WATER POLLUTION CONTROL RESEARCH SERIES
12060—04/70
FIRST
NATIONAL SYMPOSIUM ON FOOD PROCESSING WASTES
APRIL 6-8, 1970 PORTLAND, OREGON
SPONSORED BY:
FWQA Pacific Northwest Water Laboratory
USDA Western Regional Research Laboratory
National Canners Association
Northwest Food Processors Association
U.S. DEPARTMENT OP THE INTERIOR • FEDERAL WATER QUALITY ADMINISTRATION
-------
WATER POLT.UTION CONTROLRESEARCH
The Water Pollution Control Research Reports describe
the results and progress in the control and abatement
of pollution in our Nation's waters. They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Federal Water
Quality Administration, in the U. S. Department of the
Interior, through inhouse research and grants and contracts
with Federal, State, and local agencies, research institu-
tions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Head, Project Reports
System, Planning and Resources Office, Office of Research
and Development, Department of the Interior, Federal Water
Quality Administration, Room 1108, Washington, D. C. 20242.
-------
First National Symposium on Food Processing Wastes
Proceedings
Sponsored by:
Pacific Northwest Water Laboratory
Western Regional Research Laboratory
National Canners Association
Northwest Food Processors Association
April 6-8, 1970
Portland, Oregon
For sale by the Superintendent of Documents, U.S. Government Printing Office
Washington, B.C. 20402 - Price $3.00
-------
FWQA Review Notice
This report has been reviewed by the Federal Water
Quality Administration and approved for publication.
Approval does not signify that the contents necessarily
reflect the views and policies of the Federal Water
Quality Administration, nor does mention of trade
names or commercial products constitute endorsement
or recommendation for use.
-------
FOREWARD
This was the first of a planned series of conferences to discuss
current research on treatment of food processing wastes. This
Symposium was co-sponsored by Pacific Northwest Water Laboratory
of the Federal Water Quality Administration, the Western Regional
Research Laboratory of the U. S. Department of Agriculture, National
Canners Association, and Northwest Food Processors Association.
The first two days of the conference were devoted to an in-depth
review of current research and demonstration projects which have
been funded largely by the Federal Water Quality Administration.
The third day was concerned with discussions of research on
improved in-plant and in-field processing intended to reduce
the quantities of waste needing treatment, and replaced the
Annual Western Regional Research Laboratory's Collaborators
Conference.
-------
CONTENTS
SYMPOSIUM OBJECTIVES 1
James R. Boydston
FEDERAL WATER QUALITY ADMINISTRATION RESEARCH, DEVELOPMENT &
DEMONSTRATION PROGRAM 5
William J. Lacy and H. G. Keeler
STATUS OF R&D EFFORTS ON FOOD PROCESSING WASTES 17
K. A. Dostal and R. J. Burm
STATUS & RESEARCH NEEDS OF POTATO PROCESSING WASTES 27
Kristian Guttormsen and Dale A. Carlson
AEROBIC SECONDARY TREATMENT OF POTATO PROCESSING WASTES .... 39
Glenn A. Richter
USE OF FUNGI IMPERFECTI IN WASTE CONTROL 71
Brooks D. Church and Harold A. Nash
COMBINED TREATMENT OF DOMESTIC AND INDUSTRIAL WASTES BY
ACTIVATED SLUDGE 91
John L. Graham and John W. Filbert
AEROBIC TREATMENT OF LIQUID FRUIT PROCESSING WASTE 119
Larry A. Esvelt
CANNERY WASTE TREATMENT BY TWO-STAGE AERATION PROCESS 145
Leale E. Streebin, George W. Reid, Alan Hu
LIME TREATMENT AND IN-PLANT REUSE OF AN ACTIVATED SLUDGE
PLANT EFFLUENT IN THE CITRUS PROCESSING INDUSTRY 177
Richard H. Jones
SEAFOODS PROCESSING: POLLUTION PROBLEMS AND GUIDELINES FOR
IMPROVEMENT 189
M. R. Soderquist, K. J. Williamson, G. I. Blanton
CANNERY WASTE TREATMENT BY A HIGH SOLIDS ACTIVATED SLUDGE
PROCESS 227
Warren G. Palmer
CONCENTRATION OF SUGARBEET WASTES FOR ECONOMIC TREATMENT
WITH BIOLOGICAL SYSTEMS 261
Ronald W. Brenton -and James H. Fischer
-------
RECONDITIONING AND REUSE OF OLIVE PROCESSING BRINES 281
Jack W. Rails, Walter A. Mercer, Harry J. Maagdenberg
TRICKLING FILTER TREATMENT OF FOOD CANNING WASTE WATER .... 295
Walter W. Rose
FOOD PROCESSING WASTE TREATMENT BY SURFACE FILTRATION .... 311
Louis C. Glide
BIOLOGICAL TREATMENT OF FOOD PROCESSING WASTES 327
Perry L. McCarty
WURDD'S TASK FORCE ON AGRICULTURAL POLLUTION 347
J. Peter Clark
IN-FIELD PROCESSING OF TOMATOES 350
Joseph R. Wagner
"DRY" CAUSTIC PEELING OF VEGETABLES AND FRUITS 355
Robert P. Graham
PILOT PLANT EXPERIENCE OF USDA-MAGNUSON DRY CAUSTIC PEELING
PROCESS 359
Traver J. Smith
POSSIBLE USES OF UNI-FLOW FILTER 362
Karel Popper
A CASE HISTORY IN FOOD PLANT WASTE WATER CONSERVATION AND
PRETREATMENT EXPERIENCE 377
Granville Perkins
ANIMAL FEEDS FROM VEGETABLE WASTES 383
George 0. Kohler
RICE HULL UTILIZATION 387
Milton Staackmann
REGISTRATION LIST 391
-------
SYMPOSIUM OBJECTIVES
by
James R. Boydston*
This meeting beginning today marks a milestone, I think, in the
national program of research and development of water pollution
control methods for the food processing industry. The program
will, I hope, be the first of a continuing series of symposia to
make current information available in an open forum meeting of
this type and to give you an opportunity to question those experts
who are actively developing or demonstrating new or improved treat-
ment methods.
During, or following the course of this meeting, I would like to
have comments from you individually as to your opinion on how often
meetings of this type might be held and still be of value. Also,
we would like suggestions on other possible locations for future
meetings.
This meeting marks a further milestone in that it truly represents a
cooperative, coordinated program between industry and government, to
mutually solve troublesome water pollution problems. This cooperation
is demonstrated by the sponsorship of this meeting. In addition to the
Federal Water Quality Administration, the Government is represented by
the Western Regional Research Laboratory of the U. S. Department of
Agriculture, through Dr. Peter Clark; and industry co-sponsors, the
..National Canners Association by Mr. Walter Mercer, and the Northwest
;Food Processors Association through David Pahl; the Technical Program
Coordinators are Kenneth A. Dostal, Pacific Northwest Water Laboratory,
and Dr. Clark, Western Regional Research Laboratory.
It is timely that a meeting such as this be held to review some of the
latest efforts to reduce water pollution from the food processing
industry. We are all aware of the current public clamor for a cleaner
environment. Those of us who have been working in the field for a number
of years and who have been preaching the gospel of water pollution control
are somewhat surprised at the magnitude of the current tidal wave of
public support for water pollution control. Perhaps one of our new
roles as professionals in the field is to direct these energies into
channels where they will truly accomplish desirable ends. We must avoid
being stampeded into a wave of new plant construction unless we give
careful consideration to the effluent requirements to be met, the
*Chief, Waste Treatment Research Program, Pacific Northwest Water
Laboratory, Federal Water Quality Administration, U. S. Department
of the Interior, 200 S.W. 35th Street, Corvallis, Oregon 97330.
-------
economics of various treatment methods, and the possibilities of by-
product recovery.
The cause of this justifiable national concern has now become obvious
to us all. Our world population has already exceeded the optimum level
that can be supported by our finite quantities of natural resources.
Medical science can reduce death rates through prevention and control
of disease, scientific farming methods can increase the world's food
supply through more efficient growing methods, and our modern technology
can transport men to the moon and back, but we can do nothing to increase
the available quantity of water on this earth. We must, therefore, do
more and more to improve the quality of the used water we return to our
streams and lakes.
As population and demands for water increase, it is obvious that sooner
or later most industries will be faced with requirements for tertiary
treatment. As a result, direct water re-use will become commonplace.
What does all this mean to you as industry representatives? It means
you will be faced with ever higher treatment requirements and larger
expenditures of funds for construction and operation of treatment
facilities. If you listened to the President's Message to Congress
on the environment in February of this year, you heard him say in his
37 point program that he was asking Congress for "extension of Federal-
State water quality-standards to include precise effluent standards for
all industrial and municipal sources". I don't doubt that Congress
will pass the necessary legislation to accomplish the President's goals.
If effluent standards are imposed on industry, I think a word of caution
is in order. It will be easy to be lured into a trap thinking that an
effluent standard set for a particular industrial plant is the final
word and that this requirement will be in effect for all time. Obviously,
with our accelerating population and industrial growth, competition for
water will require that higher effluent standards be imposed and additional
plant construction required, often before the original treatment plant is
broken in. We must remember that standards that are set for a particular
stream are based upon the assimilative capacity of the stream and some
given set of waste loadings, usually the existing waste sources with
some added allowances for future growth. However, when additional popu-
lation and industrial growth occur without a corresponding increase in
the capacity of the stream to absorb the wastes, either processing methods
must be revised, additional treatment must be installed, or plant production
capacities limited.
The President also said that industry must recognize that waste treatment
is a cost of doing business. This statement implies that there will not
be any construction grants program for industry.
Let's consider for a moment where we stand today regarding food processing
waste treatment. Basically, wastes from your industry are not difficult
to treat. Some segments of the industry have some specialized wastes
with difficult treatment problems such as BOD concentrations in excess
-------
of 40,000 mg/& but most of the wastes can be treated by conventional
means. Many of the papers you will hear during this conference will
concern conventional, secondary treatment processes. I am not convinced
that a great deal more work needs to be done on secondary treatment of
food processing wastes. We may spend some additional effort on increas-
ing the efficiency of secondary plants or decreasing the construction
or operating costs, but in most cases we can now define a practical
method of secondary treatment.
Our research goals should include the refinement of conventional methods
of treatment; the development of processes capable of higher degrees
of treatment or completely closed loop systems where the treated
effluent is reused within the processing plant; efforts should be
expanded on processing methods to reduce the quantity of water required
per ton of product; and, finally to develop profitable byproducts from
the wastes resulting from current processing methods.
The role of FWQA in this research effort, as I see it, is more one of
coordinating and guiding a national research effort than it is to embark
on a large scale in-house research program. Our in-house activities
will be geared toward providing technical assistance to you in industry
on waste treatment problems and to demonstrate new techniques for treat-
ing these wastes. Of the Federal research dollar, about 10 percent
is currently expended on in-house research, and 90 percent of the
Federal funds available are spent on research and demonstration grants
and contracts to industry, universities, and consultants for the develop-
ment of new treatment methods. (My colleagues suggest that I refer to
this as extramural research rather than out-house activity.) The current
level of Federal expenditures for this fiscal year in the food process-
ing area is about $3 million, including joint treatment, but we expect
this will decrease in the future.
I see a further major role for FWQA in publishing reports on waste
treatment research and sponsoring additional conferences of this type
to enable you to learn of new developments as early as possible so
that you may benefit from current research in this area of water
pollution control. We recognize that there can be considerable delay
in preparing, publishing, and distributing research reports, and it is
our hope that by conferences like this some of the time delay can be
avoided. In other areas of data dissemination, FWQA headquarters
regularly publishes a summary listing of projects of the Industrial
Pollution Control Branch and the Corvallis Laboratory is now starting
to publish a regular (quarterly) review of the in-house and grant
projects.
The Western Regional Research Lab has a continuing program to develop
better methods of food processing and is directly concerned with methods
which reduce water use. Some industry associations have not only
conducted valuable research of their own, but have been most active in
providing timely information on research developments from all sources
to their members.
-------
In closing, let me say that we are faced with some serious challenges:
we have a deteriorating environment and an aroused public. We know a
few of the answers and a lot of the problems. But, I think that co-
operation such as shown here today between industry and Government
can make the food processing industry a leader in the fight for clean
water.
-------
FEDERAL WATER QUALITY ADMINISTRATION
RESEARCH, DEVELOPMENT AND DEMONSTRATION PROGRAM
by
William J. Lacy and Harold G. Keeler*
Recently, President Nixon signed into law a bill to "maintain conditions
under which man and nature can exist in productive harmony".
Americans today are reacting to the environmental chaos they see around
them. And they are speaking out about foul air. And polluted waters.
And clogged highways. And wasted resources. In general, the public is
convinced—as we all should be — that the technological genius that has
provided abundance and even conquered space ought to be able to do these
things without polluting the environment on which we all depend.
The 1970"s must be the decade when we pay our debt to the part by reclaim-
ing the purity of our environment. It is literally now or never. Com-
missioner David D. Dominick recently remarked, "The threats, both present
and future, to the environment are obvious. Our responsibilities are
equally clear. I am convinced that local, State and Federal water
pollution control programs--with industry's cooperation—can and will
clean up our water.
"If these programs are effectively meshed, coordinated and directed,
the American water clean-up drive will begin to produce substantial
results in the next five to ten years and clean, usable water will be
assured for tomorrow."
The name of this program is unusual—perhaps unique in Government. We
have a Research, Development, and Demonstration Program--a coordinated,
problem-solving program which is dedicated to exploratory research of
new and imaginative pollution control methods; the engineering development
of these methods to solve the practical problems associated with bringing
an "idea" out of the laboratory and into the real world; and, finally,
the demonstration of this new technology to go that extra, normally for-
gotten step of showing the decision-makers that new answers, new tech-
nology have really arrived and are available for use.
In FWQA, we are carrying out a highly mission-oriented research, develop-
ment, and demonstration program. Each of our projects should be respond-
ing to an identified need for an answer. These "needs" are specified
Respectively, Chief, Industrial Pollution Control Branch; and Manager,
Food and Kindred Products; Division of Applied Science and Technology,
Research and Development, Federal Water Quality Administration, Washington,
D. C. 20242.
-------
and assigned priority primarily through input from the non-research
elements of FWQA. In short, we are trying to respond to the needs of
the agency as a whole, not necessarily the litany which the researchers
themselves deem important.
There are really only two major categories of "answers" we are seeking.
First, how do we define the water quality goals to be attained? Second,
how do we reach these goals at least cost? With regard to quality
goals, research is required on the effects of pollution. What are they?
How is the degree of effect related to the amount of pollution? And,
how can we predict the level and type of effect in advance? With this
type of information we can improve and extend the water quality standards
now being established and implemented for the Nation's waters. Simply
knowing what water quality is required is not enough, of course. We
need to develop new and improved means for control of pollution in order
to reduce the cost of pollution abatement to the very minimum possible
in those cases where we already have some ability to control, but beyond
this, our quest is to develop and demonstrate means for controlling that
pollution which today is literally uncontrollable or untreatable at any
cost! Corollary to and, in fact, inseparable from this objective is the
simultaneous upgrading of wastewater quality such that used water may
be reused again—a concept of major significance in extending our
relatively dwindling fresh water supply.
To assist in managing our program and in setting priorities and resource
allocations, a problem-oriented project categorization is utilized (see
Table 1). Eight major categories exist, the first five relate to single-
source-related pollution problems from municipal, industrial, agricultural,
mining, and from "other" sources. The last three categories relate to
problems of a multiple-source nature, i.e., where the answers will be
applicable broadly to many different sources of pollution. In the single-
source category, we are working on such pollution problems as combined
sewer discharges, pulp and paper mill wastes, agricultural runoff, acid
mine drainage, petrochemical wastes, and oil pollution. In the multi-
source categories, we have programs on eutrophication., thermal pollution,
removal of nutrients and refractory organics, and effects of pollutants
on fish and aquatic life.
The mechanisms utilized in carrying out this program are threefold:
1) In-house research and development at seven laboratory
locations plus a number of associated field sites.
2) Contract projects, primarily with industry.
3) Grant projects with universities, industries, States, and
municipalities.
Contract projects are funded entirely with Federal dollars and are
utilized primarily for problem analysis, laboratory investigations, or
pilot-scale research projects which involve a high degree of uncertainty
and which are primarily aimed at determination of feasibility and
-------
SUBPROG RAJ-IS
RESEARCH, DEVELOPMENT
AMD
DEMONSTRATION PROGRAM
Structure of Review Coordinators
Effective date: 4/15/70
11 12 13 Ik 15 16 17 18
MUNICIPAL-
POLLUTIOM
CONTROL
TECHNOLOGY
1101 Foust
Sewered
Wastes
1102 Rosenkranz
Combined
Sewer
Di scharpces
1103 Rosenkranz
Storm Sewer
Discharges
llOli Rosenkranz
Lion- Sewered
Run- off
1105 roust
Hon- Sewered
Municipal
Wastes
1106 Foust
Joint
(Mun./Ind.)
Wastes
IHDUSTRIAL-
P01LUTIOH
CONTROL
TECHNOLOGY
1201 Lacy
Metal and Metal
Products
1202 Lacy
Chemicals and
Allied Products
1203 Lacy
Power Production
12Qk Lacy
Paper and Allied
Products
1205 Lacy
Petroleum and
Coal Products
1206 Lacy
Food and
Kindred
Products
1207 Lacy
Machinery and
Transportation
Eauiranent
1208 Lacy
Stone , Clay and
Glass Products
1209 Lacy
Textile Mill
Products
1210 Lacy
Lumber and Wood
Products
1211 Lacy
Rubber and
Plastic
1212 Lacy
Miscellaneous
Industrial
Sources
1213 Lacy
Joint
(Ind./Mun.)
Wastes
AGRICULTURAL-
POLLUTION
CONTROL
TECHNOLOGY
1301 Bernard
forestry and
Logging
1302 Barnard
Rural Run-off
1303 Bernard
Irrigation
Return Flows
130U Bernard
Anireal Feed
Lots
13C5 Foust
'Non-Sewered
Rural Wastes
MINING-
POLLUTION
CONTROL
TECHHOLOaY
1U01 Hall
Mine
Drainaee
11*02 Hall
Oil
Production
1U03 Hall
Oil Shale
1^04 Hall
Other
Mining
Sources
1^05 Hall
Phosphate
Mining
OTHER- SOURCES
OF- POLLUTION
CONTROL
TECHNOLOGY
1501 Bernard
Recreational
1502 Bernard
Watercraft
Wastes
1503 Hall
Construction
Projects
150k Hall
Impoundments
1505 Bernard
Salt Water
Intrusion
1506 Bernard
Natural
Pollution
15C7 Hall
Dredging
1508 Bernard
Oil Pollution
1509 Bernard
Hazardous
Material
Spills
WATER
QUALITY
CONTROL
TECHNOLOGY
1601 Joseph
Eutronhication
1602 Forziati
Physical-Chemical
Identification of
Pollutants
1603 Forziati
Biological
Identification
of Pollutants
l6ch Forziati
Source of
Pollutants
1605 Joseph
Fate of Pollution,
in Surface Caters
1666 Joseph
Fate of Pollutants
in Ground Waters
1607 Joseph
Fate of Pollutants
in Coastal Waters
1608 Fisher
Water Quality
Control
1609 Hall
Water Resources.
Data
1610 Fisher
Cold Climate
Research
1611 Hall
Water Resources
Planning
1613 Joseph
Thermal Pollution
WASTE TREAT-
MENT & ULTI-
MATE DISPOSAI
TECHNOLOGY
1701 Foust
Dis solved
Nutrient
Removal
1702 Foust
Dissolved
Refractory
Organic s
Removal
1703 Foust
Suspended and
Colloidal
Solids
Removal
170k Foust
Dissolved
Inorganics
Removal
1705 Foust
Dissolved
Biodegradable
Organics
Removal
1706 Foust
Microorganisms
Removal
1707 Foust
Ultimate
Disposal
1708 Foust
Waste Water
Renovation
and Re-use
1709 Foust
Waste
Treatment
Optimization
WATER
QUALITY
REQJJIREMEKTS
RESEARCH
1801 Allen
Municipal Uses
1802 Allen
Industrial Uses
1803 Allen
Agricultural
Uses
l80k Alle_£
Recreational
Uses
1805 Allen
Fish and Other
Aquatic Life
1806 Allen
Marine Industrial
Uses
1807 Allen.
Marine
Recreational Uses
1808 Allen
Marine Fish and
other Aq_uated
Life
Table 1. Research, Development and Demonstration Program
-------
development of design requirements. These are not the types of projects
that municipalities and private corporations will often sponsor or
co-sponsor with their own funds due to the large degree of risk involved.
The work performed under contracts often requires highly specialized
personnel and equipment and facilities having a high value over a short
period of time, but limited value in the long-term.
Notable examples of contract projects pertaining to the petroleum-
petrochemical industry include: (1) a five-year cost projection of
pollution control facilities required by specific chemical industries,
(2) a state-of-the-art report concerning petrochemical wastes to serve
as a basis for future program planning, and (3) a presentation of the
many components which constitute a preliminary wastewater treatability
study with specific reference to petrochemical and refinery wastewaters.
Grant projects require some level of matching support from the grantee.
Grants are employed in meeting objectives where it is desirable to
utilize State, municipal, academic, or industrial talents and expertise
in carrying out research, development, or demonstration efforts on a
cost-sharing basis resulting in mutual benefit to both the Federal
Government and the grantee. Demonstration projects are particularly
significant in that they permit us to carry research and development
findings on into the full-scale phase, thereby showing what can be
accomplished through use of new technology and at what cost.
Industry's pollution problems are diverse as its products (see Table 2).
From food processing to power plants, from oil refineries to coal mines,
the pollution problems vary greatly. Some industrial wastes are organic
and lend themselves to the same biological treatment long used for
municipal sewage; others are inorganic, requiring chemical or physical
treatment.
Frequently, organic wastes from food processing and pulp and paper mill
industries can be treated by biological processes. In other cases, such
as wastes from metal finishing, chemical, petroleum, and steel industries,
different treatment methods are necessary (see Table 3). In short, each
industry group has its particular circumstances which require its own
technical and economic solutions.
Today's industrial water pollution problems are magnified by the older
plants and factories many of which are marginal enterprises and tend to
retard our efforts.
Dr. David G. Stephan, the Assistant Commissioner for Research and Develop-
ment, recently stated, "Our long-range objective in the FWQA research
program is to provide the necessary treatment systems for attaining any
degree of wastewater purification which may be required at the minimum
feasible cost." The word "systems" was used in this objective. Rarely
will any advanced waste treatment process be used as the sole technique
in the purification of wastewater for reuse. Generally, systems of
selected unit processes in series or in parallel will be required to
8
-------
Table 2. Standard Industrial Classification of Industries
of Significance for Water Pollution
CODE
FOOD AND KINDRED PRODUCTS
201 Meat products
2011 Meat slaughtering plants
2013 Meat processing plants
2015 Poultry dressing plants
202 Dairies
2021 Creamery butter
2022 Natural and process cheese
2023 Condensed and evaporated milk
2026 Fluid milk
203 Canned and frozen foods
2033 Canned fruits and vegetables
2034 Dehydrated foods products
2035 Pickles, sauces, salad dressings
2037 Frozen fruits and vegetables
204 Grain mills
2.041 Flour mills
2043 Cereal preparations
2046 Wet corn milling
205 Bakery products
206 Sugar
207 Candy and related products
208 Beverages
2082 Malt liquors
2084 Wines and brandy
2085 Distilled liquor
2086 Soft drinks
209 Miscellaneous foods and kindred products
2091 Cottonseed oil mills
2092 Soybean oil mills
2094 Animal and marine fats and oils
2096 Shortening and cooking oils
-------
Table 2. Standard Industrial Classification of Industries
of Significance for Water Pollution (continued)
CODE
TEXTILE MILL PRODUCTS
2211 Weaving mills, cotton
2221 Weaving mills, synthetic
2231 Weaving, finishing mills, wool
225 Knitting mills
226 Textile finishing, except wool
228 Yarn and thread mills
229 Miscellaneous textile goods
24 LUMBER AND WOOD PRODUCTS
242 Sawmills and planing mills
2421 Sawmills and planing mills
26 PAPER AND ALLIED PRODUCTS
2611 Pulp mills
2621 Paper mills, except building
2631 Paperboard mills
264 Paper and paperboard products
265 Paperboard containers and boxes
2661 Building paper and board mills
28 CHEMICALS AND ALLIED PRODUCTS
281 Basic chemicals
2812 Alkalies and chlorine
2818 Organic chemicals, n.e.c.
2819 Inorganic chemicals, n.e.c.
282 Fibers, plastics, rubbers
2821 Plastics materials & resins
2823 Cellulosic man-made fibers
2824 Organic fibers, noncellulosic
283 Drugs
284 Cleaning and toilet goods
2851 Paints and allied products
2861 Gum and wood chemicals
287 Agricultural chemicals
289 Miscellaneous chemical products
29 PETROLEUM AND COAL PRODUCTS
2911 Petroleum refining
295 Paving and roofing materials
10
-------
Table 2. Standard Industrial Classification of Industries^
of Significance for Water Pollution (continued)
CODE
30 RUBBER AND PLASTICS PRODUCTS, n.e.c.
3069 Rubber products, n.e.c.
3079 Plastics products, n.e.c.
31 LEATHER AND LEATHER PRODUCTS
3111 Leather tanning and finishing
32 STONE, CLAY, AND GLASS PRODUCTS
3211 Flat glass
3341 Cement, hydraulic
325 Structural clay products
326 Pottery and related products
327 Concrete and plaster products
3281 Cut stone and stone products
329 Nonmetallic mineral products
33 PRIMARY METAL INDUSTRIES
331 Steel rolling and finishing
332 Iron and steel foundries
333 Primary nonferrous metal
3341 Secondary nonferrous metals
11
-------
Table 3. Industrial Source versus
Significant Constituent Pollution
Sector
Metal and metal products
Chemicals and allied
products
Power production
Paper and allied products
Petroleum and coal
products
Food and kindred
products
Machinery and
transportation
equipment
Stone, clay and glass
products
Textile mill products
Lumber and wood
products
Rubber and plastic
__ Constituent
]JOD SS TDS Acid Heavy
Metals
X
X
X
X
Btu
X X
X X
X
X
X
X
X
X
X
X
X
X
X
12
-------
meet the various reuse criteria. Included will be one or more processes
to provide for the permanent non-pollutional disposal of the inevitable
sludges and blow-down streams generated by the separation process used.
