A complete understanding of all the plant operations is
necessary to identify the cause of the high-strength discharges.
eoooh
sooo -
SMALL BUTTER PLANT EFFLUENT
9/16-9/17 1975
MID 2
IO NOON 2
TIME IN MRS
10 MID
Figure 5. Concentrations of chemical oxygen demand and total
volatile solids in the September effluent of a
small butter plant.
436
-------
3000-
2000 -
SMALL BUTTER PLANT EFFLUENT
1/21 -1/22 1976
COD
o>
1000 -
10 NOON 2
TIME IN HRS
8
10 MID
Figure 6. Concentration of chemical oxygen demand and
total volatile solids in the January effluent
of a small butter plant.
Large Cheese and Butter Plant With a Two-Stage Aerated Lagoon and Chemical
Treatment
The selection of a typical large processing plant with a private treatment
system was rather simple. As most large plants are in cities with central
sewers and a treatment plant, very few have private treatment systems. This
limitation made the selection of a typical plant easy but not necessarily
representative. The large plant selected had just installed a new treatment
system and can be used as a model to others.
The uniform water-use data, shown in Figure 7, is expected for a large,
round-the-clock processing plant. The block steps are from the recordings
and do not reflect sudden changes in the use rate. While a slight increase
was seen through the middle of each day, the pattern was quite uniform.
One would suspect that a large, rather uniform flow would dampen or dilute
high concentrations of various parameters in wastewater analysis. This trend
was true for most of the day as seen from the data in Table 4. The striking
exception is pH. The values range from 1 to above 10 with this full-scale
change taking place between 5:30 and 6:30 in the morning. This would appear
437
-------
to be suspect data, however, the wide range and sudden change was also ob-
served from the samples taken in February. This drastic range must result
from a sudden dumping or washing of one section of the plant processing equip-
ment. This range must be severely detrimental to any biological activity at-
tempting to provide organic removal in the early stages of the system.
WATER FLOW
LARGE CHEESE 8
BUTTER PLANT
8 NOON 4
TIME OF DAY
8
MID
Figure 7. Twenty-four hour water-use data for a large
cheese and butter plant.
A large batch of whey or solids was apparently dumped early on this morning as
the TS and TVS are quite high in the 4:30 a.m. sample. The values are back
to the average concentration in the next hour's sample. The orthoTphosphate
values are quite uniform through the sampling period.
438
-------
TABLE 4.
WASTEWATER ANALYSIS FROM A LARGE CHEESE AND BUTTER PLANT
DISCHARGING TO AN AERATED TWO-STAGE LAGOON WITH CHEMICAL
TREATMENT
INFLUENT
3-4 May 76
Time
5:30 a.m.
6:30
7:30
8:30
9:30
10:30
11:30
12:30 p.m.
1:30
2:30
3:30
4:30
5:30
6:30
7:30
8:30
9:30
10:30
11:30
12:30 a.m.
1:30
2:30
3:30
4:30 a.m.
AVE.
TS
mg/1
4860
3940
5480
3040
2560
3480
3600
3560
4040
3680
2940
2540
3080
4420
4400
1900
2640
6620
4240
8320
10180
22040
5071
TVS
mg/1
2460
2320
3520
1640
1520
1280
2200
1740
1640
1900
1280
1260
1440
2040
2580
880
1440
5080
2580
3560
6860
18360
3072
COD
mg/1
10165
2876
6213
3156
2689
2592
4036
3831
3060
3377
2163
4348
2870
3105
2926
5178
2007
3853
2572
8893
4992
6913
7773
7026
4442
PH
Units
10.5
1.4
5.6
5.6
5.4
4.3
4.6
6.2
9.0
5.4
9.0
5.3
5.4
5.4
5.1
5.1
7.5
5.0
5.1
4.2
5.3
12.0
2.0
4.2
5.8
Ortho
Phos.
mg/1
96
97
184
141
89
767
116
124
149
153
148
117
92
95
130
72
105
177
374
66
255
378
179
BOD
mg/1
2205
3690
5640
3 May
4 May
EFFLUENT
3-4 May 76
TS
mg/1
3000
3130
TVS
mg/1
*"*^-!