From the systems and model formulation viewpoint, the industrial pollu-
tion research priorities area is the critical place to start. Industrial
pollution as a class accounts for the largest share of the Nation's
water pollution problem. For example, the industrial wastewater volume
is 2.5 times that of the sewered population; the "five-day" BOD of
industry is 3 times; and the suspended solids over twice that of the
sewered population. Further, given the highly correlated co-location of
major polluting industry groups with population, industrial water pollu-
tion has a greater impact on the population than would be expected. In
addition, to further complicate the industrial pollution problem, in
recent years while the major polluting industries have grown essentially
at triple the rate of population growth, it appears that in many areas
the proportion of untreated industrial waste has been growing even more
rapidly. In any event, the magnitude of the problem whether based on
Industry Profiles estimates or Census-Municipal projections, is measured
in dollar terms ranging from one to two billion dollars a year in annual
outlays in the early 1970's.
If the primary job of FWQA is to get all water pollution sources to
install at least a secondary level of treatment, then the long-term task
would be how to handle the sludge disposal problems.
That wastewater renovation and water reuse are technically feasible is
assured at this point. The major challenge ahead relates to evaluation
and development of alternative separation and disposal processes which
show promise of improved performance at reduced costs.
Renovated wastewater for reuse would alleviate simultaneously two of
our Nation's major water resource problems--water pollution and water
supply.
This sounds great, but what will this solution cost? Yes, cost—now,
today. The average cost of water delivered to your sink, bathroom,
kitchen, lawn, etc. is less than $.08/ton. Milk costs about $290/ton
delivered to your door; and even the cheapest soil costs $3/ton dumped
on your lawn and topsoil costs $25/ton.
Our in-house activity forms the foundation for an effective over-all
program. Our in-house researchers must establish objectives and plans
of attack; they must review and evaluate the many, many project pro-
posals received by this agency; allocate the available funds; and
integrate the results of these efforts into usable technology. We
believe they must be contributing creative input of their own. In short,
we believe a strong in-house program is an essential foundation for an
effective grant and contract effort.
13
-------
The research and development program also involves the assignment of
specific areas of technical responsibility to each of our laboratories.
Each laboratory functions as a national focal point for research on a
given set of problems, and in this way duplication of facilities, staff,
and effort among the various laboratories is, avoided. These research
laboratories are located at Cincinnati, Ohio; Athens, Georgia; Ada,
Oklahoma; Corvallis, Oregon; College, Alaska; Duluth, Minnesota; West
Kingston, Rhode Island; and Edison, New Jersey.
Laboratory research assignments deal with the experimental application
of technology for treatment, control, or prevention of pollution from
single-source contributors. Three laboratories have "Treatment and
Control Research" assignments related to the Food and Kindred Products
industry: Robert S. Kerr Water Research Center, at Ada; Southeast
Water Laboratory, at Athens; and Pacific Northwest Water Laboratory, at
Corvallis.
All FWQA laboratories have certain research assignments from which
research findings emerge that have potential application to wastes from
the food and kindred products area. The breakdown of the industries of
significance for water pollution research programs and the waste genera-
tion coefficients in pounds BOD/employee are shown on Table 4. All the
laboratories are involved with the development of water quality criteria.
Primary effort at these laboratories is centered on developing quality
requirements necessary for propagation of various fish and aquatic life.
Special emphasis is now being placed upon temperature and oxygen criteria
because they are important to aquatic life and are commonly altered by
many types of waste discharges. But attention is also being given to
ways of establishing water quality criteria for complex industrial wastes
in order to provide improved goals for waste treatment research and
plant design and performance.
As to the future, research, development, as well as demonstration efforts
will be accelerated in the industrial and agricultural pollution control
areas. With specific reference to food processing and canning wastes, a
great deal of emphasis will be placed on the development of parameters
to measure their pollutional effects, including improved identification
techniques. Increased emphasis will be placed on the demonstration of
municipal pollution control technology, particularly in joint treatment
with industry and nutrient removal. The emphasis in our eutrophication
research will be shifting from fundamental research into the demonstration
of corrective measures for eutrophication control and lake restoration.
Also to be demonstrated under our future program will be some 'of the
unit operations not particularly common to the food industry, i.e.,
combined chemical-biological treatment, solvent extraction, ion exchange,
electrodialysis, chemical oxidation-reduction, reverse osmosis, freezing,
microscreening, airstripping, fluidized bed, and ozonation.
14
-------
Table 4. Waste Generation Coefficients
(After Bower, RFF)
Standard Industrial
Classification Sector
2011 Meat packing
2026 Fluid milk
2033 Canned fruits and vegetables
2037 Frozen fruits and vegetables
2061 Raw cane sugar
2062 Cane sugar refining
2085 Distilled liquor
2261 Finishing plants, cotton
2262 Finishing plants, synthetic
2621 Paper mills
2631 Paperboard mills
2641 Paper coating and glazing
2654 Sanitary food containers
2911 Petroleum refining
3111 Leather tanning
Pounds BOD/Employee
4596
1770
16200
24450
3336
3420
3000
33200
11580
28960
45200
4260
4500
8400
14700
15
-------
We anticipate a continual increase in the importance of research on
Water Quality Control technology, i.e., the development of non-waste-
treatment methods for preventing or controlling pollution. It cannot
be overemphasized that a very major program is anticipated in this area
as we move into the middle 70's. Unquestionably, thermal pollution will
continue to be of growing importance and interest.
In waste treatment and ultimate disposal, major emphasis will be placed
on methods of non-pollutional disposal of sludge and concentrates. This
is of great importance and one which will receive increasing emphasis.
Other areas in waste treatment where increased emphasis can be expected
are in dissolved nutrient removal, wastewater renovation and reuse, and
in microorganism removal, particularly viruses.
Water quality requirements, i.e., the effects of pollution, will continue
to be funded at higher levels. As stated above, this information is
essential to the extension and improvement of the Nation's water quality
standards. Because of the tremendous number of new compounds being
synthesized and finding their way into our environment each year,
intensive research investigations must be conducted to develop a pre-
dictive capability allowing us to project the potential pollutional
impact of these compounds in advance.
The challenge of the 70"s is tremendous. This will truly be, as the
President said, the "environmental decade." Research must play its role--
and research must succeed! The answers we seek are not just "interesting";
they are not just to satisfy intellectual curiosity; they may well be
essential to our survival.
Recently, there has been much to-do in all our news media on the sources
of pollution and the problems of ecology. I would like to end this
with a quote from the comic strip by Walt Kelly regarding who is respon-
sible for our contaminated environment. It was Pogo who said, "We have
met the enemy and they are us."
16
-------
STATUS OF R&D EFFORTS ON FOOD PROCESSING WASTES
by
Kenneth A. Dostal and Robert J. Burm*
Three FWQA laboratories have national research assignments related to
the food industry. The Robert S. Kerr Water Research Center in Ada,
Oklahoma, is concerned with wastes from meat packing. This effort
is being handled under the direction of Mr. Jack Witherow. The South-
east Water Laboratory in Athens, Georgia, has been assigned the citrus
and poultry processing areas under the direction of Dr. H. Page Nicholson.
The remaining food and kindred product areas are assigned to the Pacific
Northwest Water Laboratory in Corvallis, Oregon, within the Waste Treat-
ment Research Program. Mr. Boydston is Chief of this program which also
handles the R&D responsibilities for pulp and paper processing.
Although our organization has had rather severe manpower and monetary
restrictions since its beginning, the Corvallis Laboratory has partici-
pated in some limited field studies. Our first such effort involved
the monitoring of a full-scale aerated lagoon which is used to treat
wastes primarily from the processing of peas. Copies of a report(1)
on this study are available. Figure 1 shows a flow diagram of this
facility which was located in Ferndale, Washington. The aerated lagoon
with a surface area of 1.75 acres contained about 5.6 million gallons.
Each of the four platform-mounted surface aerators at 50 horsepower
had a rated capacity of 3,000 pounds of oxygen per day. Effluent from
the lagoon flowed through a polishing pond with a volume of 135,000
gallons. During the study period the waste flow averaged about 1 mgd
with a BOD load of about 7,000 pounds per day.
In general, the results showed a BOD reduction of 78 percent, from
820 to 180 mg/Jl. Of the 180 mg/£ in the effluent, over 90 percent
was associated with the suspended solids. The polishing pond filled
very rapidly and since there was no solids removal from it the suspended
solids passed straight through. The effluent from the polishing pond
contained an average suspended solids concentration of 520 mg/Jl as
compared to 340 mg/5, in the feed to the aerated lagoon. With a more
positive method of solids removal the system could remove over 90 percent
of the BOD.
Another study on secondary treatment of potato processing wastes was
conducted jointly by the Potato Processors of Idaho, the Idaho State
*Respectively, Chief, Food Wastes Research Branch, and Sanitary Engineer,
Food Wastes Research Branch, Waste Treatment Research Program, Pacific
Northwest Water Laboratory, Federal Water Quality Administration, U. S.
Department of Interior, 200 S.W. 35th St., Corvallis, Oregon 97330.
17
-------
Aerated Lagoon
4-50 hp Aerators
O
o
Lagoon
Influent
PolisninS Pond
Effluent to
Nooksack River
Figure 1. INDUSTRIAL WASTE TREATMENT FACILITIES
FERNDALE, WASHINGTON
18
-------
Health Department, and the Corvallis Laboratory. A report(2) on this
two-year study has been prepared and distributed.
The objective of this work was the evaluation of anaerobic and aerobic
lagoons as potential secondary treatment processes. Three pilot lagoons
(Figure 2) were used; each was 40 feet square and about 10 feet deep.
The first pond contained a 10 horsepower surface aerator, and was fed
primary clarifier effluent at various rates. Anaerobic pond II was
covered with three inch thick styrofoam. A pump was used to maintain
the solids in suspension. As with pond I, this pond was also fed
primary clarifier effluent.
Effluent from pond II was pumped to the third pond which contained a
5 horsepower floating aerator. Various hydraulic loadings were investi-
gated such that the detention time in the first pond was always equal
to the sum of the detention times in ponds II and III. Since a wide
range in loadings and detention times were investigated, a wide range
in results was obtained. It was demonstrated that a BOD reduction of
greater than 90 percent could be obtained from either primary clarifi-
cation plus an aerated lagoon or primary clarification plus anaerobic-
aerobic lagoons in series. This was accomplished without pH adjustment
or inorganic nutrient addition even though the processing plant used
lye peeling of the potatoes. These results were instrumental, at least
in part, for the design of two full-scale secondary treatment systems.
Last fall two pilot plants were operated at a cannery in Salem, Oregon.
A tank 25 feet in diameter and 10 feet deep containing a 1 horsepower
floating surface aerator was used as one pilot plant. The tank contained
a device called a tube settler which is a system for high-rate removal
of suspended solids. The objective of this study was to see if the tank
could be used as both an aeration tank and an aerobic digester, thereby
eliminating the need for separate sludge handling facilities.
The oxygenation capacity was limited due to the 1 horsepower aerator,
thus the organic loads which could be imposed on the system were
restricted. In addition, difficulties in maintaining the feed rate
were experienced due to heavy silt loads during processing of certain
root crops which either wore out pumps or clogged feed lines.
Influent BOD concentration ranged from 260 mg/£ to over 1600 mg/& which
resulted in loadings of 3 to 12 pounds of BOD per day per 1000 cubic
feet of volume. Percentage removals were generally in the 80 to 95
percent range. At this time it is difficult to say how much added
benefit could be attributed to the tube settler as a result of the
problems encountered. Hopefully, this will be answered during the
next processing season if the manpower situation is such that the
pilot plant can be operated for one more season.
19
-------
t
Pump
Pond I
O
10 hp
Aerator
Pond II
Covered
Anaerobic
e
1
Pond III
O
5 hp
Aerator
To River
Figure 2. FLOW DIAGRAM OF PILOT PLANT
20
-------
The second pilot plant located at the cannery in Salem is called a
Rotating Biological Contactor or RBC Unit. It consists of three units:
a primary clarifier, the contactor, and a secondary clarifier. The
contactor is a tank 14 inches deep, 18 inches wide, and 12 feet long
containing 10 rows with 32 discs on each row. The 1 foot diameter discs
are spaced at one-half inch intervals. Both the direction and the speed
of rotation are controlled by valves and a hydraulic drive system.
As the discs are rotated about one-half submerged in the liquid, a
biological film develops on the surface of the discs. This attached
film absorbs and oxidizes the dissolved organics. In addition, the
disc rotation aerates the liquid as it passes through the contactor.
It is reported(3) that this type of process can be loaded much higher
per unit volume than other biological systems due to the higher quantity
of biological mass that is maintained in the system.
A combination of late arrival of equipment plus electrical and mechanical
problems resulted in no reasonable length of operation of the pilot plant.
It takes several days to build up a biological film on the discs. Almost
continuous shearing of pins on the shafts containing the discs was
experienced and this caused the biological film to slough. Thus, the
unit was nearly always in the startup phase. It is hoped these problems
have been solved and another season will result in some useful data.
Nearly all of the R&D effort within FWQA in the food processing field
is in the grant program. At the present time (April 1970) there are
45 to 50 grant projects in various stages of completion. FWQA has
contributed about $9 million to these studies which have total estimated
project costs in excess of $24 million. Approximately three-fourths
of these grants are in the area of responsibility assigned to the Corvallis
Laboratory.
Twelve of these grant projects were reported on at the Symposium and
the papers are in these Proceedings. In addition to these, a brief
description of several of the other ongoing projects follows.
In addition to the two state-of-the-art summaries reported herein, three
others have been awarded. The Beet Sugar Development Foundation located
in Fort Collins, Colorado, is nearly finished with a similar document
on the sugarbeet processing industry.
Ohio State University is preparing a state-of-the-art paper on the
dairy industry and National Canners Association is handling the canned
and frozen fruit and vegetable processing industries.
In the near future a contract will be awarded to cover the beverage
industry—both alcoholic and non-alcoholic.
21
-------
In addition to the state-of-the-art grant on the dairy industry there
are five other R&D grants related to this industry.
Dairy Research and Development Corporation, New York, has a grant for
development and full-scale demonstration of a process for the conversion
of dairy whey into saleable food products. A dry powder will be pro-
duced from acid whey by evaporation and spray drying methods. The two
year project is scheduled for completion by December 1970.
Crowley's Milk Company, Inc., of Binghampton, New York, has a 30-month
grant to develop and demonstrate an ultrafiltration plant for abate-
ment of pollution from cottage cheese whey. A two-stage ultrafiltra-
tion system with a capacity of 10,000 pounds per day will be designed,
operated, and evaluated for separation and concentration of protein and
lactose or for straight acid whey concentration. Phase II will include
the demonstration of a full-scale (250,000 pounds per day) system.
Aerobic biological treatment of cheese processing rinse waters will be
demonstrated by Kent Cheese Company, Kent, Illinois. Two lagoons will
be operated in series, each with a hydraulic detention time of about
30 days. Oxygen will be supplied through submerged, 18-inch in diameter,
aeration tubes by a compressor.
RAI Research Corporation had a small grant to investigate electrochemical
oxidation of milk whey on a laboratory scale. Although it was shown that
it is technically feasible to treat whey wastes in this fashion, the
power consumption is high and the process is economically unattractive
except for possibly the smaller plants. This work has been completed and
the final report should be available in the near future.
Bench and pilot plant studies on secondary treatment of domestic sewage
plus wastes from the manufacture of cheese and associated dairy products
are underway as a result of a grant to the Village of Walton, New York.
Both the trickling filter and activated sludge processes are being inves-
tigated along with methods aimed at improving sludge dewatering.
Seven grants currently underway are concerned with various aspects of
meat slaughtering and processing.
Beefland International, Inc., Council Bluffs, Iowa, has a grant titled
"Elimination of Water Pollution by Packing House Animal Paunch and Blood".
This 18-month project will demonstrate the economic and technical feasi-
bility of completely segregating blood and paunch from slaughterhouse
operations and converting these materials into animal feed ingredients.
Two dehydrators will be installed and utilized to process the material
generated from anticipated cattle kills of 250 head per hour.
22
-------
A 15-month project to demonstrate the application of anaerobic
lagoons and two-stage trickling filters for the treatment of strong
wastes resulting from the slaughtering and processing of hogs is
nearly completed. This full-scale project is being done by Farmbest,
Inc. in Denison, Iowa.
The University of North Carolina will conduct a two-year study
for Gold Kist Poultry. It will involve changes in their processing
operations for demonstration of effective inplant control of both
water use and discharge of effluent from poultry processing.
A project titled "Construction and Study of A Demonstration Plant
Utilizing the Aerobic Channel Method for Treating Packinghouse
Wastes" is being conducted by John Morrell and Company in Ottumwa,
Iowa. The objective is to find an efficient and economical method
of treating raw packinghouse wastes so that they can be discharged
to surface waters. Solids will be removed, dried, and tested for
their possible use as an animal food supplement.
The feasibility of incineration of cattle paunch and ground manure
will be demonstrated by Illinois Packing Company. Following segre-
gation of waste streams for concentration of waste solids, a fluidized
bed incineration unit will be operated to develop the optimal perfor-
mance and related economics.
A grant was awarded to Stockton, California, for a full-scale
development and demonstration of pretreatment of packinghouse
waste using high-rate activated sludge plus in-sewer treatment
to reduce the BOD load to the municipal system by about 80 percent.
It is hoped that the in-sewer treatment will result in significant
savings in capital expenditures on the activated sludge plant.
The City of South St. Paul, Minnesota, has a grant titled "Effi-
ciency and Economy of Polymeric Sewage Clarification". The principal
objective is to determine the increased purification attainable by
treating various combinations of pa ckinghouse waste, domestic sewage,
and stormwater with polyelectrolytes and floe weighting agents.
Two projects are concerned with wastes from the beverage industry.
Widmer's Wine Cellars, Inc., Naples, New York, has a grant to design,
construct, and operate an extended aeration waste treatment plant to
treat winery processing wastewaters. Each of two parallel systems
will consist of an aeration basin, clarifier, and a sand filter.
Design loading is about 10 Ibs. BOD/day/1000 cu.ft. with a hydraulic
detention time of about 6.5 days.
The Agricultural Experiment Station, University of Puerto Rico, has
a project aimed at evaluating the efficiency and the economics of
several processes for treatment of rum distillery wastes. Pilot
plant work is scheduled on lagooning* activated sludge, and anaerobic
digestion.
23
-------
In addition to the two projects on potato processing wastes dis-
cussed at the Symposium there are two other full-scale R&D grants
concerned with these wastes.
Western Potato Service, Inc. of Grand Forks, North Dakota, will
demonstrate at full-scale the economics and the pollution reduction
characteristics of a potato "dry" caustic peeling system. A similar
full-scale processing plant which employs conventional caustic peeling
will be monitored for comparison. The dry caustic peeling process
was originally conceived and developed by the Western Regional Re-
search Laboratory of the U. S. Department of Agriculture in Albany,
California. This process has the potential of reducing the water
use of a processing plant by 50 percent and reducing the quantity
of BOD and SS discharged by 50 to as much as 75 percent.
A full-scale demonstration and evaluation of joint treatment of
municipal sewage and potato processing wastes is being done by
Grand Forks, North Dakota. Four lagoons, each with a capacity of
8.4 million gallons will be used for treatment prior to discharge of
the wastes into 640 acres of existing stabilization ponds. Two of
the lagoons will be completely mixed, covered anaerobic units and
two will be completely mixed aerobic units. Planned operation of
the facility is for one-half of the combined waste to be added to
an anaerobic cell followed, in series by an aerobic cell. The
remaining waste will be divided between the other anaerobic cell
and the other aerobic cell.
The City of Tualatin, Oregon, has a project entitled "Tertiary
Treatment of Combined Domestic/Industrial Wastes". About 25
percent of the BOD load is generated from the manufacture of dog
food. During 6 months of the year activated sludge treatment
followed by high-rate filtration and chlorination will suffice.
During the critical 6 months of low stream flow chemical addition,
flocculation, and sedimentation of the activated sludge effluent
prior to filtration will be practiced for phosphate removal.
Alternate methods of final disposal of waste activated sludge and
the chemical sludge will be investigated.
Corn Products Company is demonstrating secondary treatment of wastes
from the wet-milling industry. The three and one-half year project
entails the design, construction, and operation as well as a technical
and economical evaluation of a one million gallon per day completely-
mixed aerobic system.
"Removal and Recovery of Fatty Materials from Edible Fat and Oil
Refinery Effluents" is the title of a grant which has been awarded
to Swift and Coma ny. The 18-month study is being conducted at their
Bradley, Illinois, plant and will demonstrate the effectiveness and
economics of air flotation for removal and recovery of fatty materials
present in their processing wastewaters.
24
-------
Melbourne Water Science Institute in Australia has a grant titled
"Cannery Waste Treatment by Lagoons and Oxidation Ditch". Existing
100,000 gallons per day facilities are being used during the two-
year study to evaluate and optimize treatment of fruit and vegetable
processing wastes by anaerobic lagoons and oxidation ditches.
A grant to National Canners Association will demonstrate the effec-
tiveness of an ion exchange system for the treatment of olive brine
process water. Operating parameters as well as scale-up factors will
be determined from the 10,000 gallon per day pilot unit which will
use calcium hydroxide as a resin regenerant.
Although not directly related to the R&D effort on food processing
waste treatment it is worthwhile to list several points President
Nixon made in his February 10, 1970, message to Congress on the
environment.
The major industrial pollution problem was listed as: "Regulations
on disposal of industrial wastes have been too weak to prevent in-
creasing water pollution." The Administration proposal was: "Reform
pollution control program to greatly strengthen regulations on indus-
trial and municipal polluters and permit swift enforcement actions."
Seven of the specific items from the 37 Point Program as presented
by President Nixon are:
1. "Requirement that municipalities impose users fees on industrial
users sufficient to meet costs of treating industrial wastes.
2. Extension of Federal-State water quality standards to include
precise effluent standards for all industrial and municipal sources.
3. Provision that violation of established water quality standards
is sufficient cause for court action.
A. Revision of Federal enforcement procedures to permit swifter
court action against those in violation of water quality standards.
5. Authorization for the Secretary of the Interior to seek immediate
injunctions where severe water pollution threatens imminent danger
to health or irreversible damage to water environment.
6. Extension of Federal pollution control authority to include all
navigable waters, both inter- and intra-state, all interstate ground
waters and the United States' portion of boundary waters.
7. Provision that violation of established water quality standards
is subject to court-imposed fines of up to $10,000 per day.
25
-------
REFERENCES
1. Dostal, K. A., "progress Report: Aerated Lagoon Treatment of
Food Processing Wastes," Pacific Northwest Water Laboratory,
Corvallis, Oregon. (March 1968)
2. Dostal, K. A., "Secondary Treatment of Potato Processing Water",
Pacific Northwest Water Laboratory, Corvallis, Oregon. (July 1969)
3. Welch, R.M. and Antonie, R.L., "Preliminary Results of a Naval
Biological Process for Treating Dairy Wastes," Presented at the
24th Purdue Industrial Waste Conference, May 6-8, 1969.
26
-------
STATUS AND RESEARCH NEEDS OF
POTATO PROCESSING WASTES
by
Kristian Guttormsen and Dale A. Carlson*
INTRODUCTION
The potato processing industry in the United States has experienced
tremendous growth during the last decade, and future projections indicate
that this trend will continue. The total annual production of potatoes
in this country was about 15 million tons in 1966, an increase of 25
percent since 1956. Over the same period of time, the tonnage processed
increased by almost 150 percent(l). In 1966, approximately 35 percent
of the total crop was used for processing. In a study of the Pacific
Northwest by the Department of the Interior(2)9 a potato production
increase of 83 percent was predicted for the period from 1965 to 1985.
Over the same 20-year period, the production of frozen and dehydrated
potato products was expected to increase by approximately 250 percent.
The highest rate of growth of the potato processing industry has occurred
in Idaho, where more than 50 percent of the total production has been
processed since 1960. Unfortunately, the waste production has increased
accordingly, and potato wastes have become one of the major water pollu-
tion sources in the state. Although all of the major processors now
provide primary treatment, severe fish kills have occurred on the Snake
River during periods of low flow. As a result, the processors have been
directed by the Idaho State Health Department to provide secondary treat-
ment of their wastes(3). By the summer of 1973, all the plants will be
required to have such facilities in operation. Similar enforcement
policies are in effect across the nation.
WASTEWATER PRODUCTION
A variety of potato products is marketed today, and processing methods
and waste production vary from plant to plant. The J. R. Simplot
Company's three processing plants at Burley and Heyburn, Idaho, process
about 300,000 tons of potatoes per year. During the 1966-67 processing
season, these three plants used an average of 4,170 gallons of water per
ton of potatoes processed. The waste stream contained an average of
90 pounds of BOD and 110 pounds of suspended solids per ton of potatoes.
Total nitrogen and phosphate amounted to 3.5 and 0.6 pounds per ton,
respectively. These figures were derived from the flows entering the
*Respectively, Sanitary Engineer, Ruskin, Fisher & Assoc., Seattle,
Washington; and Professor of Civil Engineering, University of Washington,
Seattle, Washington.
27
-------
primary clarifier, and do not include fat recovered and solids screened
out. As a result of lye peeling, the raw waste had a pH of 11.3. The
suspended solids and BOD values were higher than experienced during
previous years. Average values for the potato processing industry are
about 4,200 gallons, 60 pounds of suspended solids and 50 pounds of BOD
per ton of potatoes processed(4).
PRIMARY TREATMENT
Because potato wastes contain high concentrations of settleable suspended
solids, sedimentation has been a relatively effective treatment method.
It has been estimated that after the installation of primary treatment
facilities in the Idaho processing plants, the waste load reduction was
equivalent to the waste from three million people^).
Treatment facilities range from settling ponds used for flume and waste-
water to more sophisticated mechanical clarifiers especially designed
for potato waste. Sedimentation studies conducted in Idaho concluded
that primary clarifiers should be designed for a maximum overflow rate
of 600 gpd/ft^j with detention times of 2.5 to 3 hours(5). By proper
design of the sludge rake mechanisms and control of the sludge return,
clarifier underflows with solids concentrations of 6 to 7 percent could
be obtained. Clarifiers designed according to these criteria were
expected to reduce the BOD by 60 percent and the suspended solids by
90 percent.
High removal efficiencies have been reported from field installations.
Average results from two processing seasons (1966-67 and 1967-68) at the
J. R. Simplot Company's waste treatment plant in Burley, Idaho, were 41
percent BOD removal and 73 percent suspended solids removal. The 100-foot
diameter clarifier was operated at an average overflow rate of 800 gpd/ft(4)
Simplot"s waste treatment plant in Caldwell, Idaho, consists of grease
removal facilities, three rotary screens with 4-mesh cloth, and a
clarifier 100 feet in diameter. Plant performance for the 1967-68
processing season was 62 percent COD removal, 93 percent suspended solids
removal, and 95 percent settleable solids removal. The overflow rate of
the clarifier was about 730 gpd/ft2(5).
Attempts have been made to improve primary clarifier performance by
using chemical coagulants' '. However, significant improvements over
plain sedimentation was not achieved. Some success has been reported
on flotation treatment of potato wasted). A pilot plant study conducted
by the Potato Products Waste Disposal Executive Committee for the RexJ
River Valley gave suspended solids removals in excess of 90 percent and
BOD removals as high as 75 percent.
Centrifugation has also been proposed as a means of treating potato
waste, especially flume and wastewater, but studies by the Potato
Processors of Idaho Association did not find this method a feasible
replacement for conventional clarification(5> 8)<
28
-------
Disposal of the large volume of solids recovered by primary clarification
appeared to be a major problem until it was discovered that the sludge
could be dewatered and used for cattle feed. Vacuum filtration has been
found to be the best method for dewatering the sludge. Centrifugation
and screening have been considered as well, but have not proved as
successful as vacuum filtration(^). it was found that sludge from
plants using steam peeling could be dewatered to a concentration of 12
to 16 percent dry solids. The filtrate contained suspended solids in
concentrations of 1,000 to 1,200 mg/1, most of which would settle when
the filtrate was returned to the clarifier. Sludge from plants using
lye peeling appeared to be impossible to debater because water would
not separate from the solids and the gelatinous slurry would blind the
filter cloth. By chemical treatment of the sludge with ferric sulfate
and lime, it was found that the caustic sludge could be dewatered at
about the same rate as steam-peel sludge. However, the filter cake
could not be used for cattle feed because of the chemical content, and
the chemical costs were high. It was finally discovered that by aging
the sludge in the clarifier, the sludge becomes filterable. Bacterial
action breaks down the gelatinous consistency. Close control is neces-
sary because holding the sludge too long results in bacterial decomposi-
tion of the solids with reduced filterability and lower clarifier
efficiency.
The potato solids have been proven to be at least equivalent to barley
in nutritional value, and feeding experiments have indicated substantial
acceleration of weight gain in cattle after potato solids were included
in the diet. At the present, most of the Idaho processors are selling
the dewatered sludge along with screen solids for approximately $3 per
ton of solids at about 14 percent dry solids content. According to
Grames and Kueneman(^), this is sufficient to recover the capital cost
of primary treatment facilities in two to four years.
SECONDARY TREATMENT
The majority of the potato processors do not provide secondary treatment
of their wastes at the present, and operational results from secondary
facilities are therefore practically non-existent. A considerable
amount of research has been carried out, however, in bench-scale and
pilot plant studies.
Atkins and Sproul(9) studied the feasibility of using the complete mix
activated sludge process to treat the combined waste from a processing
plant using lye peeling. At loadings above 300 Ibs BOD/1,000 cu ft/day,
BOD removals in excess of 90 percent were obtained without pH adjust-
ment or inorganic nutrient addition. The consulting firm of Cornell,
Rowland, Hayes and Merryfield conducted pilot plant studies for the
Potato Processors of Idaho Association^). It was found that the com-
plete mix activated sludge process operated best at loadings below
150 Ibs BOD/1,000 cu ft/day, BOD removals above 90 percent were obtained.
Excess sludge production was a problem as about one pound of biological
solids were produced per pound of BOD removed. It was also anticipated
that nutrient addition would be required in a full-scale plant.
29
-------
Last year, a full scale complete mix activated sludge plant was put
into operation at the Pillsbury Company's processing plant at Grafton,
North Dakota^10). The design loading was 400 Ibs BOD/1,000 cu ft/day.
The effluent which was expected to contain 200-400 mg/1 BOD was to be
discharged to a municipal sewer for further treatment.
The contact stabilization process has also been applied to potato waste,
but not as successfully as the complete mix process^, 9).
Pilot plant studies on trickling filters have been conducted by the
Potato Processors of Idaho(3). At loadings of 400 Ibs BOD/1,000 cu ft/
day, the filter, which had artificial media, was capable of 75 percent
BOD reduction. Nutrient addition was found to be advantageous, but no
pH adjustment of lye-peeling wastes was necessary. It was concluded
that addition of alkalinity might be required when treating steam-peel
waste to prevent the pH from dropping. Sludge production was found to
be on the order of 0.6 to 0.8 pounds per pound of BOD removed.
Trickling filter treatment of protein water from a starch plant gave
BOD reduction of 90 percent at loadings of 30 Ibs BOD/1,000 cu ft/day.
With a high rate filter, BOD removals of 90 percent were obtained at
loadings up to 70 Ibs BOD/1,000 cu ft/day. The rock media filter became
clogged with sloughed solids at higher loadings (H).
Recently, the anaerobic filter has received attention as a potential
treatment process for potato waste. The anaerobic filter, which is
similar to the aerobic tricking filter in appearance, has upward flow
so that the rock media is completely submerged. The anaerobic organisms
are attached to the stones as well as suspended as discrete particles
in the interstitial spaces. High solids retention time, which is an
important operational parameter in anaerobic processes, is easily
obtained in the filter and may be on the order of hundreds of days
Pilot plant studies in Idaho(^) gave BOD removals of almost 70 percent
when the filters were loaded at about 60 Ibs BOD/1,000 cu ft/day. Two
5-foot diameter filters were tested, one with a 4-foot media depth and
one with an 8-foot depth. Almost identical performance was obtained
with both filters. The gas produced had a methane content of more than
70 percent and is therefore combustible. In a full-scale plant, the
gas could be used as an energy source. The effluent from an anaerobic
filter should be passed through a short-term aeration basin to provide
additional organic removal and render the effluent suitable for discharge
to a receiving stream. With primary treatment, this system appears
capable of 90 percent BOD removal. Some possible limitations of the
anaerobic filter have been reported by Young and McCartyN-'-^).
Stabilization ponds or oxidation ponds have long been a popular and
economical method of waste treatment. Porges(l^) made a survey of
industrial stabilization ponds in the United States in 1962. He reported
three ponds treating potato processing wastes. The median loading and
detention time of the ponds were 111 Ibs BOD/acre/day and 105 days,
30
-------
respectively. No treatment efficiencies were reported, but one of the
installations was reported effective while another caused odor nuisances.
Studies by Cornell, Rowland, Hayes and MerryfieldO) concluded that
pond loadings should be kept well below 80 Ibs BOD/acre/day, and in
general ponds should be designed according to criteria used for domestic
waste.
Ponds at Grafton and Park River in North Dakota have been studied
extensively(15, 16)_ These ponds received domestic waste as well as
potato processing waste. Both ponds have been heavily overloaded by
potato waste during the processing season. During the winter, the ponds
were ice covered and acted merely as storage ponds. When biological
activity was reestablished in the spring, anaerobic conditions developed,
and severe odors were experienced. Nevertheless, it was concluded that
potato waste and domestic waste digested readily in the ponds when the
organic loading from potato wastes was 15 times or more the domestic
loading. During the summer, the combined waste could be applied at
loading rates well above the conventional design loading for northern
climates of 20 Ibs BOD/acre/day. Although no recommendations could be
made from the study because of the "diametrically opposed" operating
seasons for the pond and processing plant, it was indicated that load-
ings on the order of 50 to 60 Ibs BOD/acre/day might be possible.
Although anaerobic ponds have been used extensively for treatment of
food processing wastes, they have not yet been used to any extent by
the potato processors. Forges(^) reported that in 1962, three anaerobic
ponds were used in the United States for potato wastes. No information
was given on the efficiency of the ponds, but it was reported that odor
nuisances were associated with one of them.
Efforts by the Potato Processors of Idaho to establish design criteria
for anaerobic ponds gave encouraging results. In a completely mixed
anaerobic pond loaded at 8 Ibs BOD/1,000 cu ft/day, average BOD removals
of 70 percent were obtained-"). An unmixed pond operated in parallel
achieved only 40 percent reduction. The effluent from the mixed pond
was treated in an aerated cell for further removal and oxygenation. With
primary treatment, this system appears capable of 90 percent BOD reduc-
tion without pH adjustment or nutrient addition.
Aerated lagoon treatment has also been investigated. Reductions of
85 percent BOD were obtained in a one-acre experimental lagoon at Park
River, North Dakota, at temperatures close to 0° C. The detention time
was 14 days(^). Pilot plant studies in Idaho gave total BOD reduction
of approximately 80 percent and soluble BOD reduction in excess of 90
percent. The detention time was 7.8 days and the temperature about
7° C (4).
Spray irrigation has been used successfully in this country as well as
in Europe as a method of final disposal of potato wasted, 18, 19).
31
-------
IN-PLANT WASTE LOAD REDUCTION
The importance of in-plant efforts to reduce the total waste load can
hardly be overemphasized. Capital and operating cost of waste treat-
ment facilities are closely related to the volume and organic content
of the waste, and the cost of in-house modifications will generally
represent substantial savings in the cost of treatment facilities. Now
that secondary treatment will be enforced, in-house waste reduction
efforts will be most important.
Excessive use of water is common in many processing plants, and with
proper control the total use can often be reduced by 50 percent. Reuse
of processing water is becoming more common among the processors, and
large savings are obtained this way. Flume and wash water, for example,
are low in BOD and dissolved solids and can be reused following minimal
treatment to remove silt, debris, and large solids. A special form for
reuse is counter-current flow of process water. This scheme is espe-
cially attractive in processes with successive washes and rinses, since
the product is moved forward after each washing into water which is
cleaner than that used in the preceding wash. This principle is being
used for final purification of starch(20)f
Good housekeeping and in-plant removal of solids are very important.
Potato peel and pulp contain 0.1 mg of BOD per mg of dry solids^21), and
solid matter should therefore be removed from the waste stream as fast
as possible to avoid leaching of organics, especially in warm water.
Experiments by Sproul(22) showed that in 30 minutes 150 grams of sliced
potatoes leached about one-half gram of soluble BOD. Various types of
screens are presently used to remove solids in the plant.
The peeling operation contributes the major waste load from a potato
processing plant, and much attention has been given to improvement of
this process. Two types of peeling processes are used extensively today,
steam peeling and lye peeling. Abrasion peeling is also used. All
these processes use large quantities of water and impose large solids
and BOD loads on the waste stream.
Presently, work is underway on two newly developed "dry" peelers.
These peelers were developed at the Western Regional Research Laboratory
of the United States Department of Agriculture at Albany, California,
and at the Institute for Storage and Processing of Agricultural Produce
(IBVL) at Wageningen in Holland.
The peeler developed in this country applies infrared radiation to
caustic-treated potatoes. The softened skin is removed abrasively
without use of water, and the potatoes are given a final brush washing
before they are discharged from the peeler assembly. A pilot plant
installation in Idaho had a capacity of about 5,500 Ibs of potatoes
per hour. The water use was about 280 gallons per hour or 100 gallons
per ton of potatoes. This represents a water use reduction of more
than 80 percent with respect to steam or lye peeling. The process was
frequently compared with a conventional lye-peeler, and trim losses were
32
-------
taken as an indication of the effectiveness of peel removal. When the
efficiency of the two lines were approximately equal, it was found that
the peeling loss on the dry-peel line was 13 percent, while the loss
on the conventional' line was about 20 percent. The caustic use was
reduced by 70 percent on the dry-peel line as compared to the conventional
line. Based on the limited amounts of data collected during the study,
it was estimated that the total waste load to secondary treatment facil-
ities could be reduced by at least 50 percent(23). Additional work on
a larger scale is underway to evaluate the overall economics of the
method.
By-product recovery from potato processing and process waste has received
considerable attention. As mentioned earlier, primary sludge and
screened solids are being sold or used for cattle feed. A symbiotic
yeast process has been developed by the Swedish Sugar Corporation(-2^0 .
Process water is used to make a growth medium for the yeast Torula. The
product may be used as a feed additive or for human consumption. In
addition to the yeast production, a considerable BOD reduction was also
achieved.
Considerable work has been conducted on by-product recovery from starch
plant wastes. Presently, research is underway at the Eastern Utilization
Research and Development Division of the U. S. Department of Agriculture' -)'
The proposed treatment process would consist of five steps: concentra-
tion of dilute waste by reverse osmosis; precipitation and recovery of
protein by steam injection or other suitable methods; separation and
recovery of potassium and other inorganic cations by ion exchange;
separation and recovery of amino compounds by ion exchange; and recovery
of organic acids and phosphates by ion exchange. Promising results have
been obtained.
RESEARCH NEEDS
With the existing waste treatment technology, any reasonable effluent
quality requirement imposed on the potato processing industry can be met.
In England, where effluent requirements are considerably more restric-
tive than in the United States, potato waste treatment facilities capable
of producing effluents with BOD and suspended solids concentrations
below 20 mg/1 are in operation. In one such plant, primary sedimentation
is followed by two stages of biofiltration and activated sludge treat-
ment. Final polishing on sand filters enables reuse of a large fraction
of the effluent in the processing operations without detrimental effects
on the quality of the final product. A treatment scheme such as this is,
of course, costly, and cost is the main treatment problem facing most
potato processors. Capital as well as operating costs are high, with
little or no possibility for return on the investments.
Many possibilities exist for reducing the amount of waste from potato
processing and improving the waste treatment technology; a great
deal of work has already been done to alleviate the potato waste problem.
However, with all due respect to previous and present efforts, there is
always room for further improvements.
33
-------
In-plant changes and modifications may be the most important area of
concern. The peeling operation, which contributes from 50 to 75 percent
of the total plant waste load, must be improved in order to reduce the
waste load significantly. Very promising results have been obtained
with the dry caustic peeler previously described. However, more work
is required to perfect this process. Handling, disposal, and utilization
of the slurry produced from the peel and wastewater must be investigated.
The economics of the process compared to other peeling methods also
must be demonstrated.
The rest of the processing operations should be considered as well. Solids
losses and water usage must be reduced, since these increase the waste
load as well as the total loss from the process. Improved equipment
and methods for slicing, blanching, cooking, etc. should be developed.
In the same respect, firm guidelines for good housekeeping should be
established, and process line personnel should be educated in waste pre-
vention.
A certain loss of solids will always occur. Preferably, these solids
should not be allowed to enter the waste stream. If they do enter the
wastewater, then as much as possible of these solids should be removed
in the processing plant before considerable leaching of organics takes
place. Removal may be achieved by screening, filtration, or centrifuga-
tion. In this respect, research is needed on suspended solids particle
size in relation to screening.. Fragmentation of the particles on the
screen may increase the apparent BOD. A similar phenomenon may take
place in the centrifuge. The solid particles may be broken down by the
centrifugal force and thus remain in the centrate. It has been reported
that this indeed happens when primary potato sludge is dewatered by
centrifugation. The small particles discharged with the centrate have
poor settling characteristics and thus are not removed easily by sedi-
mentation.
Reuse of process water has been discussed earlier. Besides removal of
suspended solids, quality control of the water must be investigated.
Various methods of microbial control should be evaluated to control
deterioration of the recirculated water. The tolerance limits for dif-
ferent pollutants should be established for various steps in the process
on a health hazard basis as well as on a product quality basis.
Separation of various waste streams may be advantageous for subsequent
treatment in different systems. Waste concentration also may prove very
advantageous since concentrated waste can be treated more efficiently
than dilute waste. Concentration and separation by reverse osmosis is
one possibility which presently is being investigated.
The concentrated portion of the waste stream may be subjected to recovery
of various valuable constituents while the separated dilute portion may
be reused in the processing operations. The present research efforts
are directed towards potato starch plant waste. Feasible application of
this or similar processes to other types of potato wastewater should be
investigated.
34
-------
A variety of salable by-products can be recovered from the various waste
streams, and considerable research has been conducted in this field.
Few large-scale, economically feasible processes have been developed,
however.
Some of the compounds which may be recovered from potato waste include
amino acids, protein, organic acids, potassium, phosphates, and other
inorganic ions. All these compounds have been recovered by physical or
chemical processes.
Potato pulp and wastewater may also be used as growth media for yeasts
and molds for production of protein and antibiotics. Promising results
have been obtained with the yeast Torula.
A number of edible products for human consumption have been proposed
from larger potato solids. Utilization of potato solids for livestock
feed has found wide acceptance.
In order to take advantage of these possible by-products from potato
wastes, feasible processes and a market must be developed. The present
efforts on by-product recovery are concentrated mainly on starch plant
waste, since the potato starch industry is a rather marginal industry.
More emphasis should be directed towards recovery of salable products
from other types of potato processing waste.
The ultimate solution to the waste treatment problem would be to produce
an effluent which could be completely reused in the processing operations.
Although such a goal is rather impractical at the present time, the
possibility of effluent reuse nevertheless should be kept in mind. In
order to solve the more contemporary problems facing the potato processors,
existing and proposed technology must be improved and modified to give
more efficient, less costly waste treatment. The most pressing problem
at present is secondary treatment, a requirement which is facing most
of the processors across the nation. However, continued efforts to
improve primary treatment methods should not be ignored since primary
treatment can reduce the load on the secondary treatment facilities
significantly.
Potato processing wastes have been found easily degradable by biological
treatment processes, and considerable progress has been made in their
application to potato waste treatment. A significant amount of pilot
plant work has been done, but operational results and experience from
full-scale installations are limited. Research needs for biological
treatment processes are not unique to potato processing wastes but
apply to most industrial wastes. In general, better prediction models
for the different processes must be developed. The values of the model
factors, i.e., removal rate constants, growth rate constants, etc., and
their variations with waste quality parameters, should be determined
more precisely. More specifically, the availability of the essential
nutrients, such as nitrogen and phosphorus, for adequate treatment should
be investigated. The amount of nutrients in potato waste is generally
35
-------
sufficient for effective biological treatment, but the nutrients may
not be readily available for cell incorporation. The effects of condi-
tions or compounds directly inhibitory to biological treatment, such as
high pH, antifoaming agents and cleaning compounds, should be studied
further.
Promising results have been obtained with the anaerobic filter, the
anaerobic contact process and anaerobic ponds, and work on these pro-
cesses should be continued. Special attention should be given to start-up
characteristics, duration of start-up period, optimum operating condi-
tions, and solids-liquid separation. Solids disposal also should be
considered.
High-rate chemical oxidation processes have not been found economically
feasible at the present. Further research on these and other chemical
and physical processes should be conducted, especially with regard to
more complete reuse of treated effluents.
Conventional primary treatment and biological secondary treatment produce
excessive solids. Primary sludge has been used successfully for cattle
feed. The possiblity of utilizing secondary sludge in a similar way
should be investigated. The economics of other solids disposal methods
should be evaluated since little is known about the feasibility of
secondary sludge for cattle feed and since there may not be a market
for this product. Efficient and economical methods for dewatering
dilute sludges should be developed, along with methods for handling
sludge with high solids content^"'.
SUMMARY
Several treatment processes have been found economically feasible for
secondary treatment of potato processing wastes, and large sums of money
need not be spent to study these processes further. Efforts should be
made to reduce the capital cost and operation and maintenance cost of
secondary treatment facilities. Future research should be directed
mainly towards the processing plant in order to reduce the waste pro-
duction and recover valuable by-products.
36
-------
REFERENCES
1. U. S. Dept. of Agriculture, "Utilization of 1965 Crop with Com-
parison." Irish Potatoes, Statistical Reporting Service, Pot 1-3
(9-66), Crop Reporting Board, USDA.
2. U. S. Dept. of the Interior, "Pacific Northwest Economic Base Study
for Power Markets. Agriculture and Food Processing." Vol. 2,
Part 5, Bonneville Power Administration. (1966).
3. Cornell, Rowland, Hayes, and Merryfield, Engineers and Planners,
"An Engineering Report on Pilot Plant Studies, Secondary Treatment
of Potato Process Water," prepared for the Potato Processors of
Idaho Assoc. (Sept. 1966).
4. Dostal, K. A., "Secondary Treatment of Potato Processing Wastes."
Report No. FR-7, Federal Water Pollution Control Administration,
United States Department of the Interior. (July 1969).
5. Grames, L. M. and Kueneman, R. W., "Primary Treatment of Potato
Processing Wastes with Byproduct Feed Recovery." JWPCF 41(7):
1358-1367. (1969).
6. Cornell, Rowland, Hayes, and Merryfield, Engineers and Planners,
"A Pilot Plant Study, Anaerobic Secondary Treatment of Potato
Process Water." Prepared for the Potato Processors of Idaho Assoc.
(July 1969).
7. Francis, R. L., "Summary of Potato Products Wastes Study." Potato
Products Waste Disposal Executive Committee, Red River Valley. (1962),
8. Dickinson, D., "Treatment of Effluents from Potato Processing."
Proc. International SympqsjLum. Utilization and Disposal of Potato
Wastes. New Brunswick Research and Productivity Council, N. B.,
Canada. (1965).
9. Atkins, P. F., and Sproul, 0. J., "Feasibility of Biological Treat-
ment of Potato Processing Wastes." Proc. 19th Industrial Waste
Conference, Purdue University, 303-316. (1964).
10. Michaelson, C. H., Personal Communications. (March 1969).
11. Buzzell, J. C., Caron, A-L. J., Ryckman, S. J., and Sproul, 0. J.,
"Biological Treatment of Protein Water from Manufacture of Starch."
Water & Sewage Works. (July and Aug. 1964).
12. Carlson, D. A., "Recent Developments in Anaerobic Waste Treatment."
Proc. Symposium on Potato Waste Treatment, FWPCA and University of
Idaho. (1968).
37
-------
13. Young, J. C., and McCarty, P. L., "The Anaerobic Filter for Waste
Treatment." Tech. Report No. 87, Dept. of Civil Engineering,
Stanford University. (March 1968).
14. Forges, R., "Industrial Waste Stabilization Ponds in the United
States." JWPCF 35: 456-468. (1963).
15. Fossum, G. 0., and Cooley, A. M., "Stabilization Ponds Treating
Potato Wastes with Domestic Sewage." Proc. International Symposium.
Utilization and Disposal of Potato Wastes, New Brunswick Research
and Productivity Council, N. B., Canada. (1965).
16. Olson, 0. 0., Van Heuvelen, W., and Vennes, J. W., "Experimental
Treatment of Potato Wastes in North Dakota, U.S.A." Proc. Inter-
national Symposium, Utilization and Disposal of Potato Wastes. New
Brunswick Research and Productivity Council, N. B., Canada. (1965).
17. Olson, 0. 0., and Vennes, J. W., "Mechanical Aeration of Potato
Processing and Domestic Sewage." Park River Study, FWPCA. (1963).
18. Haas, F. C., "Spray Irrigation Treatment." Proceedings of a
Symposium on Potato Waste Treatment. FWPCA and University of Idaho.
(1968).
19. Szebiotko, K., "Total Utilization of Potatoes Including the Disposal
of Industrial Wastes." Proc. International Symposium, Utilization
and Disposal of Potato Wastes. New Brunswick Research and Productivity
Council, N. B., Canada. (1965).
20. Douglass, I. B., "The Manufacture of Potato Starch." Proc. Inter-
national Symposium, Utilization and Disposal of Potato Wastes. New
Brunswick Research and Productivity Council, N. B., Canada. (1965).
21. Carlson, D. A., "Biological Treatment of Potato Wastes." Proc. 13th
Pacific Northwest Industrial Waste Conference. Pullman, Wash. (April
1967).
22. Sproul, 0. J., "Wastewater Treatment from Potato Processing." Water
& Sewage Works 2: 93. (Feb. 1968).
23. Willard, M., "Pilot Plant Study of the USDA-Magnuson Infra-Red
Peeling Process." Paper presented at 19th National Potato Utiliza-
tion Conference, Big Rapids, Mich. (July 1969).
24. Wramstedt, S. Personal Communications. (Sept. 1968).
25. Heisler, E. G., Siciliano, J., and Porter, W. L., "Progress Report
on Starch Factory Waste Treatment." Paper presented at 19th National
Potato Utilization Conference, Big Rapids, Mich. (July 1969).
26. Guttormsen, K. and Carlson, D. A., "Current Practice in Potato Pro-
cessing Waste Treatment." Water Pollution Control Research Series,
DAST-14, 108 pp. (Oct. 1969).
38
-------
AEROBIC SECONDARY TREATMENT OF
POTATO PROCESSING WASTES
by
Glenn A. Richter*
INTRODUCTION
The purpose of this paper is to report the progress being made on
a government demonstration grant project at the R. T. French Company's
potato processing plant at Shelley, Idaho.
The primary objectives of the grant are to:
1. Demonstrate the feasibility of secondary treatment of potato
processing wastes by a fully aerobic process.
2. Determine BOD (biochemical oxygen demand) removal efficiencies
when the treatment facility is operated as:
A. A complete mix activated sludge system
B. A flow-through aeration system without sludge return
C. An intermittent aeration system with periods of
aeration, clarification, and supernatant drawoff
3. Determine the quantity and character of waste biological sludge
produced during aerobic stabilization of potato processing wastes.
4. Define the influence of: foaming, ice, temperature, pH, nitro-
gen, and phosphorous on treatment.
5. Determine the operating costs for the methods of treatment
demonstrated.
PLANT INFLUENT CHARACTERISTICS
Potato processing plant wastewater can be divided into two major
streams: (1) silt water and (2) process water. The silt water
originates from raw potato washing and fluming operations. It
contains large amounts of soil removed from the raw potatoes.
Process water flows from potato processing operations, where raw
*Project Engineer, Cornell, Rowland, Hayes & Merryfield, Corvallis
Oregon 97330
39
-------
potatoes are processed into packaged, partially cooked food products.
It contains caustic potato peeler and barrel washer discharges, as
well as all other liquid wastes from the processing operations, in-
cluding clean-up water.
Total secondary treatment plant influent contains primary clarified
process water and clarified silt water. Characteristics of the total
plant influent are described below:
Alkalinity, pH, and Temperature - Total plant influent alkalinity
varied from 70 mg/1 to 5740 mg/1, with an average of 1730 mg/1 as
CaCO . The pH varied from 6.8 to 11.5, and averaged 9.3. The
influent temperature varied from 12°C to 24°C. The average temper-
ature was 20°C.
s
BOD and COD - The BOD of the total plant influent varied from 780
mg/1 to 2990 mg/1. The average BOD was 1680 mg/1. 65 to 70 percent
of the total influent BOD was in a soluble form. The COD of the
influent varied from 760 mg/1 to 7100 .mg/1, with an average of
3050 mg/1. The average BOD:COD ratio was 0.55.
Suspended Solids - The TSS (total suspended solids) concentration
in the total plant influent varied from 320 mg/1 to 6460 mg/1. The
average TSS was 1430 mg/1. VSS (volatile suspended solids) aver-
aged about 90 percent of the TSS.
Nutrients - On the basis of a limited amount of testing, the average
nitrogen and phosphorous concentrations in the total plant influent
were not limiting factors. The average BOD:N:P ratio was equal to
100:4.5:0.6.
PLANT DESCRIPTION
Design Criteria - The criteria used for the final design of the
secondary treatment facilities at the R. T. French Company are listed
in Table 1.