740
780
COD
mg/1
1035
954
PH
Units
9-0
7.0
Ortho
Phos.
mg/1
1.5
1.4
BOD
mg/1
1752
1275
439
-------
The effluent samples shown in Table 4 may indicate the need for additional
settling. The TS values are high while the TVS and COD values reflect some
treatment. The BOD values are quite high for effluent and certainly must be
improved to meet current state standards. As this was the first year of oper-
ation, the management of treatment systems will likely improve the quality of
the effluent with time and experience.
SUMMARY
The processing plants' effluent show a wide variation of both water use and
concentration of those parameters measured. A summary of the average data is
shown in Table 5. Reviewing this table for treatment efficiencies, the butter
plant using a two-cell stabilization pond exhibited the highest treatment of
both COD and TS. The package aeration plant treating the wastewater from the
small receiving station had the poorest treatment efficiencies.
This may also support the value of a treatment system which can be operated
effectively with little or no management requirements.
As expected, the summer and fall months do a better job of treatment while the
winter and spring periods reflect the effect of the cold temperatures.
CONCLUSIONS
1. Transfer stations and small package plants have two distinct water uses,
cooling ard wash water.
2. Values of most parameters change sharply, reflecting special activities
within the plant. This points out the danger of grab-sampling dairy
processing plants.
3. Sudden changes in the concentration of some parameters, such as pH, may be
detrimental to biological treatment components.
4. Water-use rates vary widely and nonuniform flow within some operations are
undesirable for package treatment plants, but have little effect on the
efficiency of lagoon systems.
5. The two-cell stabilization pond was the most efficient treatment of TS and
COD.
6. The dairy processing industry can make very good use of treatment systems
which require low management skills. The stabilization pond and ridge and
furrow systems meet this requirement and perform very well in Minnesota.
440
-------
TABLE 5. SUMMARY OF LABORATORY ANALYSIS OF INFLUENT AND EFFLUENT SAMPLES FROM FOUR MINNESOTA DAIRY
PLANTS
Type of Plant Approx.
annual
Date of
sample
milk
received
xlO6 Ibs.
Large cheese and 499.
butter plant using
a two-cell aerated
lagoon
Small butter plant 50.
using a two-cell
stabilization pond
Small receiving 16.
station using a
package aeration
system
Receiving station 91.
trickling filter
Feb
May
Sep
Jan
Feb
May
Sep
Mar
May
Sep
Feb
May
75
76
75
76
76
76
75
76
76
75
76
76
Flow/
day,
1,000
gal.
190.2
187.0
22.0
37.1
22.9
25.8
7.0
7.4
8.4
18.2
16.6
19.8
Chemical Oxygen demand
Influent
kg /day
3,228
3,131
61.1
37.3
32.9
45.9
25.0
16.3
17.1
35.7
69.8
68.5
Effluent
kg /day
164.1
668.7
0.3
2.6
1.9
0.5
0.9
9.1
6.4
14.9
65.3
8.9
mg/1
228
945
4
25
22
5
30
326
202
217
1,040
119
Eff.
%
95
79
99
93
94
99
97
44
62
58
65
87
Total solids
Influent
kg/day
4,354
3,444
137.3
49.7
55.9
57.2
23.8
15.6
16.4
92.7
136.5
92.6
Effluent
kg/ day
1,730
2,061
2.1
8.9
6.5
1.6
6.9
24.9
29.0
53.0
55.8
52.5
mg/1
2,403
2,912
25
86
75
16
260
890
935
770
888
700
Eff.
%
60
40
98
82
97
97
71a/
0^
43
59
43
— At the time samples were taken the concentration of total solids in the effluent exceeded the
concentration in the influent. This emphasizes the variability that can occur in measures of waste
concentration.
-------
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the technical assistance provided to this
project by Mary Paul Even and Jeffrey D. Nielsen, Assistant Scientists, Agri-
cultural Engineering Department, and Steve Ziegler, former Research Specialist,
Department of Agricultural and Applied Economics, University of Minnesota,
St. Paul, Minnesota.
REFERENCES
1. Schraufnagel, F. H. Dairy Waste Disposal by Ridge and Furrow Irrigation.
Proceedings of the 12th Indiana Waste Conference, Purdue University,
Indiana Agricultural Extension Series. May, 1957.
2. Carawan, R. E. , V. A. Jones and A. P. Hansen. Water and Wastewater Man-
agement in Dairy Processing. December, 1972. Department of Food Sci-
ence, North Carolina State University, Raleigh, North Carolina 27607.