Plant Operation - An aerial photo of the new secondary waste treat-
ment facility is shown on Figure 1, and the treatment facility flow
diagram is shown on Figure 2. A general description of the flow
patterns of the plant is given in the following paragraphs.
Primary clarified potato process wastewater flows by gravity to the
influent pump pit where it is combined with clarified silt water.
This total plant influent is then pumped to the aeration basin
inlet-outlet structure, where the flow can be divided, as desired,
between the two aeration basins. The incoming wastewater is dis-
persed throughout each of the aeration basins by floating mechanical
aerators.
40
-------
FIGURE 1
THE R.T. FRENCH COMPANY
SHELLEY, IDAHO
AERIAL VIEW
SECONDARY WASTE
TREATMENT FACILITY
FWPCA DEMONSTRATION GRANT
120602 EHV
-------
MliriOH llt>N- ,
IKKT-BIHltl •
ITIUCTUtl
FIGURE 2
THE R.T. FRENCH COMPANY
SHELLEY,IDAHO
FLOW DIAGRAM
SECONDARY WASTE
TREATMENT FACILITY
PWVPCA DEMONSTRATION GRANT
120602 EHV
B4435.8
-------
Effluent from the aeration basins flows by gravity back through
the aeration basin inlet-outlet structure to the clarifier bypass
structure. The flow is normally directed to the secondary clari-
fier, but may be bypassed during an emergency at the clarifier
bypass structure for discharge to the Snake River.
Effluent from the secondary clarifier flows to the Snake River.
Table 1. Secondary Treatment Plant Design Criteria
Average Daily Flow
Waste Process Water (mgd) 1.0
Waste $.ilt Water (mgd) 0.25
Total 1.25
Peak Flow Rate
Waste Process Water (gpm) 836
Waste Silt Water (gpm) 400
Total 1,236
BOD
Waste Process Water (Ibs/day) 13,700
Waste Silt Water (Ibs/day) 400
Total 14,100
TSS
Waste Process Water (Ibs/day) 5,680
Waste Silt Water (Ibs/day) 940
Total 6,620
Settled lightweight sludge flows by gravity from the bottom of the
clarifier to the sludge collection launder. It then flows by gravity
from the sludge launder to the sludge recirculation pump pit. The
sludge is then pumped to the recirculated sludge splitter box, where
it can be divided into two portions of desired size and returned to
the two aeration basins.
Dense settled solids are pumped from the sludge hopper in the bottom
of the clarifier to either the vacuum filter or to the clarifier-
thickener. This waste sludge may also be pumped to the clarifier
bypass structure for discharge to the river.
43
-------
The waste sludge, pumped to the clarifier-thickener, is concentrated
along with silt water. Clarifier-thickener overflow is transferred
by gravity to the influent pump pit. The thickened sludge is pumped
from the bottom of the clarifier-thickener to the vacuum filter, where
it is dewatered along with any waste sludge pumped directly to the
filter.
Solids discharged from the vacuum filter are conveyed to the silt
bunker in a cake form. This cake is then hauled away by trucks for
ultimate disposal as landfill.
Unit Sizes - The major treatment units of the secondary treatment
facility are described below.
Figure 3 shows Aeration Basin 1, which has a capacity of 1.25 mgd and
holds three 50 hp floating mechanical aerators. Aeration Basin 2,
shown on Figure 4, has a capacity of 2.5 mgd and normally utilizes six
50 hp floating mechanical aerators. Both basins have a water depth of
16 feet, and a design BOD loading of 28 lb/1000 cu. ft/day.
The secondary clarifier (Figure 5) is 70 feet in diameter and has a
12-foot side water depth. The clarifier has a rapid sludge with-
drawal mechanism, and a hydraulic overflow rate of 325 gal/sq ft/day
at design flow.
Figure 6 shows the treatment facility control building. The control
building houses the vacuum filter, sampling and testing laboratory
facilities, and various sludge handling facilities. The vacuum
filter is a disc type and has a surface area of 300 square feet.
Figure 7 shows the clarifier-thickener which is 30 feet in diameter.
The side water depth is 18 feet and the hydraulic overflow rate is
860 gal/sq ft/day.
Operation Schedule - During the test period, the secondary treatment
facility received wastewater flow on a schedule parallel with the
operating schedule of the potato processing plant. The processing
plant operated almost continuously, except for a two-day shut down
on weekends for clean-up and repairs. Cleaning water flow during
processing shutdowns was directed to the treatment facility.
Primary clarified process water was directed to the secondary treat-
ment facility by adjusting slide gates in Manhole 1 (see Figure 2).
The amount of process water discharged to the facility was determined
by the aeration basin operation. All process water entered the facility
during normal operation. Screened silt water was directed to the
clarifier-thickener by silt water pumps in the storage cellars. Potato
processing seasons are about 9 to 10 months long.
PLANT PERFORMANCE
The secondary waste treatment facility has operated from September 3,
1969, through the present time. Operating data are included through
March 31, 1970.
44
-------
:. •
Figure 3. The R. T. French Company, Shelley, Idaho
AERATION BASIN 1
45
-------
Figure 4. The R. T. French Company, Shelley, Idaho
AERATION BASIN 2
46
-------
Figure 5. The R. T. French Company, Shelley, Idaho
SECONDARY CLARIFIER
47
-------
-
Figure 6. The R. T. French Company, Shelley, Idaho
CONTROL BUILDING
48
-------
Figure 7. The R. T. French Company, Shelley, Idaho
CLARIFIER - THICKENER
49
-------
Numerous mechanical problems have hindered plant performance; however,
three months of effective operation has been achieved. The operating
data from these three months of experience have been analyzed in
greater detail than the other reported data.
Activated Sludge System - Figure 8 is a plot of the best fit curve
of the frequency distribution of BOD loading to the activated sludge
system on arithmetic-probability paper. The probability shown on
this and following figures is the percent of measurement equal to or
less than, the stated class mean of the measured item. Figure 8 shows
that about 50 percent of the time the BOD loading to the activated
sludge system was equal to, or greater than, the plant design load of
14,100 Ib BOD/day; while 25 percent of the time, the BOD loading was
equal to, or less than, 10,600 Ib BOD/day.
Figure 9 shows that about 40 percent of the time the D.O. (dissolved
oxygen) in Basin 1 was less than 1.0 mg/1, which is often thought to
be a minimum level to avoid limiting biological activity. Figure 10
shows that Basin 2 was also operated at a D.O. of less than 1.0 about
40 percent of the time. Low basin D.O. was associated with mechanical
aerator problems.
Figure 11 shows that about 23 percent of the time, Basin 1 temperature
was equal to or below, 50°F (10°C), which is reported to be low
enough for filamentous bacteria to become competitive. Figure 12 shows
that a temperature below 50°F occurred about 18 percent of the time.
Low basin temperature occurred at long aeration times and low air
temperatures.
Figure 13 shows that about 25 percent of the time, Basin 1 pH was equal
to, or above, 8.0, which is the top of the normal operating range for
an.activated sludge system. This occurred in Basin 2 about 35 percent
(Figure 14) of the time, but neither basin appeared to be affected by
the high pH.
C02 production in the aeration basins during organic material breakdown
buffered the high alkalinity in the influent and resulted in a pH drop.
This is illustrated on Figure 15. The average influent pH was 9.3 and
the average basin pH was 7.8. Many of the high basin pH levels were
associated with low organic removals, which resulted from mechanical
aerator problems.
The average BOD loading (lb/1000 cu ft/day) and BOD removals (percent)
are shown on Figure 16. The various operating phases are also shown on
Figure 16. BOD reductions decreased during period of sludge bulking.
Sludge bulking was associated with mechanical aerator failures, which
lowered the available D.O. in the aeration basins below desirable levels,
Analysis of three periods of operation, each about one month long with
the plant operating as a return sludge activated sludge system,
50
-------
c
2
H
M
n
u
0
0
3
cr
w
d
^3
o
95
90
80
70
60
50
40
30
20
10
5
39
O Q
0
62
85
201
108 131 154 178
100 Ibs BOD/Day
Figure 8. PROBABILITY OF PLANT LOADING
51
224
247
-------
(C
en
0
0
n)
=
cr
Ed
j
a
'-j
—
99.9
99
95
90
80
70
60
50
40
30
20
10
7r
0.8 1.7 2.6 3.6 4.5 5.4 6.3
Dissolved 0 Basin 1, mg/1
Figure 9. PROBABILITY OF BASIN 1 D.O.
7.2
8.2
9.1
52
-------
99.9
99
95
90
80
70
60
50
40
30
20
10
o
0
c
a
-
cc
•s.
0
-
a1
<
*
>,
O
•-:
=-,
O.A7 1.15 1.83 2.51 3.19 3.87 4.55 5.23 5.91
Dissolved Oxygen in Basin 2, mg/1
Figure 10. PROBABILITY OF BASIN 2 D.O.
6.59
53
-------
99.9
c
-
DO
x
U
Q
H
a
r
~
I •
>,
-
—
0
~
~
99
95
90
80
70
60
50
40
30
JO
10
5
n
^
0
o
-O
o
c
0
S
42.6 45.2 47.8 50.4 53.0 55.6 58.2 60.8 63.4 66TT)
Temperature Of Basin 1, °F.
Figure 11. PROBABILITY OF BASIN 1 TEMPERATURE
54
-------
c
I
T.
—
I-
-
:r
W
J2
a
—
a
.9
99
95
90
80
70
60
50
40
30
20
10
:
i
(
S
c
V
x
V
I*
)
X
w
X
5
x
f c
>
^^
/
r
)
x^
C
)
x^
)
/
C
J
X
^ c
)
/*
X
>
/j
i
x<
>
H
x
c
>
/c
)
c
/
)
43.6 46.0 48.4 50.8 53.2 55.6 58.0 60.4 62.8 65.2
Temperature Of Basin 2, °F.
Figure 12. PROBABILITY OF BASIN 2 TEMPERATURE
55
-------
99.9
H
0)
0
c
;r
u
3
Vi
99
95
90
80
7u
60
50
40
30
20
LO
0.1
7.0 7.2 7.4 7.6 7.8 7.9 8.1 8.3 8.5 8.7
Basin 1 pH
Figure 13. PROBABILITY OF BASIN 1 pH
56
-------
99.9
c
to
•r,
X
J
~
cd
3
~
cd
J3
O
99
95
90
80
70
60
50
40
10
5
0.1
V
o^v
o
6.9 7.1 7.3 7.5 7.6 7.8 7.9 8.1 8.3 8.5
Basin 2 pH
Figure 14. PROBABILITY OF BASIN 2 pH
57
-------
00
Basin 1
Down
Basin 1 - 33% Basin 1 -
20% R.S.
I3asin 1 R.S
Basin 2 Down
Basin 2 Basin 2 - 67% Basin 2 -
80% Both R.S. 80% F.T.
Both R.i
Basin 2
R.S.
O Basin 1
Basin 2
Return Sludge System
Flow Through System
5 15 25
Jan.
5 15 25
Aug.
5 15 25
Feb.
1970
5 15 25
March
-------
150
•i i
u
o
o
' 1
•I
.:
...
4
-,
f >
g
ti
M
fl
i.
•J
-:i
100
Basin 2
R.S.
Basin 1
20%
9 *
Basin 2
80%
Both R.S
Basin 1 -- 33%
Basin 2 - 67%
Both R.S.
Basin 1
20% R.S.
Basin 2
80% F.T.
Basin 1
Down
Basin 2
R.S.
Basin 1 R.S.
Basin 2 Down
50
v
100
System Upset
ing Due
>w D.O.
Bulking Due
To Low D.O.
.S. = Return Sludge System
F.T. = Flow Through System
5 15 25
Aug.
Figure 16. BOD LOADING AND REMOVAL VS TIME
-------
has provided us with several operating relationships. These rela-
tionships are described below and are based on average weekly data.
The three periods of operation are: September 14 - October 4 - Novem-
ber 23 - December 20, and March 1 - March 31. The referenced
figures include least-square curves fit to the plotted data.
A relationship developed between effluent BOD and effluent TSS is shown
on Figure 17. The relationship shows the need for good solids
removal in the secondary clarifier.
Figure 18 shows the effect of MLSS (mixed liquor suspended solids)
levels on effluent TSS. Increases in MLSS increased the solids
loading to the secondary clarifier. It is believed that this was
responsible for increases in effluent TSS.
The relationship between effluent soluble BOD and effluent TSS is
shown on Figure 18. It indicates that some anaerobic metabolism may
take place in the secondary clarifier when loaded with soluble BOD,
resulting in gas production solids rising, and solids carryover.
Figure 19 is a probability plot of BOD removal for all reported data.
During 50 percent of the time, the average BOD removal was equal to,
or greater than, 83 percent.. The average BOD removal for the entire
period was 80 percent even though numerous mechanical problems limited
operating efficiency much of the time.
The effect of BOD loading (Ib BOD/lb MLVSS/day) on BOD removal is
shown on Figure 20. Figure 21 shows the effect of this same BOD
loading on effluent BOD. As the BOD loading increased, the effluent
BOD appeared to increase, while the BOD removal remained above 94
percent.
Figure 22 shows the effect of BOD loading (Ib BOD/1000 cu ft/day)
on BOD removal. As the BOD loading increased, the BOD removal
appeared to decrease slightly; however, additional testing will
be required to verify this relationship. MLSS concentrations are
not considered on this figure.
Clarifier-Thickener - Performance of the clarifier-thickener has
varied considerably. Poor underflow pump performance has prevented
frequent solids withdrawal and has reduced removal efficiencies.
Waste activated sludge discharge to the clarifier-thickenei has
hindered silt settling and has reduced removal efficiencies.
Coagulation studies have been conducted on mixed waste activated
sludge and silt water, and coagulants may be added to the waste
activated sludge in the future to increase removal efficiencies.
COD and BOD removal efficiencies of 80 percent have been obtained,
and TSS removals have exceeded 90 percent with silt water alone
flowing through the clarifier-thickener.
60
-------
160
140
120
100
M
E
H
O
pg
4J
c
"J
w
80
-
S! 60
40
20
Average Effluent Total Suspended Solids
Figure 17. EFFLUENT BOD VS. EFFLUENT TSS
61
-------
10,000
oo
E
00
00
o
SO
3
M
OJ
SJ3
a
o
=2
o
CO
O
00
8,000
6,000
4,000
2,000
80
60
40
30
20
Average Effluent TSS
Figure 18. EFFLUENT SOLUBLE BOD AND MLSS VS. EFFLUENT TSS
62
-------
c
(fl
09
M
0)
o
4J
td
^s
o
99
95
90
80
70
60
50
40
30
20
10
5
o o
A
/_
o
46 52 58 63 69 74 80
Average BOD Removal, %
Figure 19. PROBABILITY OF BOD REMOVAL
86
92
98
63
-------
100
96
92
88
• :
84
c
-
1
00
-
h
ii
80
76
0.0
0.2 0.4 0.6 0.8
Average Loading, Ib BOD/lb MLVSS/d
1.0
1.2
Figure 20. BOD REMOVAL VS. BOD LOADING
64
-------
280
210
50
B
O
O
PQ
C
U
a
11
M
10
(j
OJ
140
70
0.0
0.2 0.4 0.6 0.8
Average Loading, Ib BOD/lb MLVSS/d
1.0
1.2
Figure 21. EFFLUENT BOD VS. BOD LOADING
65
-------
100
96
92
85
•: :
DQ
>
-------
Vacuum Filtration - The solids concentration of the clarifier-
thickener underflow averaged about 48 percent and the vacuum filter
cake averaged 62 percent solids without coagulants. When an anionic
polymer was added to the silt water entering the clarifier-thickener,
the underflow solids increased to about 53 percent and the filter
cake solids increased to about 72 percent. The addition of waste
activated sludge directly to the vacuum filter has not been successful.
OPERATING PROBLEMS
Numerous mechanical problems have been encountered at the secondary
waste treatment facility. As a result, problems have developed
within the biological and physical processes. The mechanical and
process problems encountered to date are discussed below, along with
attempted solutions to these problems and their present status.
Mechanical Problems - The major mechanical problems, equipment
modifications, and the present status of the problems are shown
in Table 2.
Process Problems - Problems encountered in the activated sludge
process have been primarily associated with sludge bulking. This
has occurred several times and appears to be related to frequent
aerator problems, with accompanying large fluctuations in D.O.
levels and recycling of waste activated sludge through the clarifier-
thickener. Increased D.O. levels, increased sludge wasting, and
reduced influent BOD to basin MLVSS ratios have been used to eliminate
the bulking sludge when it appears.
Foaming has occurred on the aeration basins at low MLSS levels, but
has not been a problem at higher MLSS levels, while maintaining a
high D.O.
Addition of an anionic polymer to the silt water, ahead of the clarifier-
thickener, has improved the solids removal efficiency of the clarifier
on silt water; however, the anionic polymer has no effect on waste
activated sludge settleability. A cationic polymer may be added to the
waste activated sludge in the future to aid settleability of the waste
activated sludge-silt water mixture.
SUMMARY
Despite numerous mechanical problems, BOD removals of over 90 percent
have been obtained for sustained periods of time, demonstrating the
applicability of the activated sludge process for treating potato
processing water.
High pH values of influent process water were buffered in the aeration
basins and were not detrimental to treatment efficiencies.
Low temperatures have not caused failure of the activated sludge process.
Data obtained do, however, demonstrate the need to consider temperature
loss in system design. Temperature loss relationships will be
developed in the final report.
67
-------
TABLE 2
THE R.T. FRENCH COMPANY
MECHANICAL OPERATING PROBLEMS
MECHANICAL PROBLEMS
I AERATION BASINS
A. HYPALON TOP LINING
1. LINING WOULD NOT STAY BELOW WATER SURFACE.
2. LINING CAUGHT IN WIND AND TORE IN SEVERAL
PLACES.
3. PIECES OF LINING CAUGHT IN AERATORS.
B. PVC POND LINING
1. LINING FAILED.
2. PARTS OF LINING BECAME BRITTLE.
3. PIECES OF LINING CAUGHT IN AERATORS.
C. AERATORS
1. AERATORS OVERDREW AMPERAGE.
2. VARIATOR SYSTEMS BECAME INOPERABLE - WATER
ENTERED MOTOR SEALS.
3. ICE BUILT UP ON ELECTRIC LEADS AT AERATORS.
4. AERATOR MOTOR BEARINGS FAILED.
5. AERATOR LEGS BROKE OFF.
II SECONDARY CLARIFIER
A. SLUDGE WITHDRAWL MECHANISM ROTATION WAS JERKY.
B. WASTE SLUDGE PUMP BECAME INOPERABLE.
C. BEARINGS OF RECYCLE PUMPS.
EQUIPMENT MODIFICATIONS
LINING WAS REMOVED.
PONDS WERE DRAINED AND
LINING WAS REMOVED.
HEIGHT OF DISCHARGE PORT
WAS INCREASED.
NEW VARIATOR MOTORS WERE
INSTALLED AND VARIAC SYSTEM
WAS MOVED UP TO MOTOR MODULE.
LENGTH OF ELECTRICAL LEAD
STANDPIPE WAS SHORTENED.
BEARINGS WERE REPLACED.
LEGS WERE REPLACED.
SEALS WERE ADJUSTED AND
ADDITIONAL BRACING WAS
INSTALLED.
PUMP WAS REPAIRED AND SAND WAS
REMOVED FROM WASTE SLUDGE LINE
REPLACED BEARINGS PUMP NO.1
PRESENT STATUS
OF PROBLEM
LINING WILL BE REPLACED
WITH ROCK SURFACE
BASIN 2 LINING WAS REPLACED
3 APRIL
BASIN 1 LINING WILL BE REPLACED
N EARLY MAY.
SAME PROBLEMS REMAIN.
INOPERABLE.
OK
BEARINGS ARE STILL FAILING
FUTURE FAILURES EXPECTED
OK
OK
OK- REPLACE PUMP NO. 2
BEARING THIS SUMMER.
Ill CLARIFIER - THICKENER
A. RAKES STOPPED TURNING.
B. UNDERFLOW PUMP BROKE DOWN.
THICKENER WAS DRAINED
AND SILT WAS REMOVED.
BUILT UP PUMP SUPPORTS AND
REBUILT PUMP
SILT MUST BE REMOVED
FREQUENTLY. SCHEDULED FOR
CORRECTION THIS SUMMER.
STILL HAVING TROUBLE
PUMPING SOLIDS
IV VACUUM FILTER
A. SLUDGE WOULD NOT COME OFF CLOTH MEDIA WITH
RUBBER SCRAPERS.
B. CAKE STUCK TO MEDIA.
C. SCREW CONVEYOR HOUSING
HAD NO ACCESS FOR CLEANING.
D. OPERATION WAS NOISY.
E. SCREW CONVEYOR DRIVE MOTOR BROKE DOWN.
F. FILTER CAKE FROZE OUTSIDE BUILDING, BINDING AUGER.
G. FILTER FEED PORTS DID NOT ALLOW THICK UNDERFLOW
TO'PASS QUICKLY ENOUGH TO FILTER.
H. SCREW CONVEYOR HANGER BEARING AND STUBB
SHAFT WORE EXCESSIVELY.
I. SCREW AUGER WAS BENT WHEN INSTALLED. RESULTING
IN EXCESSIVE RUBBING AND WEAR ON AUGER HOUSING.
REPLACED CLOTH MEDIA
WITH STAINLESS STEEL AND
REPLACED RUBBER SCRAPERS
WITH PLASTIC SCRAPERS.
STARTED ADDING COAG.
TO INCOMING SILT WATER.
NEW U-SHAPED HOUSING WAS
CONSTRUCTED.
INSTALLED UNDERGROUND
MUFFLER ON EXHAUSE PIPING.
MOTOR WAS REPAIRED.
WRAPPED HEAT TAPE AROUND
CONVEYOR.
PORT SIZE WAS ENLARGED.
TEE DISTRIBUTION PIPE WAS
INSTALLED.
BEARING AND SHAFT WERE
REPLACED SEVERAL TIMES AND
BEARING LOCATION WAS MODIFIED
AUGER WAS REMOVED AND
STRAIGHTENED.
STAINLESS STEEL MEDIA
IS WEARING.
OK
NEW BELT CONVEYOR TO BE
INSTALLED.
OK OUTSIDE. BUT STILL NOISY
INSIDE. ADDING INSULATION.
OK
OK
NO GOOD.
STILL UNSATISFACTORY.
NEW BELT CONVEYOR TO BE
INSTALLED.
V MISCELLANEOUS EQUIPMENT
A. BEARINGS OF INFLUENT PUMPS FAILED.
B. SUMP PUMP BEARINGS FAILED.
REPLACED BEARINGS.
REPLACED WITH BALL BEARINGS.
OK
OK
68
-------
Silt water can readily be concentrated in a clarifier-thickener
and dewatered with a vacuum filter. Addition of waste activated
sludge to the clarifier-thickener decreased removal efficiencies,
but chemical coagulants can be added which may reduce this problem.
Operation of the clarifier-thickener, vacuum filter, and solids
conveying system has demonstrated the need for special design for
potato processing wastes.
69
-------
USE OF FUNGI IMPERFECTI IN WASTE CONTROL
by
Brooks D. Church and Harold A. Nash*
Studies have been conducted by North Star Research and Development
Institute under a grant by the Federal Water Quality Administration
and four participating food processing companies (Green Giant Com-
pany, General Mills, Inc., Central Soya Company, Inc. and Ralston-
Purina Company) to investigate the digestion of corn and soy waste
streams by fungi.
The rationale for the use of fungi in treating waste streams from
food processing plants has been that of incorporating the dissolved
and suspended nutrients into a macroscopic organism which can be
filtered out readily. A second hope is that the fungal mycelium
will have utility as a feed so that its sale will help defray treat-
ment costs. In order for a process using fungi to meet these objectives
it was considered necessary that it have the following characteristics:
1. The fungi must be capable of reducing BOD to low levels.
2. The waste digestion cycle should be a short one.
3. The fungi must be capable of maintaining themselves as the domi-
nant organism in spite of the nonsterile nature of the waste streams
and environmental variations e.g., temperature, pH, feed rate, nutrient
level, etc.
4. The mycelium must be readily recovered from the digestion mixture.
5. If the mycelium is to be an acceptable feed capable of commanding
a continuing place in the feed market, it must be nontoxic, of high
protein content, digestible, and bland in character.
*North Star Research and Development Institute, 3100 38th Avenue South,
Minneapolis, Minnesota 55406.
71
-------
METHODS AND MATERIALS
Waste streams used in the investigations were obtained from corn
canning operations and from manufacture of soy protein. Waste
stream supplies were drawn, Immediately frozen and kept in the
frozen state until just before use. Waste streams from soy protein
manufacture included one in which the protein had been precipitated
by hydrochloric acid and one in which the acid used was sulfur dioxide.
Our general approach was that of selecting cultures with promising
capabilities on the corn and soy waste streams in shake flask cultures
and testing them in detail in a continuous configuration. The
continuous culture apparatus is shown diagramatically in Figure 1.
The fermentation vessel was a topless 20-liter polyethylene jug
inverted over a 300 mm glass funnel. The feed solution was stored
in a cold room and pumped into the fermentation continuously. This
produced a continuous effluent which was passed through a nylon mesh
to remove fungal mycelium. Aeration and stirring of the fermentation
was accomplished by the air fed into the system through stone spargers.
When addition of nitrogen or phosphate was required, the additions were
made to the feed reservoir. Sulfuric acid additions were similarly
made to the feed reservoir in amounts shown by previous experience to
yield operating pH values in the range desired. Additional amounts
were added occasionally when the pH drifted upwards. No attempts
were made to maintain sterile operating conditions, and the feed was
nonsterile.
Measurements made in monitoring the fermentation included regular
measures of influent and effluent COD, occasional measures of 6005,
regular microscopic examination, regular measures of pH, temperature,
and mycelial mass. Phosphate determinations were made by the Fiske
and Subbarow method, protein determinations by the Lowry method,
carbohydrate measures by the phenol sulfuric acid method, and nitrogen
determinations by the micro-Kjeldahl method.
RESULTS
Strain Selection
Forty-eight strains from eighteen genera of fungi obtained from
Dr. Gray at Southern Illinois University, Dr. Emmons at Natick Labora-
tories, and Dr. Hesseltine at Northern Regional Laboratories of the
United States Department of Agriculture were screened for their ability
to grow rapidly, reduce BOD effectively, and efficiently convert waste
stream materials into mycelium of high protein content.
These fungal candidates were grown in sterile and nonsterile corn and
soy waste in Erlenmeyer flasks on a rotary shake rack at various pH's,
temperatures, and with or without the addition of certain inorganic
salts.