442
-------
EIGHTH FOOD WASTE SYMPOSIUM
REGISTRATION LIST
WALTER ADAMS
Sterling Colorado Beef Co.
P.O. Box 1728
Sterling, CO 80751
JAMES AGEE
EPA, Region X
1200 6th Avenue
Seattle, WA 98101
MELVIN D. ALSAGER
J. R. Simplot Co.
P.O. Box 27
Boise, ID 83707
JULIAN B. ANDELMAN
University of Pittsburgh
Graduate School of Public Health
Pittsburgh, PA 15261
JAMES S. ATWELL
Edward C. Jordan Co., Inc.
P.O. Box 7050
Portland, ME 04112
MICHAEL AYRES
Agripac Inc.
P.O. Box 5346
Salem, OR 97304
HAROLD BARNETT
National Marine Fisheries Service
2725 Montiake Blvd. E.
Seattle, WA 98112
MARTHA I. BEACH
N-CON Systems Co., Inc.
308 Main Street
New Rochelle, NY 10801
ARTHUR BENNY
Suite 405, The 400 Building
Bellevue, WA 98004
D. R. BENTLEY
H. D. Fowler Co.
Box 160
Bellevue, WA 98155
E. E. BERKAU
EPA, lERL-Ci
5555 Ridge Avenue
Cincinnati, OH 45268
DON E. BERRYHILL
Del Monte Corporation
P.O. Box 150
Vancouver, WA 98660
TRAVIS BILBREY
O'Neill Meat Co.
P.O. Box 12226
Fresno, CA 93777
JOHN J. BIRDSALL
American Meat Institute
P.O. Box 3556
Washington, DC 20007
PAM A. BISSONNETTE
Kramer, Chin & Mayo, Inc.
1917 Fisst Avenue
Seattle, WA 98101
LEE BODNAR
Protein Products
P.O. Box 328, Springbrook Road
Newberg, OR 97T32
443
-------
JOHN L. BOMBEN
USDA, Agricultural Research Service
Western Regional Research Laboratory
Berkeley, CA 94710
JOHN BONN
Nalley's Fine Foods
3303 South 35th Street
Tacoma, WA 98411
WILLIAM BOON
INPRO Systems Division
Drawer 940
Bock Island, IL 61204
H. BOROW
Standard Brands Ltd.
550 Sherbrooke W.
Montreal, CANADA
G. R. BOWES
Alaska Packers Association, Inc.
P.O. Box 3326
Bellevue, WA 98009
JAMES R. BOYDSTON
EPA, Food and Wood Products Branch
200 SW 35th Street
Corvallis, OR 97330
DON E. BRITTON
National Fruit Canning Co.
P.O. Box 9366
Seattle, WA 98109
GERALD BROWN
Star-Kist Foods
582 Tuna Street
Terminal Island, CA
JOHN W. BUCKLEY
R. W. Beck & Associates
200 Tower Building
Seattle, WA 98155
MAX S. CAMPBELL
H. D. Fowler Co. Inc.
13440 SE 30th
Bellevue, WA 98006
DAVID CARTER
Tyson Foods Inc.
P.O. Drawer E
Springdaie, AR 72764
ROBERT E. CEROSKY
General Foods Corporation
250 North Street ', ,
White Plaines, NY 10625
SOT CHIMONAS
J. R. Simplot Co.
P.O. Box 1059
Caldwell, ID 83605
JIMMIE A. CHITTENDEN
TASCO
P.O. Box 11175
Amarillo, TX 79109
RICHARD D. CHUMNEY
New Jersey Department of Agriculture
P.O. Box 1888
Trenton, NJ 08625
BROOKS D. CHURCH
University of Denver
Department of Biological Sciences
Denver, CO 80210
J. M. CONDIT
Arthur L. Benny & Co.
Suite 405, The 400 Building
Bellevue, WA 98004
JOE COGAR
Stayton Canning Co. Coop
930 Washington Street
Stayton, OR 97383
RON COOK
Stephan Thurlow Co.
Box 9520
Seattle, WA 98109
J. A. CURRY
ARA Corporation
2844 Cascadia Avenue
Seattle, WA 98148
444
-------
GARY W. DAVIS
Brown & Caldwell
1501 N. Broadway
Walnut Creek, CA 94596
ROGER A. DECAMP
National Canners Association
1600 S. Jackson Street
Seattle, WA- 98144
JEFF D. DENIT
EPA, Effluent Guidelines Division
Waterside Mall, 401 M Street, SW
Washington, DC 20460
MARSHALL DICK
EPA, Office of Research & Development
Waterside Mall, 401 M Street, SW
Washington, DC 20460
ROBERT E. DIEHL
Van Camp Sea Food Company
11555 Sorrento Valley Road
San Diego, CA 92121
TED DiNOVO
Battelle Columbus Laboratories
505 King Avenue
Columbus, OH 43201
S. J. A. DOBBERSTEIN
Foremost McKesson, Inc.
One Post Street
San Francisco, CA 94104
KENNETH A. DOSTAL
EPA, Food and Wood Products Branch
200 SW 35th Street
Corvallis, OR 97330
CHARLES M. DOUCETTE
Doucette Consulting Engineers
1900 Donovan Avenue
Bellingham, WA 98225
STEPHEN W. DVORAK
Packer!and Packing Company
P.O. Box 1184
Green Bay, WI 54305
MAURENE W. EHLERS
Western Starch Div., Western Polymer Corp.
P.O. Box 488
Tulelake, CA 96134
LARRY ENNINGA
Pacific Egg and Poultry Association
5420 West Jefferson Boulevard
-Los Angeles, CA 90016
EUGENE E. ERICKSON
Midwest Research Institute
North Star Division
10701 Red Circle Drive
Minnetonka, MN 55343
DAVID B. ERTZ
Edward C. Jordan Co. Inc.
P.O. Box 7050, Downtown Station
Portland, ME 04112
LARRY A. ESVELT
Esvelt Environmental Engineering
E. 7905 Heroy Avenue
Spokane, WA 99206
THOMAS M. ETHEN
N.W. Food Processors Association
2828 SW Corbett
Portland, OR 97212
JOHN FARQUHAR
American Frozen Food Institute
919 18th Street, N.W.
Washington, DC 20006
R. P. FARROW
National Canners Association
1950 Sixth Street
Berkeley, CA 94710
MONTY FERGUS
Protein Products
P.O. Box 328, Springbrook Road
Newberg, OR 97132
JOHN 0. FINNIE
Ministry of the Environment
Pollution Control Branch
Legislative Buildings
Victoria, BC V8V 4S5
445
-------
E. H. FORSHT
EPA, Effluent Guidelines Division
Waterside Mall, 401 M Street, SW
Washington, DC 20460
W. E. GALLAND
WEMCO Division, Envirotech Corp.
P.O. Box 15619
Sacramento, CA 95813
R. A. GALLOP
Food Science Department
University of Manitoba
Winnipeg, Manitoba, Canada R3T 2N2
WAYNE D. GILHAM
Prison Industrial
625 South Industrial Way
Seattle, WA 98108
LANCE GILKEY
Whitney Fidalgo Seafoods
2360 W. Commodore Way
Seattle, WA 98199
JO-ANN GODDARD
Western Washington State College
Environmental Health Department
Bellingham, WA
GEORGE W. GOODWIN
Fort Lewis
Post Engineers DFAE
Ft. Lewis, WA 98433
KEN GRAY
B. F. Goodrich
500 S. Main
Akron, OH 44133
T. R. GREGG
Environmental Marketing Association
3331 N.W. Elmwood
Con/all is, OR 97330
RICHARD W. GREILING
Department :>of Ecology
Southwest Regional Office
Olympia, WA 98504
DEB K. GUHA
H. D. Fowler Co. Inc.
13440 S.E. 30th
Bellevue, WA 98009
C. FRED GURNHAM
Gurnham and Associates, Inc.
223 West Jackson Boulevard
Chicago, IL 60606 sS-
SUZAN A. GUTTORMSEN
Brown and Caldwell
100 W. Harrison
Seattle, WA 98119
AHMED HAMZA
Higher Institute of Public Health
Alexandria University
165 El Horn'a Avenue
Alexandria, Egypt
ARTHUR W. HANSEN
Del Monte Corporation
P.O. Box 3575
San Francisco, CA 94119
JANE HARDING
S.W. Arkansas Education Center
302 W. DeQueen
DeOueen, AR 71832
TOM HARDING
Mountaire Poultry, Inc.