72
-------
Pressure gage
LO
Oilless
air
compressor
(The latched lines represent that
part of the apparatus maintained
at 4°C)
Nylon
filter
Secondary
continuous
digestor
Figure 1. FERMENTOR APPARATUS FOR CONTINUOUS DIGESTION OF CORN AND SOY WASTE BY FUNGI
-------
The fungal strains best meeting the criteria described above are
shown in Table 1 together with their optimal growth conditions.
Table 1. Characteristics of Corn and Soy Fungal Digestion Systems
Substrate
Corn
Soy
Fungi
Optimal pH
Optimal temperature °C
Additional supplements
*Trichoderma
viride 1-23
3.5
30
Nitrogen
Phosphate
Gliocladium
deliquescens
1-31
3.5
30
None
I numbers are William D. Gray's strain designation.
Continuous Culture - Corn
Continuous culture treatment of corn wastes with Trichoderma viride
1-23 was highly successful. The operation of the fermentation over a
140-day period is shown in Figure 2 in terms of COD in the effluent
solution after filtering out the mycelium on a nylon mesh. The COD of
the feed was initially 3700 to 3900 mg/1 and in the latter period of the
run was 5200 mg/1. The feed rate was 15 to 18 ml/min, which corresponds
to a turnover time of 17 to 20 hours. The temperature was in the range
25-28°c. The COD of the effluent was generally maintained at levels
below 200 mg/1 but with occasional escapes to higher levels. These
escapes were associated with definite events in the feed of the fermenta-
tion. At points A and B, the feed line clogged, and the culture went
into starvation with some lysis. At points C and D, the feed line valve
stuck in the open position (feed was controlled by intermittent operation
of a solenoid), and the fermentation was overwhelmed with.fresh influent.
In area E, excellent stability was achieved by decreasing the phosphate
additions to a lower level. Our accumulated experience has been that
feed interruptions of longer than 12 hours were likely to result in
increased COD in the effluent.
74
-------
J I I I I I I I I I I I I I
3.5
3.0
2.5
o
rH
x 2.0
a
o
1.5
1.0
0.5
10 20 30 40 50 60 70 80 90 100 110 120 130 140
DAYS
Figure 2. CONTINUOUS DIGESTION OF CORN WASTE BY T. VIRIDE
75
-------
An analysis of fermentation behavior during period E is shown in
Table 2. The removal of 6005 is seen to have been over 99 percent.
Phosphate and nitrogen were removed to low levels. Sulfate was
present in higher concentrations in the effluent than in the influent
because of additions of sulfuric acid and ammonium sulfate. Even so,
48 percent of the total sulfate was removed in the mycelium. The
mycelial dry weight was equivalent to about 55 percent of the BOD5
consumed.
Table 2. Continuous Culture Digestion of Corn Wastes by a Strain
of Trichoderma viride
Test «
COD
BOD
Carbohydrate (total)
Protein (Folin)
Nitrogen (Kjeldahl)
Phosphate (total soluble)
Sulfate
Chloride
Mycelial weight
Solids in feed
Solids in effluent
Ash
Milligrams per liter
Feed
5200
3976
3500
200
96
32
120
784
0
4000
980
Feed
Addition
22a
13a
280
Effluent
196
35
64
7.5
2.4
0.15
210
55
2200
760
510
Percent
Reduction
96.2
99.2
98.2
96.0
98.8
99.6
48.0
93.9
81.0
52.0
aAmmonium sulfate added as nitrogen source to give mycelium of high
protein content and NaH2P04 was the form of phosphate addition. Values
in table correspond to actual ni-trogen and phosphate mg/1 added.
Continuous Culture - Soy
Soy waste streams were of two types, depending on the acid used in
reprecipitating soy protein. In one stream hydrochloric acid had been
used and in the other, sulfur dioxide. Relatively little effort was
devoted to the soy whey from hydrochloric acid precipitation of soy
protein, and success was only moderate. Eighty to ninety percent
reduction of COD was obtained, but difficulty was encountered in
maintaining stable fermentations. Fermentations began well enough
but soon deteriorated, and competing organisms appeared. The mycelium
showed evidence of ageing and starvation, even though the COD of the
effluent was above the levels we hoped to attain. Difficulty was also
encountered with mycelium gathering at the top of the fermenter in
bulky masses.
76
-------
Since S02~soy whey was judged to be the more important problem
commercially, the fermentation difficulties were pursued with this
medium with the rationale that successful solutions for the S02~soy
whey could be applied to the HCl-whey. Tolerance to SC»2 was low
for many Fungi Imperfect! strains, but it proved possible to select
strains of Aspergillus oryzae and Gliocladium deliquescens which
would tolerate S02 levels as high as 700 mg/1. Repeated passage
through media containing incrementally higher levels of SC>2 resulted
in improved tolerance to S02- The fermentations destroyed S02« The
strains which tolerated the highest S02 levels and would give good
COD removal were selected for further study.
As in the case of HCl-soy whey, difficulty was encountered in
maintaining a stable fermentation. After successful start-ups,
cultures showed evidence of lysis, and yeast became prominent.
Two expedients were found successful in preventing such culture
deterioration. Both were based on the premise that the deterioration
resulted from starvation of the fungus even at moderately high COD
levels. One expedient was to remove fungal mass from the fermentation
to keep the total biomass in the fermentation relatively low. The
other was to increase the feed rate whenever the mycelium appeared
slightly vacuolated and granular.
The results of a continuous fermentation of S02~soy whey using
Gliocladium deliquescens and the technique of regularly removing
part of the mycelial mass are shown in Figure 3. Mycelium was
removed whenever the level climbed above 4 g/1. This removal was
in addition to that achieved by continuous overflow from the fermenter.
The feed rate was 5 ml/min of a feed containing 12,200 mg/1 COD.
Since the fermenter was of 18 1 size, this represented a retention
time of 60 hours. The S02 level was adjusted by mixing HC1 soy whey
with the S02~soy whey. The S02 level of the feed was increased in a
stepwise fashion as indicated at 3 points in Figure 3. The tempera-
ture was in therangv. 25-28°C. The only chemical addition was sulfuric
acid to adjust the pH.
The fermentation was begun by an inoculation dilution technique in
which the fermenter was filled with soy whey diluted to five percent
of its normal concentration; mycelium was added, and feeding with soy
whey b,egun. The changes in COD expected if no microbial action occurred
are shown by the line marked "theoretical".
The effluent from the fermenter was passed through a nylon mesh to
remove mycelium and then through a second stage fermenter consisting
of a 4-liter flask inoculated with a fungal and a bacterial isolates
obtained from a soil enrichment culture. The medium of the soil enrich-
ment culture was effluent from the Fungi Imperfecti fermentation. This
second stage was aerated but not otherwise attended. COD levels from
the second stage effluent are also shown in Figure 3.
77
-------
12,000
12,000
10,000 _
- 10,000
8,000 -
8,000
60
E 6,000
6,000
00
4,000
2,000
Solid line represents the theoretical COD
if no removal was achieved. The fungal mass dry
weight is represented by t • , the secondary
stage COD reduction by 0 0, and the primary
G^. deliquescens) stage COD A A. BOD's were
determined as shown by arrows. The S02 concen-
Theoretical
trations in mg/1 in the feed are also indicated
by the appropriate arrows.
Feed Rate: 5 ml/min
S02
760 mg/1
S02
570 mg/1
10
3
M
3
_ 4,000
- 2,000
30
days
02 46 8 10 12 14 16
Figure 3. CONTINUOUS DIGESTION OF S02 SOY WHEY BY G_. DELIQUESCENS. FUNGAL MASS CONTROL OF FERMENTATION.
-------
The fermentation was successfully operated for 30 days before
being discontinued. COD levels were reduced from 12,000 to about
1,100 in the first stage. BOD5 levels were reduced to about 250.
The second stage reduced COD and BOD5 levels to about half these
values. Total mycelium produced was about 5 grams per liter. An
analysis of the performance on day 7 is shown in Table 3.
Table 3. Reduction of the Chemical Components of S02 Soy Whey
by G. deliquescens
Test
•
COD
BOD5
Carbohydrate
Protein
Phosphate (total)
Nitrogen
(Kjeldahl)
S02
Chlorides
Fungus
Solids
Ash
Feed
Soy Whey
mg/1
12,230
8,537
5,450
3,950
144
1,514
525
848
-
11,650
2,397
After Fungal
Digestion
mg/1
Pri.
Stage
808
235
215
420
43
148
20
550
3300
2800
1620
Sec.
Stage
648
125
% Reductions
Pri.
Stage
93.4
97.3
97.0
89.4
70.0
90.2
96.0
35.0
-
76.0
32.0
Sec.
Stage
94.7
98.6
Stabilization of the fermentation by controlling feed rates in
accordance with appearance of the mycelium resulted in feed rates
equivalent to 24 hours turnover time and COD levels in the- effluent
of about 1500. COD levels of the influent were 12,200. Results are
shown in Figure 4. The buildup of mycelium to extremely high levels
must be explained by a filtering action at the overflow point. Some
difficulties were experienced with foaming. The rise in COD at the
end of the experiment was due to starvation of the mycelium. In
retrospect, it is apparent that feed levels were reduced when they
should have been increased.
79
-------
12,000
10,000
8,000 -
6,000 _
00
o
00
E
f
O
O
o
4,000 -
2,000 _
I i i
I l
Feed rate changes indicated
by arrows pointing 5-7, 9-11, 11-13, __
13-15 ml/min were used to control the
fermentation. The solid line indicates
the theoretical COD which shows
break where the feed rate was
increased to 7 ml/min. The fungal dry
weight is indicated by the 0 0 line
and the actual COD by the A A line.-
Tneoretical
SP, on the fungal curve, indicates the
area where fungal sporulation occurred.
SO 2
760 mg/1
S02
570 mg/1
SOn 400 mg/1
I I I I I I I
I I I I
12,000
_ 10,000
8,000
_ 6,000
tn
D
so
a
_ 4,000
- 2,000
DAYS
Figure 4. CONTINUOUS DIGESTION OF S02 SOY WHEY BY G. DELIQUESCENS.
-------
Aeration^Requirements
Aeration requirements were calculated from the disappearance of
dissolved oxygen after interrupting aeration of a stabilized
fermentation. The assumption required in this method is that the
COD reduction during the period of measurement be the same as the
average COD reduction per unit of time before the aeration was
interrupted. A mechanical stirrer was used to provide gentle
agitation and homogenization of the digester components during
interruption of aeration. Curves for the consumption of dissolved
oxygen are shown in Figure 5. In the case of corn waste, the feed
rate was 20 ml/min of material containing 3650 mg/1 COD. The
effluent assayed 204 mg/1 COD. Oxygen usage in an 18 1 fermentation
was seen to be 10.2 mg/min (0.57 mg/min/1 x 18 1). COD reduction
was 70 mg/min. This calculates to a usage of one pound of oxygen for
every seven pounds of COD removed.
Data for calculation of oxygen usage for the soy fermentation were
a feed rate of 4 ml/min into a fermentsr of 15 1 volume. The in-
coming feed contained 10,320 mg/1 COD, and the effluent contained
2600 mg/1. The oxygen exhaustion curve shows disappearance of
0.33 mg 02/min/l. These data indicate 5.5 pounds of COD removed
per pound of dissolved 02 used.
Properties of Mycelium
The mycelium yielded by the fermentations was threadlike in nature
and readily filtered from the effluent by screens of up to 50 mesh
size. It drained by gravity to about 12 percent solids and with
vacuum filtration yielded a mass of 15 to 17 percent solids. The
composition of _T. viride mycelium grown on corn canning wastes is
given in Table 4.
Table 4. Proximate Analysis of T_. viride Mycelium on a
Dry Weight Basis
Percent
Nitrogen 9.5
Protein a 42
Fiber 9
Ash 9
Fat 4
O
calculated from amino acid analysis
81
-------
B
c
-------
The amino acid composition together with those of casein and high
lysine corn for reference are given in Table 5. It is to be noted
that the analyses are in terms of amino acids per 16 grams of nitro-
gen. Since a substantial part of the nitrogen of the fungal mycelium
is in materials other than protein, the total amino acids per 16 grams
of nitrogen is considerably less than the expected 100 grams. In the
case of T_. viride 1-23, the total is only 67 grams and in the case of
(J. deliquescens, 83 grams. This means that the amino acid values given
in the table for T_. viride can be multiplied by a factor of about 1.5
to give the grams of amino acid per 100 grams of protein. The amino
acid distribution is quite favorable and compares well with high-lysine
corn. Especially notable are a good lysine content, a high tryptophan
content, a favorable leucine-to-isoleucine ratio, and a high threonine
content.
Feeding Studies
The possible usefulness of the mycelium as a feed was probed in feeding
experiments with weanling rats. The experiments were minimal in scope
in that they used only three rats per group and were of three weeks dur-
ation. The principal objectives were to determine protein digestibility
and utilization. Secondary objectives were to test palatability and
look for any evidences of toxicity. Fungus was used as the sole pro-
tein source in the test diets except that selected amino acids were
added to adjust the amino acid composition of the test diet to equal
that of the casein control diet. Amino acids added were of L-methion-
ine, L-serine, L-valine, L-leucine, L-tyrosine, and L-glutamic acid.
Additions totaled less than 5 percent of the amino acids of the test
diet. The total protein of test and control diets was 23 percent.
The remainder of the diets was made up of starch, vegetable oil, min-
erals, and vitamins. Growth curves on individual rats are shown in
Figure 6. It is seen that the average growth rate of control rats
and test rats were identical. The digestibility of the fungal protein
as determined by feed and fecal analyses was 90 percent. The net nit-
rogen utilization as determined by feed, feces, and urine analyses was
50 percent. That of the casein diet was 75 percent. It is important
to note that about one-third of the nitrogen was nonprotein and so
would not be expected to be utilized. It is also to be pointed out
that the digestibility may be unrealistically high compared with that
to be expected of a commercial product. The mycelium was broken in
a Waring Blendor, then freeze-dried. On the other hand it has been
observed that lysis can readily be produced by exposure to distilled
water. It may be possible to release the protein before drying even
in a commercial product.
83
-------
Table 5. Amino Acid Analysis of Two Fungal Strains Compared to
Several Standard Proteins (grams/16 grams nitrogen)
Amino Acid
Lysine
Histidine
Arginine
Aspartic Acid
Threonine
Serine
Glutamic Acid
Proline
Glycine
Alanine
Cystine
Methionine
Valine
Isoleucine
Leucine
Tyros ine
Phenylalanine
Tryptophan
Trichoderma
viride
on Corn Waste
3.94
1.67
2.98
6.49
3.94
3.51
8.98
4.34
3.88
4.76
1.38
1.20
4.48
3.52
5.35
2.44
2.76
1.80
Gliocladium
deliquescens
on Soy Whey
6.15
2.33
5.17
8.41
4.86
4.71
9.47
4.25
4.39
5.99
1.42
1.19
4.85
4.06
6.18
3.29
3.96
2.31
Casein
8.0
3.0
4.0
7.0
4.7
6.7
25.0
11.0
2.5
3.0
1.0
3.5
7.7
6.5
9.7
6.5
5.9
1.2
Soy
Bean
Meal
6.6
2.5
7.0
8.3
3.9
5.6
18.5
5.0
3.8
4.5
1.2
1.1
5.2
5.8
7.6
3.2
4.8
1.2
Opaque-2
Corn
4.2
3.5
6.8
10.0
3.3
4.3
18.7
8.6
4.8
6.5
1.7
1.4
4.9
3.2
8.4
3.9
4.4
1.3
The protein content of these fungal mycelia is between 42 and 45 percent,
based on the amino acid analyses.
Black and Boiling, Amino Acid Handbook, 1960, was used for the amino
acid reference values.
84
-------
150
CO
O
•a
•rl
-------
Aeration cost estimates are based on the laboratory finding that 0.14 Ib
of dissolved oxygen was used per Ib of COD destroyed. The assumption
was made that 1 hp hr will provide 2 Ib of dissolved oxygen. The $500
investment per horsepower is meant to cover the cost of the aeration
equipment, the lagoon, and costs of control equipment. This estimate
seems reasonably conservative. The investment has been amortized over
10 years at 8 percent interest. It was assumed that the COD load is
2500 mg/1 and the equipment is in use 50 days per year.
Labor costs were calculated assuming eight hours of labor a day at
$100 cost per day (including overhead) to operate a 2,500,000 gal/day
installation.
Filtering and drying costs represent a gross estimate and are meant
to cover labor, capital equipment, power, and other costs associated
with this operation.
Sales returns assume the product would bring the same price as soy
oil meal with which it compares in protein content and quality.
Estimates for soy waste processing have similarly been attempted
(Table 7). Nitrogen and phosphate supplies are probably adequate
in the incoming feed and so do not need to be added. Aeration
requirements are similar to those of corn per pound of COD removed,
but the costs are lower because the amortization is spread over
constant operation instead of over fifty days operation per year. The
constant operation does raise the need for heat in the winter in northern
climates. No attempt has been made to estimate heating costs because it
is not known whether waste heat would be available from processing
operations.
Table 7. Economy of Soy Waste Treatment
Item
H2S04
Aeration
Investment
Labor
Heat
Filtering and Drying
Selling Price
Amount
0.1 Ib
0.28 Ib 02 /Ib
(1 hp hr =
(Power cost
at $500/hp
$200 a day
product
2 Ib DO)
= 1.5c/kw hr)
Subtotal
Total
Cents/lb
product
$ 0.14
0.16
0.12
0.17
—
0.54
2.00
2.66
$ 3.75
86
-------
Rat feeding experiments with G_. deliquescens mycelium from soy why
were less salutary. The rats refused to eat mycelium grown on SO^-soy
whey. They did consume mycelium from HCI soy whey, but in only moderate
amounts. Growth rates were lower than with the casein control diets,
but only enough mycelium was available for a 7-day test. The results
must therefore be considered inconclusive.
ECONOMIC ESTIMATES
Only crude estimates of the cost of waste treatment by fungi can be
made from laboratory data alone. Larger-scale trials will be required
as a basis for more accurate estimation. Estimates for costs of appli-
cation of fungi to corn processing wastes are summarized in Table 6.
The estimates were based on recovery of 0.5 pound of dried mycelium
for each pound of COD utilized. This is a conservative estimate based
on accumulated experience.
The amount of ammonium sulfate added is sufficient to yield a product
with 60 percent protein if all the nitrogen is converted to protein.
The amount of sodium dihydrogen phosphate was selected to give the
phosphate-to-nitrogen ratio that has given the best control of fer-
mentation. The amount of sulfuric acid is based on experience with
the Green Giant corn waste stream. The amount actually required will
depend, in some degree, on the amount and kind of materials in the
water used in the plant.
Table 6. Economy of Corn Waste Treatment
Cents/lb
Item Amount fungal product
N (NH4)2S04 0.45 Ib $ 0.67
PO. " NaH2?04 0.022 Ib 0.20
H SO, 0.10 Ib 0.14
2 4
Aeration 0.28 Ib dissolved oxygen
(1 hp hr - 2 Ib DO) 0.16
(Power cost = 1.5
-------
A plant handling 3,500,000 gallons of waste per day with a COD
load of 8,000 mg/1 has been assumed.
It has been assumed that one pound of dry product is obtained
per two pounds of COD removed.
CONCLUSIONS
The use of Fungi Imperfecti to remove BOD^ from food processing
waste streams with production of a mycelium with value as a by-
product appears promising from these studies. In the case of the
corn stream, the several objectives of achieving a low BOD^ in the
effluent, attaining a short digestion cycle, attaining a stable
fermentation, producing a mycelium that is easily recovered, and
producing a mycelium that has potential feed value seem to have
been satisfactorily fulfilled. In the case of the soy stream, the
BOD removal was less satisfactory (although still above 97 percent),
and the promise of the mycelium as a feed remains to be confirmed.
The economics of the process appear attractive. In the case of
corn canning wastes, it appears that sale of the dried mycelium as
a feed should cover about 85 percent of the treatment costs. In
the case of soy wastes, a profit should be realized. The more
favorable economics in the case of soy wastes is due to year-round
operation and the adequacy of nitrogen and phosphate in the waste
stream. In either case, the economics depend very much on drying
costs, and these are gross estimates. Arrangements such that the
mycelium could be fed without drying would allow very favorable
economics. A lesser factor in economics is the low aeration
requirements. Our estimates of these requirements are based on
laboratory measurements only and so cannot be considered proved.
The measures of aeration needs were more favorable than expected.
It must of course be realized that actual use of the mycelium in
feed will require more extensive tests to prove its safety. Use
as a human food of high protein content and good protein quality
is an ultimate possibility.
SUMMARY
Forty-eight species of eighteen genera of the Fungi Imperfecti were
screened for those fungal candidates best able to rapidly convert
soluble and suspended organic material (as measured by BOD) from corn
and soy food-processing waste streams to mycelial protein. Rapidly
growing fungal strains were selected which were readily removed
from the digested waste effluents by coarse filtration. Trichoderma
viride, Gliocladium deliquescens, and either Aspergillus oryzae or
88
-------
(J. deliquescens gave the best results on corn, soy and
containing soy wheys, respectively. Optimal growth conditions
included pH of 3.2 to 3.5, and a temperature of 30°C. Oxygen
requirements were relatively low (1 Ib 62/6 to 7 Ib COD removed).
Nitrogen and phosphate additions were required for the corn
digestion system, and additions of sulfuric acid were necessary
to adjust the pH. These studies were done in 125 ml flasks contain-
ing nonsterile corn and soy wastes. The growth conditions that
resulted in the highest fungal yield and greatest reduction in BOD
and total solids were incorporated into 20-liter continuous culture
digestions. Corn waste was reduced from an initial BOD level of
1600 mg/1 to 25 mg/1 in 24 hours. Soy wastes were reduced from
6200 mg of BOD/1 to 125 mg of BOD/1 in 36 hours in incubation.
Studies of rapid fungal digestion of soy whey containing 700 mg/1
of S02 resulted in selction of A., oryzae and G_. deliquescens
strains which removed S02 from the medium. Mycelial yields were
approximately 50 to 60 g of dry mycelium per 100 g of COD utilized.
The stability of the continuous fermentation with corn waste was
demonstrated in a fermentation run of 140 days' length. Runs of
30 days' length have bean conducted with soy whey. The protein
content of mycelium recovered from the continuous culture corn
digestion system was 42 percent. The recovered mycelium was light tan
in color and bland in taste and smell. Feeding trials in weanling
rats using TT. viride grown in corn waste as the protein source gave
a growth response equal to that seen with a standard casein rat diet.
Digestibility was 90 percent, and no toxicity was observed in a
three-week trial. Feeding trials were inconclusive with rats fed
G_. deliquescens fungal protein from the soy whey fermentation due to
a palatability problem. Economic estimates based on the experimental
results showed the fungal product to be comparable in cost to soil
oil meal.
Results on both soy and corn wastes gave definite encouragement that
the commercial use of selected strains of certain species of Fungi
Imperfecti to remove BOD in a readily harvested form is practical.
89
-------
COMBINED TREATMENT OF DOMESTIC,
AND INDUSTRIAL WASTES BY ACTIVATED SLUDGE
by
John L. Graham and John W. Filbert*
INTRODUCTION
Purpose of Project
Many small cities provide wastewater treatment services to local
industry. In the Pacific Northwest, food processing plants located
within cities often seasonally discharge a greater pollutional load
to the municipal sewerage system than all domestic and commercial
discharges combined. The construction and operation of conventional
wastewater treatment plants to treat the combined domestic and indus-
trial flows can be a financial burden to the citizens and the industry.
In addition, most conventional approaches to treatment plant design
have been found to be inadequate to provide the necessary degree of
treatment, the operational flexibility, and the system stability
required.
A new approach is needed in the design of small treatment plants
which results in minimum costs, maximum effluent quality, and max-
imum flexibility to meet the peak industrial waste loads frequently
encountered. One possibility which may fit these criteria is the
completely aerobic treatment system. A system of this type has
been constructed by the City of Dallas, Oregon, with the aid of a
municipal research and development grant provided by the Federal
Water Quality Administration.
The treatment system constructed at Dallas has not previously been
proven in full-scale operation. There are, for this reason, several
areas in which more information is required to refine design and oper-
ational criteria. The main purpose of this demonstration project is
to provide this information.
Project Objectives
The major objectives of this project are as follows:
To demonstrate the economics and efficiency of the completely aerobic
treatment method when applied to the treatment of combined domestic and
*Cornell, Rowland, Hayes & Merryfield, Corvallis, Oregon 97330
91
-------
industrial wastes, primarily food processing in origin.
To demonstrate the application of aerobic digestion, mechanical
surface aeration and earthen embankment construction.
To demonstrate the flexibility of the system in handling rapidly
changing industrial waste loadings due to the seasonal nature of
canning operation.
GENERAL TREATMENT DESCRIPTION
Waste flow from the City's collection system enters the plant through
a 36-inch diameter influent sewer at the pump station (1). (See Figure
1 for locations). The raw waste is pumped through an 18-inch pipeline
to the headworks (2). The waste is shredded by the comminutor (3)
and then flows to the aeration basin splitter box (4) where it is
discharged to one or both of the 124-foot square aeration basins (5).
The overflow from the aeration basin flows by gravity to the 60-foot
diameter final clarifier (6) where the activated sludge is settled.
The activated sludge to be returned to the aeration basin is removed
from the clarifier by means of six suction pipes mounted on the re-
volving clarifier mechanism and flows by gravity to the return sludge
pump station located nearby (7). Two sludge pumps, sized to return
25 to 100 percent of the plant dry weather design flow, lift the
activated sludge back up to the splitter box (4).
Excess, or waste sludge is collected on the bottom of the clarifier
by a revolving scraper arm and pumped directly to the 90-foot diameter
aerobic digester (8) through one or both sludge pumps located in the
control building (9).
The effluent from the clarifier flows by gravity through flow measure-
ment box (10) to the chlorine contact channel (11), and is finally dis-
charged through the outfall (12) into LaCreole Creek.
The digested sludge is pumped through a sludge pump located in the
control building to the two, 67,000 square foot, humus storage ponds
(13) where it is further stabilized and dried. The supernatant from
the humus ponds flows by gravity to the plant influent line (1).
A schematic diagram of the system is shown on Figure 2. Figures 3,
4 and 5 show the final effluent from the chlorine contact channel, the
aeration basins and the clarifier.
The system treats a combination of cannery wastes, domestic wastes,
a limited quantity of slaughterhouse wastes and an occasional flow
of glue spreader wastes from a plywood plant. The plant was designed
to meet the Oregon State Department of Environmental Quality require-
ments, which state that the effluent to the receiving stream may con-
tain not more than 20 mg/1 each of BOD and suspended solids.