123 W. Park
DeQueen, AR 71832
JAMES HARRIS
Department of Plant Science
and Technology
North Carolina A & T State University
Greensboro, NC 27411
HERBERT H. HART
Snokist Growers
2506 Terrace Heights Road
Yakima, WA 98901
RICHARD C. HEIMSCH
Dept. of Bacteriology & Biochemistry
University of Idaho
Moscow, ID 83843
446
-------
BRIAN W. HEMPHILL
Neptane Microfloc, Inc.
P.O. Box 612
Corvallis, OR 97330
JAMES L. HETRICK
Combustion Engineering Bauer
P.O. Box 722
Dana Point, CA 92629
JOHN H. HETRICK
Dean Foods Company
1126 Kilburn Avenue
Rockford, IL 61101
DANIEL HOBE
Castle & Cooke
50 California Street
San Francisco, CA 94111
DOUG HOLBEGK
Harbor Seafoods Company, Inc.
Box 908
Wrangell, AK 99929
JEFFREY A. HOWARD
Moore, Wallace & Kennedy, Inc.
1915 1st Avenue
Seattle, WA 98101
JERRY HUANG
Eutek, Inc.
1828 Tribute Road, Suite H
Sacramento, CA 95815
MEL JACKSON
University of Idaho
Department of Chemical Engineering
Moscow, ID 83843
DOUGLAS JACOBSON
Hammond, Collier & Wade-Livingstone
4010 Stone Way N.
Seattle, WA 98103
ART JOHNSON
Gray & Osborn, Inc.
P.O. Box 2795
Yakima, WA 98902
DENNIS L. JOHNSON
Swift & Co.
115 W. Jackson Boulevard
Chicago, IL 60604
EUGENE S. JOHNSON
Nalco Chemical Co.
2901 Butterfield Road
Oak Brook, IL 60521
BRIAN KELLY
North Pacific Processors
2155 N. North!ake Way
Seattle, WA 98103
RALPH KENWORTHY
John Inglis Frozen Foods Co.
P.O. Box 3111
Modesto, GA 95353
RIC KERIN
Philip M. Botch & Assoc., Inc.
1021 - 112th N.E.
Bellevue, WA 98004
LLOYD H. KETCHUM, JR.
University of Notre Dame
Department of Civil Engineering
Notre Dame, IN 46556
ALLAN D. KISSAM
Washington Sea Grant
3716 Brooklyn N.E.
Seattle, WA 98155
JAMES K. KOELLIKER
Oregon State University
Agricultural Engineering Department
Corvallis, OR 97331
ARTHUR KOLBERG
Alaska-Shell, Inc.
4215 21st Avenue West
Seattle, WA 98199
E. G. KOMINEK
EIMCO PMD Division, Envirotech Corp,
P.O. Box 300
Salt Lake City, UT 84110
447
-------
R. KUILBOER
Ver Krachtwerktuigen
P.O. Box 165
Amersfoort, Holland
HAROLD KUMMER
Kummer Meat Company
P.O. Box 159
Hlllsboro, OR 97123
CHARLES P. LAND
E. Kahn's Son's Co.
3241 Spring Grove Avenue
Cincinnati, OH 45225
RICHARD LANSDOWN
International Co-op
P.O. Box 1378
Grand Forks, ND 58201
JOHN W. LEE, JR.
CH2M Hill
1500 - 114th Avenue S.E.
Bellevue, WA 98004
EILEEN LEITE
Michigan State University
Department of Food Science
E. Lansing, MI 48824
CARLTON LEWIS
SLAC
800th Street N.W.
Washington, DC 20006
CHI EN LIU
RCL Industries, Inc.
State Pier, P.O. Box 1153
Gloucester, MA 01930
DUGAL R. MacGREGOR
Canada Agriculture
Research Station
Summer!and, BC VOH 1ZO
W. K. McALEER
Peter F. Leftus Coop
Chamber of Commerce Building
Pittsburgh, PA 15243
R. F. McFEETERS
Michigan State University
Department of Food Science and Nutrition
E. Lansing, MI 48823
C
JACK McVAUGH
Envirex Inc.
P.O. Box 1067
Waukesha, WI 53186
ROBERT B. MAGUIRE
Agripac Inc.
P.O. Box 5346
Salem, OR 97304
REGINALD E. MEADE
Anderson - IBEC
19699 Progress Drive
Strongsvilie, OH 44136
J. G. MEENAHAN
John & Anderson, Inc.
P.O. Box 1166
Pontiac, MI 48056
WALTER MERCER
National Canners Association
1950 Sixth Street
Berkeley, CA 94710
SAMUEL P. MEYERS
Louisiana State University
Department of Food Science
Baton Rouge, LA 70803
J. RONALD MINER
Oregon State University
Department of Agricultural Engineering
Con/all is, OR 97331
MICHAEL MINNER
City of Tacoma
818 South Yakima, Suite 202
Tacoma, WA 98405
JERRY MINOR
KCM/Environmental Associates Inc.
535 S.W. 4th Street
Con/all is, OR 97330
448
-------
JAMES C. MITCHELL
O'Neill Meat Co.
P.O. Box 12226
Fresno, CA 93777
FREDERICK L. MONROE
Ore-Ida Foods, Inc.
P.O. Box 10
Ontario, OR 97914
JAMES A, MOORE
Agric. Engineering Department
University of Minnesota
3576 Siems Court
St. Paul, MN 55112
PETER MULYK
Stanley Associates Engineering Ltd
505 - 5th Avenue. S.E.
Calgary, Alberta Canada
ARNOLD M. MUNTER
Jeno's, Inc.
525 Lake Avenue S.
Duluth, MN 55802
CARL E. NALL
Pacific Egg & Poultry Association
5420 W. Jefferson Boulevard
Los Angeles, CA 90016
ROBERT C. NEAL
C. E. Bauer
P.O. Box 968
Springfield, OH 45501
GUY R. NELSON
EPA, Envir. Res. Info. Ctr.
26 W. St. Clair
Cincinnati, OH 45268
RICHARD W. NELSON
National Marine Fisheries Service
2725 Montiake Boulevard E.
Seattle, WA 98112
MORTON NEMIROFF
Castle & Cooke Foods
P.O. Box 3380
Honolulu, HI 96801
HARRY NEUMANN
USDA, Western Regional Research Ctr.
800 Buchanan Street
Albany, CA 94710
L. B. NISLE
J. R. Simplot Co., Food Div.
P.O. Box 130
Burley, ID 83336
JAMES H. DATES
J. R. Simplot Co.
Box 1059
Caldwell, ID 83605
CHANDLER ODELL
City of Tacoma
818 South Yakima, Suite 202
Tacoma, WA 98405
FRANK C. OLMSTED
Vulcan Dehydrators
1038 E. Ft. Lowell Road
Tucson, AZ 85718
CARL OLSON
Stokely-Van Camp, Inc.
P.O. Box 456
Mount Vernon, WA 98273
JERRY ONGERTH
Brown & Caldwell
100 W. Harrison
Seattle, WA 98119
J. G. ORTENGREN
ALWATECH
Oslo, Norway
RICH OTOSKI
Kenics Corporation
4105 S.W. Hocken #201
Beaverton, OR 97005
BRYAN PERKINS
Louisiana State University
Department of iood Science
Baton Rouge, LA 70803
449
-------
ROBERT L. PERRY
Perry Bros., Inc.
500 Terry Avenue N.
Seattle, WA 98105
JOHN B. PETERS
University of Washington
213 Fisheries Center
Seattle, WA 98195
LARRY L. PETERSEN
METRO
410 W. Harrison
Seattle, WA 98119
JOHN P. PILNEY
Midwest Research Institute
North Star Division
10701 Red Circle Drive
Minnetonka, MN 55343
JAMES J. PLAZA
Carborundum
P.O. Box 1269
Knoxville, TN 37901
STEPHEN POLONCSIK
EPA, Region V
230 South Dearborn Street
Chicago, IL 60604
MICHAEL P. PRICE
City of Tacoma
818 South Yakima, Suite 202
Tacoma, WA 98405
ERNEST R. RAMIREZ
Swift & Co.
1919 Swift Drive
Oak Brook, IL 60521
DUANE RASMUSSEN
Jacobs Engineering Co.
2401 Stanwell Drive
Concord, CA 94520
ELDON RICKMAN
P.U.D. Noil of Chelan Co.