92
-------
o
10
FIGUR-E 1
DALLAS WATER POLLUTION CONTROL PLANT
1. RAW SEWAGE PUMP STATION 8.
2. HEADWORKS 9.
3. COMMINUTOR 10.
4. AERATION BASIN SPLITTER BOX 11.
5. AERATION BASINS (2) 12.
6. FINAL CLARIFIER 13.
7. RETURN SLUDGE PUMP STATION
AEROBIC DIGESTER
CONTROL BUILDING
FLOW MEASUREMENT BOX
CHLORINE CONTACT CHANNEL
OUTFALL SEWER
SLUDGE LAGOONS (2)
-------
VO
c
HEADWORDS
1—CD
RAW SEWAGE
INFLUENT
LEGEND
CONTINUOUS & AUTOMATIC
SAMPLING POINT
OTHER SAMPLING POINTS
CONTINUOUS FLOW METERING
CONTROL VALVE OR GATE
CONSTANT SPEED PUMP
AERATION
BASIN ,
TO PLANT
INFLUENT
EMERGENCY
DIGESTED SLUDGE OVERFLOW _^
CONTROL BUILDING
PUMP ROOM
DIGESTED
SLUDGE
D
AEROBIC
DIGESTER
UNDIGESTED
SLUDGE
AERATION
BASIN
SPLITTER BOX
SLUDGE RECIRC-
ULATION PUMPS
HUMUS POND
NO. 1
SLUDGE
RECIRCULATION
CHLORINE CONTACT
CHAMBER
7
TREATED EFFLUENT
TO RIVER
AERATION/
BASIN
NO. 2
CLARIFIER BYPASS
FIGURE 2
CITY OF DALLAS. OREGON
TREATMENT SYSTEM
SCHEMATIC PLAN
-------
Figure 3. CHLORINE CONTACT CHANNEL AND FINAL EFFLUENT
95
-------
Figure 4. AERATION BASINS
96
-------
Figure 5. FINAL CLARIFIER
97
-------
DESIGN CONCEPT AND CRITERIA
The design was based on the need to minimize the following shortcomings
often associated with conventional activated sludge designs: high con-
struction costs; high operating labor costs; need for highly trained
operators; susceptibility to upset by shock organic, hydraulic and toxic
loads; and waste solids handling, stabilization and disposal problems.
System design was undertaken to eliminate or reduce the foregoing
shortcomings. The primary clarifier and grit removal facility were
eliminated. The aeration capacity was increased to compensate for
the greater organic load to the aeration basins. Grit will be set-
tled out in these basins to be removed periodically by heavy equip-
ment.
Aerobic digestion was employed in place of the higher cost conventional
anaerobic digestion.
Earthen basins, with shotcrete sideslope linings were used in place
of higher cost reinforced concrete construction.
Automation of many plant operations which would normally require
operator attention helps to reduce operating costs.
Employing the complete-mix activated sludge principle, maintaining a
conservative range of organic loadings, and adequate hydraulic surge
capacity within the aeration basins serve to minimize susceptibility
to process upset by shock organic, hydraulic and toxic loads.
Waste solids handling and disposal problems have been minimized by
the use of aerobic digestion and automatically controlled sludge
pumping to the digester and from the digester to the humus storage
ponds.
The design concepts used more nearly parallel those for industrial
waste treatment systems as opposed to strictly domestic waste treat-
ment; e.g., omission of primary clarifiers, use of mechanical surface
aerators and earthen basin construction.
Table 1 lists the treatment plant design criteria.
STUDIES AND OPERATION
Operation
The operation of the plant has been varied to facilitate deter-
mination of treatment efficiency and flexibility. Among these
variations are the following: operation at MLSS levels ranging
from 700 mg/1 to 3,000 mg/1; aeration times between 3.6 hours and
98
-------
60 hours (due to variations in flow); and digester detention times
from 5 days to 30 days.
Sampling and Testing
Influent and effluent samples are composited automatically in pro-
portion to plant flow. Additional samples (see Figure 2) are com-
posited manually during the 8-hour operating shift. All samples
are refrigerated prior to analysis. A specified minimum number of
tests per week are run on an established schedule.
Table 1. Treatment Plant Design Criteria
Population
Present
Design
Biochemical Oxygen Demand (BOD.)
Canning season
Non-canning season
Population equivalent (@ 0.17 Ib BOD/
person/day)
Canning season
Non-canning season
Flow
Average
Maximum
Aeration Basins
Aeration time at design flow
Clarifier
Surface overflow rate
5,500
10,400
7,080 Ib/day
2,080 Ib/day
41,600
12,200
2.0 mgd
6.0 mgd
24 hours
700 gal/day/ft"
99
-------
PRELIMINARY RESULTS AND DISCUSSION
The demonstration program is about 50 percent complete and the
plant has been in operation for about six months. The preliminary
data shows the effluent quality well within the requirements set
by the State Department of Environmental Quality. The average efflu-
ent BOD for the six-month period was 7 mg/1 and the average efflu-
ent suspended solids 12 mg/1. These results were obtained with
organic loading rates ranging from 0.03 to 0.42 Ib BOD5/ Ib MLVSS.
The relationship between organic loading and effluent BOD,, is
shown on Figure 6. The data was plotted on this figure by com-
puter and the curve generated by the method of least squares fit.
The effluent BOD,, was affected very little by changes in the organic
loading. This provides an indication of the flexibility and effici-
ency of the system under various loading conditions.
A further indication of plant flexibility is shown by Figure 7.
The effluent BOD- was little affected by wide variations of influ-
ent BOD,-. The influent waste strength has varied from nearly 300
mg/1 BOD- to 30 mg/1 BOD5.
The chronological correlation between flow rate and effluent suspended
solids (SS) is shown on Figure 8. Suspended solids concentration in
the effluent was generally not affected by changes in the flow rate.
The high effluent SS near the end of November occurred during a period
of relatively low flow. However, this peak coincided with a shock
loading of phenolic glue waste containing a high concentration of
poorly settling solids which were readily carried over the clarifier
weir. The effluent SS peak near the middle of December coincided
with a very high flow rate. The shock load of glue waste adversely
affected the settleability of the activated sludge. The system was
not fully recovered when the high flow occurred and a portion of the
activated sludge was carried over the weir. However, by the last
week in December, the system had completely recovered even though
the flow remained high.
The relationship between pH and alkalinity in the aeration basin is
shown on Figure 9. The correlation appears to be reasonably consis-
tent. That is, a pH drop corresponds to a reduction in alkalinity.
The periods of peak alkalinity coincide rather closely with high
flows. This is because, during high flow, a higher total amount of
alkalinity enters the aeration basin even though the concentration
is less. However, the same amount is removed through nitrification
in the basin as during low flow periods, so a higher concentration
is measured in the basin.
The data shows that a significant amount of nitrification has occurred
in the system, however, it appears that the alkalinity in the influent
flow is enough to sufficiently buffer the system, since the minimum
pH recorded was 6.5. A statistical analysis of the data (Figure 10)
shows that 90 percent of the time the aeration basin alkalinity was
25 mg/1 or greater.
100
-------
o
CO
Ocsj
O
CD
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
BOD LORDING (LB BOD/LB MLVSS)
FIGURE 6
CORRELATION BETWEEN EFFLUENT BOD AND BOD LOADING
101
-------
3s.
INFLUENT
,£FFIUEIU
SEPTEMBER
OCTOBER
NOVEMBER
DECEMBER
UNUIRT
FEBRUHRY
FIGURE 7
CHRONOLOOICflL COMPARISON OF INFLUENT flND EFFLUENT BOD
C5002.7
-------
o
c
o
V)
UJ
O-
Crto
ZD —
CO
Q
C)
r:
\
SEPTEHBER
OCTOBER
NOVEMBER
UNHURT
DECEMBER
FIGURE 8
CHRONOLOGICflL PLOT OF FLOWRATE AND EFFLUENT SUSPENDED SOLIDS DATA
FEBRUARY
C5002.I
-------
Zl/>
»—*
_J
cr
_j
cr
CO
cr
a)
\
X^
A
SEPTtlBER
OCTOBER
NOVEVBER DECEtBER
FIGURE 9
AERflTION BASIN pH AND ALKALINITY
UNUIRY
FEBHUHRV
C500J. )
-------
PROBHBILITY
(PERCENT) g 8
•— MU)-*^cnu>^JC9toco CD * i
noooaoQooocn co CD «
^t
i
0.1
0.01,
1
;
/
)
=x
)
X
«
F ^r
t
{ X
/^
^
' ^
1
4
J
^
•/
4
,x
t
q
4
••/
••A
/>
s
^
5
»***^
c-^
}
L-^O
)
i
3MfOMtrtaJo>«j)ci>ci»*-*-*-*.*-*.wcncno
*<8<»a»«p — tj»*--o«p~M».
-------
Figures 11 and 12 show data for Total Kjeldahl Nitrogen (TKN),
Ammonia Nitrogen (NH_-N), and Nitrate Nitrogen (NO--N) in both
influent and effluent flows. Figure 11 shows a significant reduc-
tion in both TKN and NH -N through the plant, and Figure 12 shows
an increase in NO»-N. The decrease in NH--N with a corresponding
increase in NO«-N shows the occurrence of nitrification in the
system.
Figure 13 shows data for total and orthophosphate (T-PO, and 0-PO,).
The T-PO, curves are difficult to analyze due to missing data. How-
ever, it appears that a slight reduction in T-PO, has occurred. The
average removal over the period was about 10 percent and the average
effluent T-PO^ was 3.8 mg/1.
The 0-PO, curves show a consistent conversion of T-PO, to 0-PO,
4 4,4
as may be expected. The effluent 0-PO, averaged 3.2 mg/1. How-
ever, there was a net reduction in total phosphorous content.
The plant was operated at various MLSS levels, as mentioned previously,
to determine its flexibility. Figure 14 shows the probability of
occurrence of the various MLSS levels. Seventy percent of the time
the MLSS level was between 1,000 mg/1 and 2,200 mg/1. The median
MLSS value for this period was 1,400 mg/1.
Sludge Volume Index (SVI) was used as an indication of the settle-
ability of the activated sludge. A statistical analysis of the
data (Figure 15) shows that the SVI was 90, or less, 92 percent of
the time. However, no apparent increase in effluent suspended solids
was observed with SVI values up to 130.
Figure 16 shows that about 50 percent of the time the flow was 2.0
million gallons per day (mgd), or greater. This is significant be-
cause the average plant design flow was 2.0 mgd. These high flows
were due to the large quantity of rainfall received during December
and January, and a high degree of infiltration into the collection
system.
The high flows were also partially responsible for the low strength
of the influent wastewater as shown on Figure 17. This curve indi-
cates that about 70 percent of the time the influent BOD,, was 110
mg/1, or less.
The plant efficiency is confirmed by the statistical data plotted
on Figures 18 and 19. The total effluent BOD- (Figure 18) was 16
mg/1, or less, 96 percent of the time. About 95 percent of the time
the dissolved BOD5 in the effluent was 7 mg/1, or less (Figure 19).
Waste sludge accumulated at the rate of about 1,800 pounds per day.
This figure is based on calculations using average data obtained
and McKinney's activated sludge equations. The average BOD removal
per day was about 1,800 pounds resulting in approximately one pound
106
-------
-INFLUENT
V
EFFLUENT
en
•r.
.z
u-
-INFLUENT
V
EFFLUENT
C5002.7
SEPTEMBER
OCTOBER
NOVENBER
DECEMBER
JANUARY
FEBRUARY
FIGURE 11
TOTflL KJELDflHL NITROGEN AND flMMONIA NITROGEN DOTfl FOR INFLUENT ftND EFFLUENT FLOWS
-------
tn
13
-EFFLUENT
-INFLUENT
UPTCIBED
OCTOBER
HOVEIIBER
DECEIIEI
IIMUIRT
FEBRUARY
C5002.7
FIGURE 12
INFLUENT AND FTFLUENT NITRflTE NITROGEN DflTfl
-------
-INFLUENT
-
UJ
•f
-EFFLUENT
^•EFFLUENT
-INFLUENT
SEPTEMBER
OCTOBER
NOVEMBER
DECEMBER
)»HU»RY
FEBRUARY
FIGURE 13
TOTAL HMD ORTHOPHOSPHATE DATA FOR INFLUENT AND EFFLUENT FLOWS
C5002.7
-------
95
g*T
i Sfl
t—
>- ^
»- y «•
CD -51 *°
S >- 40
POM
Q_
1—
r^ 20
0
CC
UJ
°- ID
1
01
C
/
5
»*
i
' a
«
•
i
j
«
^
x
c
•
•
) ••
•
k •—
1
/
• s
i
/
! §
t r
i »
/
\ i
* •-
•
7
-
a
x^5
i
! i
X'
> r
CJ
> «
x<
i
> K
i *•
> a
» •-
i
> K
• K.
1 8
^
i 2
) a
'
! &
• A
t »
^
> ^
i a
•>
k .—
^
\ *.
> •.
1 S
\
s
MIXED LIQUOR S- SOLIDS (MG/L)
FIGURE 14
STATISTICAL ANALYSIS OF MIXED LIQUOR SUSPENDED SOLIDS DATA
110
-------
99.g
96
21
iso
co
00 7D
an
j UJ
M so
CD i_
ff^
b_ 4U
9" 30
0. i— ^
. , on
cc
UJ
Q_
* — ' in
j}
Oi
U\
C
•
>
: i
•
• •
*
- 4
• «.
) A
^
s
- a
i *
> a
7
; s
t a
4
II C
J u
I -
*
i
t
{ S
*
4
/
1
t
\ 0
k «,
J
3 a
•
3 C
|
' !
> c
/;
i
i
9 <
' (
> a
)
I S
» a
]
; — —
3 »•
. c
>
: ;
• »•
3 G
3 fl
5- — )
• »
I e
!' — 'J
• *•
• *•
9 a
•
, *
• to
• !».
J »•
•
_>
• r-
' ?
1^^^^^
I 4
• •-
? i!
•
:
i
• —
) 0
R3
Q» —
a us
SLUO&E VOLUME INDEX
FIGURE 15
STATISTICAL ANALYSIS OF SLUDGE VOLUME INDEX DATA
ill
-------
MB
•O
PROBflBILITY
(PERCENT OF TIME LESS THAN)
— i\> u> *• en u> -j w co »
i o ooaaooa a
:^J
;^ *
:^>
r^o
p**'
^^5
*
^
/
3 ,— »*• *- rvj o fo i4 u) u) *• ^ * cn en on o>~o) a
FLOWRRTE (HDD)
FIGURE 16
STATISTICAL ANALYSIS OF FLOi^RATE DATA
112
-------
99.9
99
95
s
T rtn
l—
00 70
>_ 00 'U
r_ uj
« -1 60
-> LU
•~* 5" SO
cr i
o u.
? ° 30
t—
i • i 20
Ld
in
3
;
0.01
»•
0
«
1
/
K
1 -
]
> *
1 »
^
/
: 8
i
<
/
\ ~
^
) Q
> *
; ^
? 2
q
^^
l •-
• (^
>
*
\ **
t
Q
a
^f^"^ t
\ 0
<
i 8
4
}^^
' a
, ^
\ !i
i ik
^
> !
\
) a
! 5
> c
^
c
1 *
•
) N
I 8
r^
> M
! 3
INFLUENT BOO (MG/L)
FIGURE 17
STATISTICAL ANALYSIS OF INFLUENT HASTE STRENGTH DATA
113
-------
Mn
OK
_ in
oo *u
>- UJ
_J LU
•— « ^~ co
**^
^p* K—
p^ 4 y
°°30
°~ ^
ct:
UJ
•«•'
N— <" IU
• 1
0.01.
9
^ t
X
a
i
, ^
r
i
7
n ^
j
i ^^^
•
^
x"
j <
:X
n n
*
^f t
x:
c
\
x
XN
4
«
t
*
1 S
f •*
^ -• ' t
C -3
1- — 3
i^^^^^
>
•
3 K
» •
CO
y,
5S5S5S6bgggggggggggg
"82S8SS"
EFFLUENT TOTRL BOD (MG/L)
FIGURE 18
STATISTICAL ANALYSIS OF EFFLUENT TOTAL BOD DATA
114
-J »
-------
•V***
99.9
99
96
90
cr
nr 80
1—
00 70
.*_ CO
£ ^
« ~"80
-J UJ
5s50
g^40
O 1_L
^ °30
i—
ul 20
cr:
UJ
°- in.
^^ lu;
5
1
O.I
0.01,
•
t
4
i
/
i
'/
*.
t f *
/
\
~?
\
,^
S*
t
/
1
> ^ <
•
2
^3
4
_^--
: —
— -3
C 5
•
' rr"^^?'?>?'f?'?1?l?l???>rir4r
EFFLUENT DISOLVED BOD (MG/L)
FIGURE 19
STATISTICAL ANALYSIS OF EFFLUENT DISSOLVED BOD DATA
115
-------
of waste solids produced per pound of BOD removed. This value is
somewhat greater than would normally be expected for an activated
sludge system because the lack of primary clarification allows a
larger quantity of non-biodegradable solids to accumulate in the
system.
No odor problems have been associated with the humus storage ponds
during this period. This indicates that the digested sludge has been
well stabilized. The drying characteristics of the sludge have not
been determined as yet. This will be studied during the second half
of the program.
The total construction cost for the Dallas plant, including engine-
ering, was $506,300. Comparing this with a value of $950,000 for a
2.0 mgd conventional activated sludge plant obtained from Smith's*
curves shows a capital cost saving of about 47 percent. Operation
and maintenance costs at the Dallas plant for the six-months' period
averaged about $20 per million gallons treated. This compares to
a Smith curve value for a 2.0 mgd conventional plant of about $75
per million gallons treated, or a difference of approximately 70
percent.
Power cost, the largest single operation expenditure, has averaged
about 20 percent less than anticipated. This is due to a low dis-
solved oxygen requirement resulting from small BOD loadings.
Operation labor costs have been minimized by operation on a 5-day per
week, 8-hour per day basis.
The coliform data shows that adequate disinfection of the effluent
has been provided. Dye tracer studies were run on the chlorine con-
tact channel which showed that the actual detention time was greater
than 90 percent of theoretical.
PRELIMINARY CONCLUSIONS
On the basis of the operation data obtained so far, the conclusions
listed below can be made. However, these conclusions are subject
to modification when the complete data has been obtained at the end
of the demonstration program.
1. The process as designed is flexible enough to withstand high
shock organic loads and abrupt changes in loading.
2. Elimination of primary clarification and grit removal causes no
special operating problems.
_ _
Smith, Robert. Cost of Conventional and Advanced Treatment of
Wastewaters. U. S. Department of the Interior, FWPCA, Cincinnati,
Ohio. July 1968
116
-------
3. The effluent quality has been consistently good despite wide
variations in flow and organic loading.
4. Aerobic digestion is an adequate stabilization method for
this combination of sludges.
5. A treatment system of this type can be constructed and oper-
ated at a substantial reduction in cost.
117
-------
AEROBIC TREATMENT OF LIQUID FRUIT PROCESSING WASTE
by
Larry A. Esvelt*
INTRODUCTION
Subsequent to an order from the Washington Water Pollution Control
Commission to remove 90 percent of BOD and suspended solids from
processing effluents, Snokist Growers cannery undertook a two year
construction and research program. The cannery which processes
pears, peaches and apples as major products, obtained a research
and development grant from the Federal Water Quality
Administration to assist in the work. This paper summarizes the
results of the study carried out.
The Snokist cannery had a processing capacity for approximately
250 tons per day of pears or peaches or approximately 100 tons
per day of apples during the study. Their entire product was put
out in cans and glass containers. Apple products include sauce,
slices, and rings. The processing waste was greatest during the
2 1/2 to 3 months of pear and peach processing each fall when flows
of about 2 mSd and BOD outputs (after screening) in excess of 20,000
pounds per day were experienced.
FACILITIES
The existing waste treatment facilities at the start of the study
included vibrating screens with lift pumps, a 1.5 mg earth diked,
asphalt lined lagoon with two 30 horsepower mechanical surface aera-
tors, a small rectangular clarifier with hopper bottom, sludge re-
circulation pump and waste outfall to the Yakima river.
Prior to the first season of study a 6 million gallon earth diked
basin was constructed and piped in parallel with the existing fac-
ility. This basin was lined with 0.020 inch PVC sheet and had
installed four 60 horsepower surface aerators. Subsequent problems
arose with the PVC liner when gas accumulated underneath it which
carried it to the surface in bubbles. The bubbles were punctured
as a temporary expedient to allow continued operation. Later, the
basin was drained and the liner, which had had several seams split
apart, allowing sludge underneath it, was repaired and covered with
a layer of round river rock.
* University of California, Berkeley. At the time of the study:
Sanitary Engineer, Gray & Osborne, Consulting Engineers, Yakima,
Washington.
119
-------
Following the first season of study designs were drawn to add con-
struction to make the system capable of meeting the State's standards.
This consisted of clarification and sludge recycle for the large aera-
tion basin, additional piping so recycled sludge could be reaerated in
the small basin and additional aeration capacity for the large basin.
A 90 foot diameter by 9 foot side depth clarifier with suction sludge
removal was constructed. Two 1750 gallon per minute variable speed
sludge recirculation pumps were installed and a 150 horsepower sur-
face aerator was added to the large aeration basin. In addition a
30 foot diameter compressed air flotation sludge thickener was installed
for wasting biological sludge.
A temporary laboratory was equipped for use prior to the completion of
a new building. After completion the new lab was used to conduct all
testing for the study.
OPERATION
During the first season of study the waste flow was split between the
older small aeration basin and the newer basin, approximately 15 to
20 percent and 80 to 85 percent respectively, during pear and peach
processing (Figure 1). The smaller basin with its clarifier was oper-
ated as an activated sludge treatment unit with biological sludge re-
cycling. Considerable sludge went over the clarifier wiers and it
was difficult to build up the mixed liquor solids level. The larger
basin operated as a flow through unit or "completely mixed aerated
lagoon" with biological solids passing out in the effluent. Nitro-
gen and phosphorus were added to the raw waste stream throughout the
treatment.
At the beginning of the second study year the clarifier was not yet
completed so the large basin once again acted as an aerated lagoon.
After clarifier completion sludge was recirculated to the large
basin to comprise an activated sludge treatment. During the last
portion of the pear processing season sludge was recirculated through
the small basin for reaeration (Figure 2). Nutrients were added
throughout this season also.
Testing throughout the study was according to Standard Methods (1).
Settleable solids were read following one hour settling time in a
1000 ml graduated cylinder. Suspended and volatile suspended solids
were run using glass fiber filters (2). Total phosphorus was performed
according to an alkaline ashing procedure (3). Oxygen depletion was
measured for oxygen uptake determination with the use of a D. 0. meter
and stirred erlenmeyer flask (4).
120
-------
Screened
Waste
Metering a Flow
Distribution Box
I
To River
Activated
Sludge
Aeration
Basin
Clarifier
Aerated Lagoon
Waste Flow
Return Sludge Flow - Activated Sludge
Figure 1. SCHEMATIC FLOW DIAGRAM 1967-1968 PROCESSING SEASON
121
-------
Screened
Waste
Metering & Flow
Distribution Box
I
To River
Sludge
Reaeration
Basin
Aeration
Basin
Waste Flow
Return Sludge Flow—Activated Sludge
Return Sludge Flow—Activated Sludge—with Sludge Reaeration
Figure 2. SCHEMATIC FLOW DIAGRAM - 1968 PROCESSING SEASON
122
-------
RESULTS
The operation of the treatment facilities during the two processing
seasons definitely indicated that a good quality effluent is obtain-
able in the treatment of fruit processing waste. The three "methods"
of treatment tried all demonstrated workability and confirmed that
all three flow sheets were in reality just variations in configur-
ation of a process in which the mechanism of organic removal follows
the same laws of action. All three flow sheets, aerated lagoon,
activated sludge and activated sludge with sludge reaeration deve-
loped a biological solids population in the aeration basin which
acted on the waste according to the same principles. The signifi-
cant differences in effluent quality were due to the amount of
suspended solids (biological solids) contained therein.
Table 1 and Table 2 summarize the results experienced during the
two years of operation of the treatment facilities. The values
given are average for the treatment process and intervals indicated.
As it can be seen from these tables BOD removal is consistently
well above seventy percent during aerated lagoon operation and
well above ninety percent when sludge was recirculated which allowed
better solids removal due to the lower sludge organic loading.
Table 1. Treatment System Performance
1967-1968 Processing Season
Screened Waste Aerated Lagoon Effluent
Flow BOD COD BOD COD
MGD mg/1 mg/1 mg/1 mg/1
Pear Proc. 1.03 2040 3050 370 1040
Peach Proc. 1.36 1810 2150 340
Apple Proc.
(Nov. Dec.) 0.60 1230 1520 190 760
Apple Proc.
(Jan. Feb.) 0.52 950 1400 110 620 470
Activated Sludge Effluent
Pear Proc. 0.21 2040 3050 250 490 375
Peach Proc. 0.30 1810 2150 360 370
Apple Proc.
(Mar. Apr.
May) 0.40 1390 1830 130 570 470
123
-------
Table 2. Treatment System Performance
1968-1969 Processing Season
Screened Waste
Pear Processing
Aerated Lagoon
Peach Processing
Aerated Lagoon
Peach Processing
Activated Sludge
Pear Processing
Activated Sludge
Pear Processing
A.S. with Sludge
Apple Processing
Flow
MGD
1.48
2.02
2.02
1.87
1.58
BOD
mg/1
2080
860
860
1600
2150
COD
mg/1
2870
1510
1510
2290
2910
Effluent
BOD
mg/1
460
115
20
9
4
COD
mg/1
1050
710
120
55
23
Reaeration
0.43
1190
1500
5
28
Activated Sludge
DISCUSSION
Waste Characteristics
The screened waste strength and volume can be seen from Tables 1 and
2. The unit waste contribution as derived from waste testing and
plant production records is shown on Table 3. Table 3 also contains
the ratio of measured BOD to COD byproduct and the average nutrient
available in the screened raw waste as a ratio to the BOD. As it can
be seen by traditional standards the waste is nutrient deficient for
biological treatment without nutrient addition.
124
-------
Table 3. Unit Waste Contribution
Product
Pears
Peaches
Purple Plums
Tomato Juice
Apples
Product
Pears
Peaches
Apples
Flow*
COD**
BOD**
5800 gal/ton
Raw Product
7500 gal/ton
Raw Product
7000 gal/ton
Raw Product
3200 gal/ton
8200 gal/ton
Raw Product
* gal per ton of
** Ibs per ton of
BOD
COD
0.74
0.67
0.75
105#/Ton
Raw Product
98#/Ton
Raw Product
95#/Ton
Raw Product
78#/Ton
Raw Product
66#/Ton
Raw Product
64#/Ton
Raw Product
Negligible compared
to concurrent pro-
ducts being run
71#/Ton
Raw Product
raw product
raw product
P
BOD
0.0012
0.0027
0.0010
53#/Ton
Raw Product
N
BOD
0.006
0.011
0.0023
125
-------
Cost of Treatment
The $543,000 total cost of construction of the facilities can be
broken down by treatment components and amortized to arrive at
unit costs which when added to 0 & M costs and divided by the
waste load become unit costs of treatment (4). These costs shown
in Table 4 reflect the waste levels and season length experienced.
The raw COD and BOD received for treatment each season was approx-
imately 2.0 and 1.5 million pounds, respectively.
Table 4. Estimated Unit
Costs of
Aerated Lagoon
Construction Costs
Annual Cost of Construction
(20 years @ 7% interest)
Power Cost
Nutrient Chemicals
Operation and Supervision
Maintenance Labor & Materials
Laboratory Supplies
Major Repairs and Improvements
Annual Cost of Treatment
Assumed % COD Removal
Cost per #COD Removed
Assumed % BOD Removal
Cost per #BOD Removed
$205,000
19,400
7,000
10,500
6,000
1,500
1,500
3,000
$ 48,900
70%
$ 0.035
80%
$ 0.041
Treatment
Activated
Without
Sludge
Reaeration
$470,000
44,400
10,000
10,500
12,500
2,000
2,500
5,000
$ 86,900
90%
$ 0.048
95%
$ 0.061
Sludge
With
Sludge
Reaeration
$557,000
52,600
10,000
10,500
12,500
2,000
2,500
5,000
$ 95,100
90%
$ 0.053
95%
$ 0.067
126
-------
Substrate Removal
The removal of organic substrate (soluble BOD or COD) by a biological
population is presently generally believed to follow a kinetic model
similar to the Michaeles - Menten equation for enzyme reactions (5).
The use of this equation (equation 1) is dependent upon the assump-
tion that the aeration basin functions closely as a completely mixed
basin. Volatile suspended solids is assumed to represent a measure
of active biological mass.
R •
R = Substrate removal rate; mg BOD, COD/mg MLVSS-day
F = Maximum removal rate
S = Substrate concentration at R = 1/2 F
s = Substrate concentration; mg BOD, COD/1
Since S , is usually large compared to s in a normally operating treat-
ment system, where R is normally much less than one, this equation can
be reduced to a first order expression such as equation (2) .
R = fs (2)
f = First order removal rate constant or "removal coefficient";
mg BOD or COD removed _
mg MLVSS-day-mg/1 BOD or COD
The data derived from this study using BOD as the substrate is pre-
sented on Figure 3, with data corrected to 20°C. A straight line
corresponding to equation (2) is plotted through the data. Consider-
ably more data was available for COD removal and substrate concen-
tration than BOD. This data was plotted by temperature ranges and
the coefficients resulting from each group of data is plotted against
its mean temperature on Figure 4. As it can be seen the points fit
closely the line described by equation (3) . The correction factor
found here was used to correct the BOD removal data for construction of
T— 90
fT = 0.017 x 1.16 (3)
fT - COD removal coefficient at T °C
Figure 3. Figure 3 then yielded the coefficient for BOD removal at
20°C and equation (3a) is derived.
T— 9fl
f = 0.068 x 1.16 (3a)
BOD
f = BOD removal coefficient at T °C
BOD
127
-------
00
0>
&
CO
o
pp
Q
§
i.o -
0.5 -
0
fBODT = °'068 x l'
(T-20)
I
10
Pears - Aerated Lagoon
Pears - Activated Sludge
Pears - Act. SI. w/Sludge Reaeration
Peaches - Aer. Lagoon & Act. Sludge
Apples - Activated Sludge
20
Substrate BOD, mg/1
Figure 3. BOD REMOVAL RATE VS CONCENTRATION AT 20°C
-------
§
o
00
co
CO
>
00
8
o
00
e
B
0)
•H
O
vl
in
IM
<0
O
o
O
Q
O
(J
0.02 _
0.01 _
0.008 _
0.006 _
0.005 -
0.004 _
0.003 -
0.002 -
0.001
f = 0.017 x 1.16
(T-20)
i
5
Q.Q17
Log 0.064 =1.16
O
D
A
Pear Processing
Peach Processing
Apple Processing
Pear Processing -
Activated Sludge with
Sludge Reaeration
10
Temperature, °C
I
15
20
Figure 4. COD REMOVAL COEFFICIENT VS TEMPERATURE
129
-------
There did not appear to be variations in "f" according to product
being processed but data scatter could have obscured small deviations.
Biological Growth
A material balance of MLVSS across a completely mixed basin yields
equation (4). [5] [6] [7]
Net Sludge Growth = c(BOD or COD removed/day)k'MLVSS (4)
c = Yield coefficient, mgVSS formed/mgBOD or COD removed
k = Endogenous respiration rate, mgVSS destroyed/mgMLVSS-day
MLVSS = Mixed liquor volatile suspended solids
Equation (4) can be divided by the mixed liquor volatile suspended
solids and since equation (5) is the definition of removal rate,
equation (6) can be formed.
R = BOD or COD removed/day (5)
MLVSS
G - cR-k (6)
G = Net sludge growth rate, mgVSS/mgMLVSS-day
MLVSS here refers to the total weight of volatile suspended solids
in the treatment system and the net sludge growth rate (G) consists
of volatile suspended solids accumulated in the system, those which
are wasted from the system as carryover in the effluent and those
intentionally removed by wastage, divided by the total VSS in the
system. The reciprocal of the net sludge growth rate is the "sludge
age" of the biological solids in the system.
The data gotten during this study were separated and plotted according
to major product being processed and according to temperature range.
Net sludge growth was plotted vs. COD removal rate. The yield coef-
ficient varied with product being processed but remained constant
with temperature change. The yield coefficients obtained are as follows:
Product Being Processed mgVSS/mgCOD Removed mgVSS/mgBOD Removed
Pears 0.49 0.66
Peaches 0.46 0.69
Apples 0.57 0.76
c = yield coefficient
130
-------
The endogenous respiration rate obtained from the zero removal rate
intercept on the sludge growth axis (negative) varied with tempera-
ture. The values obtained are plotted on Figure 5 vs. temperature.
It can readily be seen that the points very nearly fall on the line
described by equation (7).
1^ = 0.115 x 1.14(T-20) <7)
k^ = Endogenous respiration rate at temperature T,°C
On Figure 5 the next to highest point was obtained from peach proces-
sing data while the lowest was from apple processing data so apparently
no difference in endogenous respiration rate exists between sludge
developed on pear, peach or apple processing waste. There was no
difference in growth or respiration characteristics between sludge
developed in aerated lagoon treatment and activated sludge.
Oxygen Requirements
The oxygen requirements for an aerated system depend upon the amount
of organic matter removed from the waste stream and on the amount of
volatile suspended solids in the aeration system as described by
equations (8) and (9). [5] [6] [7]
02 = a-BOD or COD removed + b-k'MLVSS (8)
02/MLVSS = a'R+b- k (9)
In these equations "k" is the endogenous respiration rate, "R" is
the substrate removal rate and "a" and "b" are constants.
The oxygen uptake rate studies from individual days resulted in
data for various substrate removal rates and at various temperatures.
Using the temperature correction factor for the endogenous
respiration rate, these individual data points were corrected to
20°C and plotted on Figure 6. From Figure 6 the constants a and b
were obtained.
a = 0.34mg02 required/mgCOD removed
a = 0.46mg02 required/mgBOD removed
b = 1.2mg02 required/mgVSS destroyed by endogenous
respiration
131
-------
0.20-
«J
I
to
CO
t>0
CO
CO
00
6
o>
C
O
0)
H
O.
co
-------
10
z
UJ
CD
X
o
0.4-
1
I
CO
en
>
o>
CSt
O
o>
E
ci
o
O
CM
I
LJ
CE 0.2-
LU
0.34
0.2
0.4
0.6
C.O.D. REMOVAL RATE - mg C.O.D. / m g VSS - DAY
Figure 6. OXYGEN UPTAKE RATE VS COD REMOVAL RATE AT 20°C
0.8
-------
Using the constants obtained and loadings experienced, the oxygen
supplied by the surface aerators was computed. They were capable
of supplying 2 pounds of 62 per horsepower hour with a residual of one
mg/1.
Sludge Characteristics
The biological sludge developed on the fruit processing waste was
quite different from sludges normally found in domestic waste treat-
ment plants. The solids were about 90 percent volatile and exhibited
a very slow settling velocity at high concentrations. Figure 7 shows
sludge settling curves for various dates during the 1968 season. The
COD removal rate is given on each plot along with the mixed liquor
suspended solids on that date. With the exception of 9-4-68 it can
be seen that settling improved with a lower removal rate. On 9-4-68
the system was still operating as an aerated lagoon. The supernatant
of the settled sample remained quite cloudy indicating a large portion
of the solids did not settle and demonstrated what has been termed
"deflocculated" sludge [8]. The samples other than 9-4-68 were from
the system when sludge recirculation from the clarifier was being
practiced and very clear supernatant was experienced both from the
settled samples and clarifier.
The clarifier for this treatment system was designed for an overflow
rate of 400 gallons per day per square foot after reviewing the
settling data from the first year of operation. It operated at about
this value or less throughout the study. Solids removal was excel-
lent with a very clear effluent. Return sludge concentrations
exceeded 6000 mg/1 at times and MLSS got as high as 3500 mg/1. During
the 1968 season MLSS got up to 5000 mg/1 [10].
Figure 8 shows the COD and BOD of the biological sludge as mgCOD or
BOD per mgVSS plotted against COD and BOD removal rate. The plot
yields the relationship:
mgCOD/mgVSS =1.39
mgBOD/mgVSS = 0.3 + 0.28 x BOD removal rate
Figure 9 shows the organic nitrogen and phosphorus content of the
sludge measured as volatile suspended solids. The lowest four points
of data for nitrogen and lowest two points of data for phosphorous
were gotten when the residual dissolved nutrient was zero so
deficiency could not be ruled out. It must be concluded that
Organic N/mgVSS = 0.087 mg
Organic P/mgVSS = 0.016 mg
Process Control
Control of the treatment process is mandatory for achievement of
consistent and reliable treatment results. The maintenance of a
134
-------
E
I
tr
ui
UJ
o
Q
Z>
_l
CO
u.
o
UJ
1000
500-1
1000-
500-
1000
500-
S.S. = 2610
10-4 -68
mg C.O.D. REM
R = 0.26
mg V S S -DAY
\
4
\
6
8
S.S = 2590
10- 26- 68
mg C.O.D. REM
R »O.I8
mg V S S- DAY
I
4
I
6
T
8
1000-
500-
1000-
500-
1000
500-
S.S.= 920
9-18-68
mg C.O.D. REM
R= O.67
mg V SS - DAY
I I I I
2468
S.S.= 3260
0- I I- 68
ro g C. 0. D. REM
R= 0.19
mg V S S - DAY
I
2
I
4
i
6
\
&
12- 13 - 68
mg C.O.D. REM
mg VSS- DAY
S.S.= 1680
SETTLING TIME - HOURS
Figure 7. SETTLING CURVES OF AERATION BASIN SUSPENDED SOLIDS
135
-------
CO
co
bo
B
^
Q
0
U
00
B
1
1
1
1
.5 -
.4 _
.3 -
.2
0°
9&>o
"goo o
o
o
00
o o o
1.39 mg COD/mg VSS
00
0 «
CO
CO
I
a
o
oo
0.6 -
0.5 _
0.4 -
0.3 _
0.2
mg BOD/mgVSS = 0.3 + 0.28 x BOD Removal Rate
0.1
0.2
0.3
0.4
0.5
0.6
0.7
o'.s
tf.9
1.0
BOD, COD, Removal Rate, mg BOD/mg MLVSS - Day, mg COD/mg MLVSS - Day
Figure 8. BOD, COD EQUIVALENT OF VSS
-------
C/2
>
e
£5
00
o
e
w
C/l
60
s
p-l
00
V-i
o
00
E
.11 _
.10 _
.09 -
.08 _
.07 -
.06 -
.05 -
.04 -
.03 -
.02 -
.01 .
0
O
0 0&>0° ppo ° 0.087 mg Org N/mg VSS
DCK O O
*"^J V
O
o
0
o
o
A 0.016 mg Org P/mg VSS
._ 9^-,p f n O
°0
o
o
1 1 I 1 1 1 1 1 1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
COD Removal Rate
Figure 9. NITROGEN AND PHOSPHORUS CONTENT OF BIOLOGICAL SLUDGE
-------
high degree of treatment, uninterrupted by conditions of upset which
are capable of affecting the effluent is necessary for assurance of
undamaging influence on the receiving water.
The pH in these treatment units remained virtually constant at near
7 throughout both processing seasons. The nearly completely mixed
condition of the aeration basins and the buffering provided by the
biological system prevented acid or alkaline conditions from occurring.
The influent pH varied from 6 to 8 during pear and apple processing
and from 10 to 11 during peach processing due to the alkaline peel
operation.
Nutrient feeding is necessary with this waste to achieve satisfac-
tory biological treatment. The total nutrient (nitrogen and phos-
phorus) requirement is the amount necessary to supply the biological
yield at the percentage of the constituent incorporated into the
MLVSS. Nitrogen and phosphorus are released by the MLVSS through
organism breakdown by endogenous respiration but it has been observed
that about 20 to 25 percent of the amount in the respired sludge is
not released [9].
Nutrient feeding is normally related to the waste constituents
(COD or BOD) and others have given desirable ratios for N and P
to COD or BOD. Taking the N and P content of MLVSS found and the
yield coefficient a maximum requirement per amount of substrate
removed can be obtained. The total nutrient requirement could then
be reduced by the amount being released per day from organism
destruction by endogenous respiration, which is governed by substrate
removal rate and temperature.
A short run at nutrient deficient conditions demonstrated impairment
of substrate removal efficiency and biological solids (MLVSS) which
exhibited no settling.
Startup of the treatment system results in a lower quality effluent
for about a week. Initially the clean water-filled basin dilutes
the wastes. As solids build up a portion are lost over the clari-
fier wier. By the end of one week the system has usually reached
stability and even during the week of startup the effluent BOD and
COD are considerably lower than in the influent. Temporary shut-
downs of waste feed do not cause upsets in the system so long as the
basins are continually aerated. Shutdowns of a week seem to be
permissible. Changes in product being processed did not adversely
affect the treatment.
Control of solids level must be practiced in the form of solids
wasting to avoid exceeding the capacity of eithe^- 1) the aeration
138
-------
system to support them or 2) the clarification-sludge return system
capacity for solids removal and return. In this system solids are
wasted to a flotation thickener for subsequent land disposal.
Operation of the system in extremely cold weather can lead to prob-
lems. During late December, 1968, and January, 1969, sub-zero
temperatures were experienced. No processing was occurring and the
aeration basin sub-cooled and subsequently froze over. Prior to
their shutdown the aerator platforms had accumulated a considerable
amount of ice. It was not determined whether the aerators could
have been kept in operation if run continuously. Processing
operations did not begin until the basin had thawed.
Sludge Thickening
Sludge disposal is a major problem from an operation of this type.
A flotation thickener with pressurized recycle was installed for
this facility.
During the operating season testing done on the thickener indicated
that the float concentration would not exceed about 2 percent D*S..
During special testing performance of sludge thickening was evaluated
[11] during a period when BOD rate was about 0.4 day "^ and the
volatile content of the solid about 87 percent. Air:solid ratio was
about 0.6 and the feed sludge was 6000 to 7000 mg/1 solids.
Solids Loading Hydraulic Loading Float Solids
lb/hr./ ft2 gpm/ft?
0.4
0.4.
0.5
0.65
0.5
0.6
0.43
0.6
2.5
2.1
2.1
1.9
For a comparison with this sludge thickening performance, the Selah,
Washington municipal sewage treatment plant receives from 30 to 70
percent of its BOD load as screened fruit processing. The combined
waste does not receive primary clarification and is treated by acti-
vated sludge process. Waste activated sludge is also thickened by
compressed air flotation. During 1969 the sludge thickened here
averaged 3.8 percent P.S, without significant fluctuations between
active and less active industrial season [12]. The BOD removal rate
was similar to that at Snokist but sludge volatility ranged from 66 to
76 percent volatile. Solids thickening rate ranged from less than
one to greater than two pounds per hour per square foot and hydraulic
loading rate ranged from 1.35 to 2 gpm per square foot without seeming
to affect the solids concentration. It would appear that
139
-------
the domestic component beneficially affects fruit processing waste
activated sludges thickening characteristics even though other
influences were undoubtedly present.
Summary and Conclusions
In summary, the conclusions drawn from the study of treatment of
fruit processing wastes are as follows:
1. Activated sludge or contact stabilization treatment of fruit
processing waste at low rates (below 0.4 mgCOD removed per day
per mgMLVSS) will provide greater than 90 percent removal of
organic load and removal of solids from the effluent. The activated
sludge process does as adequate a job as contact stabilization and
would be recommended because of requiring less facilities and thereby
lower cost.
2. Aerated lagoon treatment can provide greater than 70 percent BOD
removal but suspended solids remain in the effluent and are the
principle source of effluent BOD and COD. This treatment is con-
siderably less expensive and is recommended if the effluent quality
can be tolerated.
3. "Completely mixed" aeration basins provide effective buffering
to avoid pH fluctuations.
4. Final clarification for activated sludge or contact stabilization
is successful at low surface loading rates (below 400 gallons per
square foot per day), made necessary by the slow-settling sludge
developed, and with vacuum type sludge removal.
5. Nutrient addition is necessary to achieve successful treatment.
Nutrient savings can be made by increasing the amount of biological
sludge in the system, thus decreasing the waste loading rate on the
sludge. This allows a greater destruction of sludge by endogenous
respiration which returns nutrient to the system and cuts down the
amount necessary to add. Activated sludge or contact stabilization
treatment is necessary to accomplish this.
6. Surface aerators accomplished adequate mixing at 0.3 hp/1000
cubic feet and provided 2 pounds of oxygen transfer per horsepower
hour Under operating conditions.
7. PVC sheeting is a suitable basin lining material provided adequate
diligence is expended in assuring sound field welds at all joints. It
is recommended that the bottom be covered with sufficient material to
prevent it from floating even if gas should accumulate beneath it.
140
-------
8. The biological sludge developed seemed to be independent of the
type of fruit processing waste and
a. about 90 percent volatile
b. settled very slowly
c. settling improved at low COD and BOD removal rates
d. mg COD/mg VSS =1.39
e. mg BOD/mg VSS = 0.3 + 0.28 x BOD removal rate
f. mg Organic N:mg Organic P:mgVSS = 0.087:0.016:1
9. Pressurized-recycle flotation thickening of waste activated sludge
from fruit processing waste yields only about 2 percent D.S. thick-
ened sludge. A separate system treating 30 percent or greater domestic
waste along with fruit processing waste yielded a thickened activated
sludge of nearly 4 percent D.S.
10. Net Sludge Growth/day = C x Substrate removed/day -kfc x MLVSS
where
c=0.49 mg VSS formed for Pear processing waste
mg COD removed
c=0.46 mg VSS formed for Peach processing waste
mg COD removed
c=0.57 mg VSS formed for Apple processing waste
mg COD removed
kt=0.115 x 1.14(T~20) mg VSS destroyed
mg MLVSS-day
ft=0.017 x 1.16(T 20) mg COD removed
=Endogenous respiration rate at T °C
11. Substrate removal rate = ft x Substrate Concentration
mg COD removed
mg MLVSS-day-mg/1 COD
(T-201
=0.068 x 1.16V ' mg BOD removed
mg MLVSS-day-mg/1 BOD
= Substrate removal rate coefficient
12. Oxygen required/day = a x substrate removed/day + b x kt x MLVSS
a=0.34 mg Q£ required/mg COD removed
=0.46 mg 02 required/mg BOD removed
b-1.2 mg Q£ required/mg VSS destroyed by endogenous respiration
13. Cost of treatment is estimated to be:
$0.041/#BOD removed for Aerated Lagoon
$0.061/#BOD removed for Activated Sludge
$0.067/#BOD removed for Contact Stabilization
141
-------
ACKNOWLEDGMENTS
The studies reported on herein were supported in part by Research
and Development Project 12060 FAD of the Federal Water Quality
Administration, Department of the Interior. The staff and management
of Snokist Growers and Gray and Osborne, Consulting Engineers are
acknowledged for their valuable contributions.
142
-------
REFERENCES
[1] Standard Methods for the Examination of Water & Wastewater,
12th Edition American Publich Health Association, Inc., New York,
N. Y. 10019.
[2] Jenkins, D., "An Improved Volatile Solids Determination", Water
& Waste Treatment Journal (London) August/September, 1962.
[3] Menar, A. B., Private communication, January 9, 1968.
[4] Esvelt, L. A., Aerobic Treatment of Fruit Processing Waste,
Final Report on WPRD 58-01-68, Department of the Interior, Federal
Water Pollution Control Administration, September, 1969.
[5] Pearson, Erman A., "Kinetics of Biological Treatment", paper
presented at Special Lecture Series on Advances in Water Quality
Improvement, University of Texas, April, 1966.
[6] Eckenfelder, W. W. Jr. and D. J. O'Connor, Biological Waste
Treatment, Pergemmon Press, 1961.
[7] Stewart, Mervin J. , "Activated Sludge Process Variations, The
Complete Spectrum", Water and Sewage Works, Reference No. 1964.
[8] Pipes, Wesley 0., "Ecology of Sphaerotilus in Activated Sludge",
Third Annual Report, Department of Civil Engineering, Northwestern
University, Evanston, Illinois, December, 1967.
[9] Eckenfelder, W. W. Jr., "Theory of Biological Treatment of
Trade Wastes" JWPCF, 39, 2 February, 1967.
[10] Snokist Growers waste treatment operating records, 1969.
[11] Poston, J. C., Private communication, February, 1970.
[12] Selah, Washington sewage treatment plant operating record, 1969.
143
-------
CANNERY. WASTE TREATMENT
BY TWO-STAGE AERATION PROCESS
by
Leale E. Streebin, George W. Reid, and Alan Hu*
The Stilwell Canning Company, located in Stilwell, Oklahoma, appro-
ached the Oklahoma Economic Development Foundation in 1965 to deter-
mine if there were government grants available for waste treatment.
Being consultants to the foundation, the authors were called upon
to assess the problem.
This company employs approximately 450 persons ten months a year,
and it is located in an economically depressed area; therefore, to
qualify for government assistance, it was only necessary to deter-
mine the need. After meeting with the Stilwell city officials and
personnel from the canning company, this was not difficult to est-
ablish. It was readily apparent that if the canning company was
forced into waste treatment without government assistance, it would
be forced to close down. It was also apparent that without some
improvement in its waste handling procedures, it would not be able
to continue to operate.
At that time the waste was divided into two streams. The strong
wastes (wastes from Irish and sweet potatoes) were pumped out
about three-fourths of a mile over a 500 foot hill to a lagoon
and then sprayed on an orchard. This system was not operated pro-
perly and, as a result, several problems existed, not the least of
which were odors which at times were noticeable more than a mile
downwind. To control odors, sodium nitrate was added to the lagoon,
which killed some of the fruit trees. At this point the orchard
owner issued an order to stop irrigation.
The weaker wastes (wastes from green products) were discharged to
the city system which was designed for 6,000 people and was treating
the waste from 2,600 people plus the cannery waste with a population
equivalent to from 7,200 to 15,000 which completely overwhelmed the
treatment plant. The effluent was discharged into Caney Creek, a
spring-fed stream with a flow during the critical summer months
roughly equivalent to the waste flow. After a flow of sixteen miles,
Caney Creek discharges into Lake Tenkiller, a major water recreation
* Respectively, Associate Professor, Director, Research Assistant,
School of Civil Engineering and Environmental Sciences, University
of Oklahoma, Norman, Oklahoma.
145
-------
area. The stream had no trace of D. 0. for a reach of more than five
miles while the cannery was in operation and as a result there were
numerous complaints of odors and in general a bad rapport between the
city residents and the cannery even though the cannery hires 450 people
in a town of only 2,600.
PROBLEM
The problem was to provide an integrated disposal system for the can-
ning plant capable of treating the highest strength, most unbalanced,
and largest volume wastes to a degree compatible with the receiving
stream. This was no small task because organic load per shift varies
by more than 60 fold, from a volume of 0.389 mgpd and COD of 150 mg/1
for spinach wastes to greater than 1.91 mgpd and a strength 5,500 mg/1
while processing sweet potatoes. To solve the problem, a research
study undertaken which included an initial survey; a detailed analysis
of waste orgin, strength and volume; literature review; review of survey
data taken by the Oklahoma State Health Department and others; a bench
scale model study to verify loadings and nutrient requirements; and
finally, recommendations for process selection and scope design.
Based on the information gathered and reported elsewhere, (1) the prob-
lem was to develop a process that would most economically reduce an
unbalanced waste (C/N/P = 1200/11.6/1) ranging from 0.4 mgd to 2 mgd
and from 150 to 5,500 mg/1 COD to a level compatible with the receiving
stream. Several systems were considered including trickling filters,
minimal solids, moderate solids, and extended solids aeration and bio-
sorption systems.
The incoming waste in question has a very high strength; therefore,
trickling filters and biosorption systems were deleted frort consid-
eration, because they were not capable of meeting the required standards
without polishing ponds or other tertiary treatment systems. Moderate
solids or extended aeration systems, though efficient in terms of per-
centage removals, would suffer from size diseconomies; therefore, a
dual or two stage aeration process was chosen.
The first is a high rate process with high loading capability, followed
by a more efficiently polished-effluent-producing process, i.e., min-
imal solids followed by extended aeration. Minimal solids processes
have a short sludge retention time (SRT); a high AM/AF ratio (AM =
change in active solids and AF = change in organic concentration),
and a high loading rate (greater than 2.0 Ibs COD/Day/lb VSS).
With this process the removal rates are extremely high, the
removal efficiencies low, and the mixed liquor suspended solids
146
-------
highly dispersed; therefore, the process, though efficient in terms
of dollars per pound of COD removal, must be followed with another
process.
The process chosen, an extended aeration process, has a loading rate
of about l/100th the rate of a minimal solids system, a AM/AF ratio
much lower, and SRT perhaps 100 times greater. The two stage aera-
tion system combines the desirable characteristics of both. The high
rates of removal are provided by the minimal solids basin and polished
effluent and aerobic sludge digestion by the extended solids aeration
basin.
RECOMMENDED PROCESS
The two stage aeration system designed (shown in Figure 1) consists
of screens to remove gross solids which are trucked to a previously
existing landfill. After screening the waste flows to a minimal
solids basin, which was designed to remove 50 percent of the soluble
BOD, and then to two low rate extended aeration basins, designed as
polishing units. Aeration is followed by final clarification. Pro-
visions are built into the system to return the sludge to the front
of all aeration tanks and to maintain a recirculation ratio of approx-
imately one to one. Occasionally it is necessary to withdraw excess
sludge from the system which is routed to sludge retention basin for
anaerobic digestion.