327 N. Wenatchee Avenue
Wenatchee, WA 98801
MARTYN J. RIDDLE
Fisheries and Environment Canada
Water Pollution Control Directorate
Ottawa, Ontario K1A OH3
GEORGE H. ROBERTSON
USDA - Western Regional Research Lab,
800 Buchanan
Berkeley, CA 94710
THOMAS C. ROONEY
Rexnord Inc.
5101 W. Beloit Road
Milwaukee, WI 53214
WALTER W. ROSE
National Canners Association
1950 Sixth Street
Berkeley, CA 94710
JOHN ROSENAU
University of Massachusetts
Department of Food Engineering
Amherst, MA 01009
JOHN S. RUPPERSBERGER
EPA, Food and Wood Products Branch
200 S.W. 35th Street
Corvallis, OR 97330
MICKEY ROWE
East Central University
School of Environmental and
Health Sciences
Ada, OK 74820
D. S. SANDFORD
Eco-Research Ltd.
6940 Fisher Road, S.E.
Cllgary, Alta T2H OW3
LARRY W. SASSER
R. W. Beck & Associates
200 Tower Building
Seattle, WA 98101
JOHN E. SCHADE
USDA, Western Regional Research Ctr,
800 Buchanan Street
Albany, CA 94710
450
-------
EDWARD J. SCHAEFER
Bepex Corporation
1021 112th Avenue, N.E.
Bellevue, WA 98004
W. J. SCHELL
Envirogenics Systems Co.
9255 Tel star Avenue
El Monte, CA 91731
W. G. SCHULTZ
USDA, Western Regional Research Ctr.
800 Buchanan Street
Albany, CA 94710
RONALD D. SCINOCCA
Jeno's Inc.
525 Lake Avenue Co.
Duluth, MN 55802
K. L. SIRRINE
R. T. French Company
434 So. Emerson
Shelley, ID 83274
NORMAN W. SMITH
Hallanger Engineers, Inc.
1621 - 114th Avenue, S.W.
Bellevue, WA 98004
JAMES E. SMITH
Envirex Inc.
8300 Rex Road
Pico Rivera, CA 90660
RICHARD M. SMITH
Food Engineering Service
9544 E. Rush Street
So. El Monte, CA 91733
IRV F. SNIDER
Carborundum
P.O. Box 1269
Knoxville, TN 37901
W. W. SOLOMON
Alaska Packers Association, Inc.
P.O. Box 3326
Bellevue, WA 98008
JOHN E. SOMERVILLE
James M. Montgomery Consulting Eggrs.
1301 Vista Avenue
Boise, ID 83705
STEVEN SPANGLER
Van Camp Sea Food Company
11555 Sorrento Valley Road
San Diego, CA 92121
PETER SPECK
Agriculture Canada
Research Station
Summer!and, EC VOH 1ZO
PATRICK M. STANLEY
Safeway Stores, Inc.
425 Madison Street
Oakland, CA 94660
RICHARD W. STERNBERG
EPA, Office of Enforcement
401 M Street, S.W.
Washington, DC 20460
FRED STONE
USDOC - Nat^l. Marine Fisheries Service
2725 Montiake Boulevard
Seattle, WA 98112
JOHN SULLIVAN
BioMed Research Labs
1115 E. Pike
Seattle, WA 98122
MICHAEL D. SWAYNE
SCS Engineers
4014 Long Beach Boulevard
Long Beach, CA 90807
T. T. TAKEOKA
Alaska Packers Association, Inc.
P.O. Box 3326
Bellevue, WA 98009
ROBERT TEDROW
BioMed Research Labs
1115 E. Pike Street
Seattle, WA 98122
451
-------
DONALD J. THIMSEN
General Mills Inc.
P.O. Box 1113
Minneapolis, MN 55440
HAROLD W. THOMPSON
EPA, Food and Wood Products Branch
200 S.W. 35th Street
Corvallis, OR 97330
R. GORDON THOMPSON
Environmental Protection Service
Kapilano 100 Park Royal South
West Vancouver, BC
STEVE THURLOW
Stephan Thurlow Co.
Box 9520
Seattle, WA 98109
PEPPE F. TIANO
Environmental Health
Huxley College/WWSC
Bellingham, WA 98025
CALVIN R. TININENKO
Farmland Foods, Inc.
10700 West Wave!and Avenue
Franklin Park, IL 60131
TED TREPANIER
Hammond, Collier & Wade-Livingstone Assoc,
4010 Stone Way N.
Seattle, WA 98103
ROBERT E. TROWBRIDGE
Ore-Ida Foods, Inc.
P.O. Box 10
Ontario, OR 97914
ROBERT J. UFFEN
Fofct Lewis
Post Engineering DFAE
Ft. Lewis, WA 98433
LANCE VAN BROCKLIN
Vita Food Products, Inc.
P.O. Box 427
Bellingham, WA 98225
RICHARD V, VANCE
Anheuser-Busch, Inc.
721 Pestalozzi Street
St. Louis, MO 63118
452
ROBERT H. VICKERMAN
R. R. Engineering Ltd.
P.O. Box 261
Hardisty, Alberta TOB 1VO
TOM VILLMAN
Stayton Canning Co. Coop
930 Washington Street a
Stayton, OR 97383
r.
GARY WAINWRIGHT
Illini Be£f Packers, Inc.
P.O. Box 245
Geneseo, IL 61254
JOHN F. H. WALKER
Arthur G. McKee
100 S. Riverside Plaza
Chicago, IL 60606
RICHARD A. WALKER
Fluid Systems Division, UOP Inc.
2980 N. Harbor Drive
San Diego, CA 92101
DAVID WELKER
Michael A. Kennedy Consulting Engineers
W. 1625 Fourth Avenue
Spokane, RA 09204
W. JAMES WELLS, JR.
Bell, Galyardt, Wells Inc.
5634 S. 85th Street
Omaha, NE 68127
JENNIFER L. WILKINS
Environmental Health
Huxley College/WWSC
Bellingham, WA 98225
H. KIRK WILLARD
EPA, Food and Wood Products Branch
200 S.W. 35th Street
Con/all is, OR 97330
LOUIS WILLIAMS
Tyson Foods, Inc.
P.O. Drawer E
Springdale, AR 72764
M. K. WINTER
WEMCO Division
Envirotech Corporation
P.O. Box 15619
Sacramento, CA 95813
-------
JACK L. WITHEROW
EPA, Food and Wood Products Branch
200 S.W. 35th Street
Corvallis, OR 97330
KENNETH W. WONG'
Castle & Cooke Foods
P.O. Box 647
Monroe, WA 98272
PING-YI YANG
University of Hawaii at Manoa
Agricultural Engineering Department
3131 Maile Way
Honolulu, HI 96822
453
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-77-184
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
5. REPORT DATE
Proceedings Eighth National Symposium on Food
Processing Wastes
August 1977 issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Food and Wood Products Branch
Industrial Environmental Research Laboratory
Corvallis, Oregon 97330
10. PROGRAM ELEMENT NO.
1BB610
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Lab-Cinti., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF RE.PORT-AND PERIOD COVERED
Symposium Proceedings
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
Additional sponsors include: National Canners Association,
American Meat Institute, Southeastern Poultry and Egg Assoc., Pacific Egg & Poultry
Asso_c_. .__Western States Meat Packers Assoc., NW Food Processors Assoc., Nat'1. Indep.
16. ABSTRACT
Meat Packers Assoc., American Frozen Food Institute
The Proceedings contains copies of 29 of the 31 papers presented at the Eighth
National Symposium on Food Processing Wastes. Subjects included: processing modi-
fications, product and by-product recovery, wastewater treatment, water recycle and
water reuse for several segments of the food processing industry. These segments
included: red meat and poultry, seafood, dairy, fruit, and vegetable.
Attendance at the two and one-half day Symposium was approximately 200 with good
representation by industry, universities, consulting firms, as well as state and
Federal agencies.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
Industrial Wastes, Waste Water, Food
Processing, Byproducts
Process Modifications,
Water Reuse, Water
Recycle
13/B
Release Unlimited
^•^•••••^^•^••••••^•••^•B
EPA Form 2220-1 (9-73)
19. SECURITY CLASS (ThisReportT
Unclassified
21. NO. OF PAGES
462
i. bECURITY CLASS (Thispage)
Unclassified
22. PRICE
*U.S.WHIIWKm PRINTING OfTO 1977-757-056/6537
------- |