This system is so designed that it can accomplish a high degree of
treatment at the design capacity without a primary clarifier. With
the addition of a primary clarifier, which will remove approximately
50 percent of the BOD, and a retention cell for solids disposal, the
capacity of the system will be increased by a factor of approximately
two.
This plant is so designed that any one of the treatment units may
be operated individually or in conjunction with other units. This
modular design makes the plant ideal for optimal operation and re-
search. It can be operated as a minimal solids, moderate solids,
or extended solids aeration process or a combination of any of these
processes. Other parameters that can be controlled are solids reten-
tion time, hydraulic retention time, and power input to the surface
aerators.
The recommended design criteria is given in Table 1.
DEMONSTRATION GRANT
Benefiting from the grant-in-aid program of the Federal Water Quality
Administration of the U. S. Department of Interior and a grant from
the Economic Development Administration Office of the U. S. Department
of Commerce, the plant was constructed and started operation in May, 1969,
147
-------
Raw Waste
To Truck Loading
Hopper
Screen
Sludge Vy
Beds jT
Return And Excess Sludge
Parshall F!
' 1 >
JX
Primary7 N\ Sludge
Clarifier ' /
^"' '
L
1
Minimal Solids 2 Surface '
Aeration Aerators @75 HP
1
1
Lume
1
1
1
1
u" 1— 1
"T T^
Vacuum Filter
Nutrients
By-Pass
1 1
3 Surface 3 Surface
Aerators Aerators
@ 40 HP @ 40 HP
t
^^ f \ Effluent To Caney Creek
Sludge Punp
Final Clarifier
Present Facility
Future Facility
Figure 1. FLOW DIAGRAM
148
-------
Table 1. Design Criteria
1. Design Flow =1.5 MGD
BOD = 1,500 mg/1
2. Minimal Solids Unit
Vol = 40,584 ft = 0.30 MG
Depth = 8 ft
Top = 116 ft x 66 ft
Bottom = 84 ft x 34 ft
HRT = ^4 =4.87 hrs
-L • j
3 Organic loading, COD = 20,000 Ibs COD/day
4. HP provided = 2 surface
aerators @ 75 HP
total 150 HP
5. Extended Aeration
_ 2.29 MG _
HRT ~ 1.5 MGD ~ 36 hrs
Vol = 2.29 MG
HP provided = 240 HP
6. Final Clarifier
Overflow rate based on
1.5 MGD = 800 gpd/ft
Depth = 8 ft
Diam. = 50 ft
DT 1.5 MGD = 1.89 hrs
Weir loading 1.5 MGD = 10,000 gpf/day
7. Return Sludge Pumps
2 sets @ 500 gpm
with TDK = 48 ft
No sludge flow meter provided
8. Chemical Feeder
1 set day feeder
@ 1,000 Ibs/day
149
-------
A demonstration grant was secured from FWQA to study the system
and to show that a high strength, large volume, nutritionally
unbalanced, cannery waste can be successfully treated to a high
degree by the two stage aeration process. The results will be
used in establishing design criteria for the treatment of fruit
and vegetable waste treatment using a two stage aeration process.
The scope of this research includes:
1. Collection and analyses of the various vegetable and fruit
wastes as they are produced and discharged from the cannery.
2. Determination of waste flows for each product processed.
3. Studies on the waste characteristics and the removal efficien-
cies for each modular process and the entire system.
4. Evaluation of receiving water response to discharged wastes.
The Stilwell Canning Company cans and freezes a wide variety of
vegetables and fruits which include spinach, strawberries, greens,
beans, yellow squash, okra, peas, white potatoes, and sweet pota-
toes with potatoes being the dominant product. A production sch-
edule is shown in Table 2.
Within this variety of products being processed during the canning
season, it is evident that the organic strength and flow fluctuates
widely. Seasonal changes and moreover changes between and within
shifts are reflected in the waste stream makeup. In order to cope
with changing production, the activities within the cannery were
followed closely and recorded at the time of sampling.
SAMPLING
Because the primary purpose of this project was to study treatment
efficiency and cannery waste characteristics, the sampling program
had to be developed to monitor the changes in waste flow and strength
and to establish the effects of these changes on treatment efficien-
cies.
In any sampling program, considerable thought must be given to the
location of sampling points. Sampling points should conform to
hydraulic suitability, that is, points of high turbulence that
assure good mixing should be selected. The points must also be
located so that individual process unit efficiencies could be deter-
mined. Examination of the plant layout indicated that sampling from
150
-------
Processing
Time
Apr - Dec
Apr - Dec
Apr - Dec
Apr - Dec
May - Nov
Table 2. Production Record
Water
Production Used Peak
Cases
Year
1965 1,494,266
1966 1,650,000
1967 1,750,000
1968 1,800,000
1969 1,979,000
MGD
1.5
0.84
Month
Oct
Sept
Peak
Production
Cases
234,311
268,000
Sept
Oct
294,741
271,294
Case = 24 @ 303 can
1969 Production Schedule
Month Products
May Spinach, Mustard Green, Collard Green, Turnip Green,
Strawberry, Irish Potato
June Green Beans, Irish Potato, Squash, Peas, Blackberry
July Green Beans, Okra, Squash, Irish Potato, Peas
August Irish Potato, Okra, Peas, Sweet Potato
September Sweet Potato, Irish Potato, Okra, Peas, Butter Beans,
Squash, Green Beans
October Sweet Potato, Irish Potato, Okra, Squash, Butter Beans,
Lima Beans, Peas, Turnip Green, Mustard Green
November Sweet Potato, Collard Green, Turnip Green, Spinach
151
-------
the Parshall Flume, the minimal solids aeration unit, the extended
solids aeration unit, and after the final clarifier would yield homo-
geneous samples that could be used to determine waste characteristics
and plant efficiencies.
Sampling and flow determinations were also considered within the can-
nery but were not run because the various waste streams within a
process were integrated with streams from other simultaneously oper-
ating processes. Referring to the flow diagrams in Figures 2 through
10, notice that there is a minimum of four waste streams (excluding
blackberry) per product processed. Therefore, a complete analysis
and flow determination of waste streams from each product processed
was not practical because it would have necessitated the complete
analysis of a minimum of twelve streams plus the analysis of the
waste treatment system. A complete analysis of the treatment system
included the following:
1. Raw waste
2. After screening
3. Minimal Solids
4. Extended Aeration
and Final Clarifier
5. Sludge
Total and soluble COD
Total, Settleable, Suspended and Volatile
Solids
Total and soluble COD
Total, Settleable, Suspended and Volatile
Solids
Kjeldahl, Nitrite, Nitrate, and Ammonia
Nitrogen
Total Phosphate
Alkalinity, pH, temp.
Chlorides
Total and soluble COD
MLSS and MLVSS
Dissolved Oxygen
Kjeldahl, Nitrite, Nitrate, and Ammonia
Nitrogen
pH, temp
Total and soluble COD in effluent
MLSS and MLVSS
Dissolved Oxygen
Kjeldahl, Nitrite, Nitrate, and Ammonia
Nitrogen in tank and in effluent
Settling Study
Total phosphate in tank and in effluent
Effluent Suspended and Volatile solids
pH, temp.
% Volatile Material and Sludge Concentration
152
-------
IRISH POTATOES
I
Dry Screen
4-
Wet Screen —
Blanching
Wet Screen
i
Steam Peeling
T
Wet Screen and Spraying
Polisher (Abrasive Peeling)
4-
Spraying
4-
Sorting and Rinse
FREEZING PROCESS
->• Waste Flow*
-»• Waste Flow
Waste Flow*
-> Waste Flow*
-*• Waste Flow*
-»• Waste Flow
-*• Waste Flow*
-* Waste Flow*
CANNING PROCESS
Packing -*-
Freezing and
Storing
* Major Stream
Canning
->• Waste Flow
Cooking
Y
Cooling
Y
Dry Storage
Figure 2. FLOW SHEET OF IRISH POTATO PROCESSING
153
-------
SWEET POTATOES
I
Dry Screen
r
Wet Screen
I
Blanching
T
Wet Screen
i
Steam Peeling
Wet Screen and Spraying-
-* Waste Flow*
-*• Waste Flow
-»• Waste Flow*
-*• Waste Flow*
-*• Waste Flow*
Polisher (Abrasive Peeling)
Spraying
Sorting and Rinse
I
Sugar and/or Other Additives
-*• Waste Flow
-> Waste Flow*
-> Waste Flow*
FREEZING PROCESS
CANNING PROCESS
Packing
Freezing and
Storing
*Major Stream
Canning
Waste Flow
Cooking
Cooling
Dry Storage
Figure 3. FLOW SHEET OF SWEET POTATO PROCESSING
154
-------
*Major Stream
OKRA
Stem Remover
Blanching
Water Bath
Packing
Freezing and Storing
Waste Flow
•> Waste Flow*
-> Waste Flow*
Waste Flow
Figure 4. FLOW SHEET OF OKRA PROCESSING
155
-------
GREEN VEGETABLE
J
Dry Screen
T
Pick Belt
Spraying
-> Waste Flow*
Water Bath
Waste Flow
CANNING PROCESS
Cutting
Canning —
Cooking
Cooling
Dry Storage
Blanching
-»Waste Flow*
FREEZING PROCESS
Cooling
4-
Cutting and Chopping
« Packing
Waste Flow
Waste Flow
*Major Stream
Figure 5. FLOW SHEET OF GREEN VEGETABLE PROCESSING
156
-------
GREEN BEANS
Dry Pick Belt
i
Dry Screen
Vine Remover
Water Bath and Spraying
Snipper
•r
Grader
I
Cutter
Waste Flow*
Grader
i
Spraying and Pick Belt T-
r
Hopper
Blanching
1
Canning
Cooking
Cooling
->• Waste Flow
-»• Waste Flow*
-> Waste Flow
Dry Storage
*Major Stream
Figure 6. FLOW SHEET OF GREEN BEAN PROCESSING
157
-------
PEAS and BEANS
Husk Remover
Air Cleaner
Flotation and Spraying
r
Froth Flotation Cleaner
I
Blanching
-*• Waste Flow*
Waste Flow
-»• Waste Flow*
Holding Tank
-*• Waste Flow*
Blanching
Air Cleaner
Packing
Freezing and Storing
-*• Waste Flow
-*• Waste Flow
*Major Stream
Figure 7. FLOW SHEET OF PEAS AND BEANS PROCESSING
158
-------
* Major Stream
SQUASH
Pick Belt
Lye Wash
Rinse
Cutter
i
Spraying
v
Blanching
Cooling —
i
Dewatering
Packing
4-
Freezing and Storing
-> Waste Flow*
Waste Flow*
Waste Flow*
Waste Flow*
Waste Flow
-> Waste Flow
Waste Flow
Figure 8. FLOW SHEET OF SQUASH PROCESSING
159
-------
BLACKBERRIES
*Major Stream
Spraying
Pick Belt
Canning
Cooking
Cooling
Dry Storage
-> Waste Flow*
-»• Waste Flow
Figure 9. FLOW SHEET OF BLACKBERRY PROCESSING
160
-------
STRAWBERRIES
Vibrator and Pressure Wash >• Waste Flow*
Pick Belt
Spraying
Packing
Freezing and Storing
*Major Stream
-> Waste Flow*
Slicing and Sugar Addition >• Waste Flow
-> Waste Flow
Figure 10. FLOW SHEET OF STRAWBERRY PROCESSING
161
-------
Equal in importance to collection of a good sample is the preserva-
tion of this sample between the time of sampling and analysis. Waste
water samples are unstable due to their chemical, physical, and bio-
logical characteristics. Since all analyses were conducted in a
mobile laboratory at the treatment plant, refrigeration was used
for preservation.
FLOW DETERMINATION
Estimates of flow volumes in cannery operation are needed for design
and planning purposes. It was desired to use the data taken in the
study of the cannery to develop and verify a prediction scheme which
then might be used to predict flows at the Stilwell site or used for
prediction for design purposes at other locations. The best means
was believed to be estimating the flow per case for each vegetable
type in the cannery by use of a predictive equation.
The data available for this study was a continuous flow record at the
entry to the waste treatment plant and production figures for each shift
(night and day) by vegetable type in cases of #303 cans.
The flow measured at the treatment plant was a combination not only
of all waste streams but of infiltration' into approximately 1.5 miles
of line between the cannery, the treatment plant, and domestic sew-
age for about eighteen families.
The periods of production usually included several vegetable types
during a shift. Not enough periods of processing of single vege-
table types could be obtained during the six months of study to ex-
amine even the most commonly occurring vegetable types separately.
Therefore, the indicated approach was decomposition of the flow into
portions associated with each of the vegetables. The approach chosen
was the linear model listed below with multiple regression estimates
of the parameters:
F = b + al + b1?l + b2P2 + b P
F is flow volume in gallons during the period for which production
is reported; b is a constant flow associated with the plant being
in operation for the period and represents internal waste water,
cooling water, cleanup water, etc., which cannot be associated
with individual processes, etc; a is a coefficient; I is the infil-
tration and domestic sewage input below the cannery; P. is the pro-
duction in cases for the ith vegetable type, and b. is the fpc for
the ith vegetable type. I was estimated by base flow during non-
production periods.
162
-------
Initial analysis was done on the basis of separate shifts. The
shift values were generally reasonable although the estimated fpc
for squash was negative for both shifts and for okra in the night
shift. The break between shifts was based on information provided
by the company which was used for payroll data.
Observing the processes indicated that the time of division between
shifts was not always reliable, because the actual processing might
proceed or follow the nominal shift by several hours and cause sig-
nificant error in the flow breakdown between shifts. To avoid this
source of error, the shifts were combined and each day run as an in-
dividual data value. The estimates of fpc were all positive, the
estimate of the constant b was smaller for the work day than for
either shift, and the multiple correlation coefficient was higher
than that for either of the individual shifts. The results of this
run are shown in Table 3.
The coefficient "a" would normally be 1.0 but was allowed to be
fitted because it was desired to see if the estimates of infiltra-
tion by using periods of no production were reliable. Values for
the coefficient much less than or greater than 1.0 would indicate
unreliability. In the studies made to date," the coefficient has
ranged from 1.0 to 1.6 and in future studies will be assumed 1.0
and not estimated.
The error analysis for this run is shown in Table 4. The sources
of error in estimating this relationship are:
1. Error in flow measurement
2. Error in reporting production
3. Variability in fpc
4. Variability in nonprocess water (constant term)
The first source is normally about 5%-10% based on the listed per-
formance for the flow measuring device. However, several days during
the study period, the capacity of the flow measuring device was exce-
eded and the flow measured manually. What the error was in these
cases is unknown.
The second source is uncertain in quantity. The production figures
should be exact; however, it is possible some processing was partly
completed and the vegetables stored and finished and reported the
next day.
The third and fourth sources of error are errors which are due to
estimating averages for values which actually show a distribution.
There is certain to be some variation in fpc because of the dif-
ferences in raw product quality for a vegetable type, possibly sea-
sonal variation, variation due to operating personnel and that due
163
-------
Table 3. Estimates of Parameters
t Value*
Constant b
Infiltration coefficient a 1.30 4.71
Vegetable FPC (gals)
Irish Potatoes 153 12.17
Okra 6 51.00
Peas 26 2.81
Sweet Potatoes 147 13.15
Beans 72 3.06
Squash 5 .39
Green Beans 41 8.17
Mustard Greens 109 5.97
Turnip Greens 69 7.54
Collard Greens 35 1.71
Spinach 61 7.98
*t = 1.29 @ 10%, t = 1.66 @ 5%, t = 2.36 @ 1% levels of
significance
Table 4. Error Analysis
Multiple Correlation Coefficient .906
Standard Error of Estimate 148,600 gals
Average Flow per Workday 696,900 gals
Prediction Error for 147 days
(error is residual/observed value)
percent less than 10% 55.1
percent less than 20% 84.4
percent less than 50% 95.2
average percent error 14.8
164
-------
to a change in quantity of product processed per unit of time. The
amount of water flowing during a shift which is not process water
will vary also.
This model may be improved. Variation in flow may be estimated by
taking periods of seven, fourteen, or more consecutive days and esti-
mating the parameters for the different periods and seeing what change
there is from one period to the next. These results will be included
in the final report.
The estimates of fpc for the eleven vegetable types are reasonable in
magnitude and are in general agreement with published figures. Since
the constant term is only about 10% of the average flow, the estimates
of fpc account for most of the water used in processing. Thus, the
values of fpc are fair estimates of the average amount of water assoc-
iated with each case of a particular vegetable processed in this cannery,
The model can be used to predict flows at this site or used at another
site by estimating the number of cases processed. The constant term
can be estimated as a percentage of the average predicted flow based
on assumed production and estimated fpc. The infiltration term would
be estimated if there are flow sources outside the plant which need to
be added (as for the cannery at Stilwell).
DETERMINATION OF STRENGTH
To better predict and design waste treatment processes, it is desired
to predict quantities of pollution based on the production figures as
was done in the volume studies. The technique will be used to predict
mass rate of flow of the pollutant under consideration during the time
interval set for the study (i.e., shift or day). The mass of pollutant
produced during the time interval (the mass rate of flow times the inter-
val) will be related to production during the same interval using a model
similar to the flow volume model. The estimated flow and mass of pollu-
tant will then be used to determine loading rates for the treatment system.
Flow and production data available are the same as listed above under
flow determination. Additional data are chemical analyses performed
one to four times daily on samples taken at the flow measuring point.
The prediction of mass rather than concentration is based on the belief
that the variation in mass rate of flow of pollutant is less variable
than concentration since a constant mass flow would show variation due
to dilution.
RESULTS
The two stage aeration system designed to treat 1.5 million gallons at
a strength (COD) of 2250 mg/1 (BOD- = 1,500 mg/1) performed above expec-
tations. The system was designed tor what was expected to be ultimate
capacity; however, due to unforseen circumstances by the time the treat-
ment plant was constructed, it was already operating above design capacity.
165
-------
During the six-month study, the flow reached one instantaneous peak
of 2.4 mgd and a peak daily flow of 1.9 mgd. The waste strength while
processing sweet potatoes peaked 5,500 mg/1. Even at these higher
loading rates the soluble COD removal efficiencies fpr the system
exceeded 95 percent. The removal efficiencies based on settled sam-
ples were also above 95 percent except during the period of sludge
bulking when it dropped to as low as 66 percent; Sludge bulkinp was due
to nutrient deficiencies. Nitrogen was not added to the system while
processing the unbalanced sweet potato waste until bulking occurred.
The waste strength and flow varied tremendously as can be seen in
Figures 11 through 13. With this radical variation the system had
to be capable of receiving shock loads without being adversely affec-
ted. Figures 11 and 12 and Table 5 demonstrate the stability of the
two stage aeration system. The soluble COD in the plant effluent,
except for one period during October when the system was nutrient
deficient, was always less than 140 mg/1. The minimal solids system
removed from 66 percent to 99 percent of the soluble COD while the
loading rate varied from slightly less than 1 Ib COD/1b MLVSS to
greater than 20 as shown in Figure 14. This is somewhat misleading
because a high percentage (up to 60 percent) was converted to cell-
ular material which resulted in a load on the extended aeration system.
The analyses are not complete on the extended aeration units; how-
ever, the soluble COD removal efficiencies exceeded 94 percent except
during periods of upset due to nutrient deficiencies. The volatile
solids removal efficiencies was generally greater than 90 percent.
Further results will be reported in the final report.
The waste treatment system protected the receiving stream. The
stream, with a flow during the summer months approximately equi-
valent to the waste flow, maintained a D.O. concentration greater
than 4.0 mg/1 for its entire length during the entire canning sea-
son. However, during the period of sludge bulking the stream did
suffer some damage due to sludge deposits.
CONCLUSIONS
1. The two stage aeration system is very stable and capable of
accepting shock loads without being adversely affected.
2. The system is flexible. The various units can be operated
separately to match the flow and strength variation.
3. The minimal solids can be started up readily by recycling the
mixed liquor from the extended aeration system.
166
-------
6 -,
M
O
O
O
Q
O
CJ
5 -
3 -
2 _
1 _
Minimal Solids
Soluble Effl
Plant Soluble
Effl
\
June
31
July Aug. Sept. Oct.
Figure 11. TREATMENT PLANT OPERATION DATA
30
Nov.
-------
oo
5 .
A -
3 -
00
6
o
o
o
Q
O
u
2 -
1 -
Minimal Solids
Effl
June
July
Figure 12. TREATMENT PLANT OPERATION DATA
-------
2 -,
Ch
60
U)
tfl
1 _
30
31
31
30
31
June
July
Aug.
Sept,
Oct.
Nov.
T~
30
Figure 13. FLOW VARIATION
-------
Table 5. Plant Performance Data
Min.
Max.
Min.
Max.
Min.
Max.
A. Products Processed:
otato, okra, peas, squash, beans, turnip
and one or more of the other products.)
green, and cobbler. (The daily process included
sweet
1. Plant Influent Characteristics
Solids
SS DS VSS VDS
1150 1300 860 1110
2540 3120 2400 2850
2. Performance
Ibs COD
MLSS MLVSS Ibs MLVSS/day
2500 2000 4.00
4200 3600 9.26
COD
Total Settled Soluble pH T°C
3060 1700 1190 5.3 23
5500 3380 3240 7.9 33
in Minimal Solids Unit
Albs COD
Ibs MLVSS/day COD Removal, %
settled soluble settled soluble
4.00 4.05 74 76
7.51 7.75 99 99
TKN
14
59
SVI
150
270
3. Performance of Extended Aeration Unit, including Final Clarifier
Solids
COD Removal %
SS VSS Total Soluble
8 5 66 94
180 83 99 99
2915* 2390*
Effluent COD Removal %
Total Soluble SS VSS SVI pH
19 Trace 87 94 140 6.
186 128 99 99 300 7.
3190* 268
1
8
6
*Bulking Sludge, solids unloaded
-------
Table 5. (Cont'd.) B. Products Processed
Irish potato, okra, peas, beans, green beans
1. Plant Influent Characteristics
Solids COD
SS DS
Mln. 840 876
Max. 2036 1728
MLSS MLVSS
VSS VDS
Total Settled
Soluble pH T°C
810 667 2016 1000 911 4.6 29
1924 1360 4229 1550 1390 7.2 37
2. Performance in Minimal Solids Unit
Albs COD
Ibs COD
Ibs MLVSS /day
Min. 840 700 2.56
M
M Max. 4270 3521 10.96
3. Performance of Extended
COD Removal %
SS VSS
Min . 8 2
Max. 212 210
Total Soluble
86 95.6
99 99.0
Ibs MLVSS/day
Settled Soluble
COD Removal %
Settled Soluble
2.42 2.44 63 77
8.65 9.98 98 99
Aeration Unit, including Final Clarifier
Solids
Effluent COD Removal %
Total Soluble
30 10
400 120
SS VSS SVI
84.5 85 80
99.0 99 240
SVI
50
256
TKN
77
pH
6.80
7.65
-------
Min.
Max.
H
VJ
Min.
Max.
Min.
Max.
Table 5.(Cont'd.) C. Products Processed
Green Beans, Squash, and Green Vegetables
1. Plant Influent Characteristics
ss
18
378
Solids
DS VSS VDS
221 15 116
664 220 415
COD
Total Settled Soluble pH
112 82 82 6.1
495 495 445 10.2
T°C
22
31
TKN
14
24
2. Performance In Minimal Solids Unit
MLSS
123
1690
3.
Solids
SS VSS
Ibs COD
MLVSS Ibs MLVSS/day
63 2.39
880 14.16
Performance of Extended
COD Removal %
Total Soluble
3 Trace 56 64
60 50
99 99
Albs COD
Ibs MLVSS/day COD Removal %
Settled Soluble Settled
1,05 2.39 33
12.23 13.84 86
Soluble
66
98
SVI
43
200
Aeration Unit, including Final Clarifier
Solids
Effluent COD Removal %
Total Soluble SS VSS
5 Trace 69 74
167 134 99 99
SVI
46
170
£H_
7.1
8.0
-------
LO
60
o
01
•H
t>0
n)
Q
§
CJ
20 -
10 .
Minimal Solids:
Ibs COD
Ibs MLVSS
Ibs COD
•/ day
June
July Aug. Sept.
Figure 14. MINIMAL SOLIDS' LOADING RATE
Oct.
Nov.
-------
4. The loss of sludge during the processing of sweet potatoes was
a result of a nitrogen deficiency.
5. The size of the clarifier should be increased and the overflow
rates decreased from those recommended for domestic waste treatment
systems.
6. The receiving stream maintained a D. 0. of above 4 mg/1 for the
entire stream reach for the entire canning season.
7. The two stage aeration system provides high rates of removal and
high treatment efficiencies.
ACKNOWLEDGMENTS
This investigation was supported by Demonstration Grant No. 12060DSB
from the Federal Water Quality Administration, for which,
we express our sincere appreciation.
The cooperation of the personnel at the Stilwell Department of Util-
ities, and that of Stilwell Canning Company is gratefully acknowledged.
Thanks are extended to Mr. John Palafox who helped in performing
numerous laboratory analyses.
174
-------
REFERENCES
1. Reid, G. W., Streebin, L. E., Klehr, E. H. and Love, 0. T. ,
"Water Treatment Facilities for A Large Canning Company," Center
for Economic Development - State of Oklahoma, Publication No.l,
November, 1966.
175
-------
LIME TREATMENT AND IN-PLANT REUSE
OF AN ACTIVATED SLUDGE PLANT EFFLUENT
IN-THE CITRUS PROCESSING INDUSTRY
by
Richard H. Jones*
The citrus industry of Florida is a $2 billion business which produces
75 percent of the oranges and 80 percent of the grapefruit grown in the
United States. Some 50 percent of the grapefruit and 80 percent of the
oranges are processed for sale as products other than whole fresh fruit,
for example, as frozen concentrates. A recent survey found there were
52 citrus processing plants in the state which discharge 150 mgd of
wastewater having an organic loading of 400,000 pounds BOD per day.
This is equivalent to the wastewater from 2 million people.
The Winter Garden Citrus Products Co-operative was cited by the Depart-
ment of Air and Water Pollution Control for discharging their wastewater
into Lake Apopka untreated for over 20 years. They were given six weeks
in which to find an acceptable method for treating their waste or they
would not be allowed to continue operation. Environmental Engineering,
Inc., was retained as their consulting engineers and was presented with
the task of developing, in a six-week period, an acceptable waste treat-
ment method including control of effluent nutrients. Environmental
Engineering, Inc., was also presented the task of obtaining a research,
development, and demonstration grant from the FWQA.
DISCUSSION
In the past, the only partially successful method of citrus waste treat-
ment was by land irrigation. This method required large areas of
specially suited land, which was seldom available. The Winter Garden
Citrus Products processing plant, located in the city of Winter Garden,
was therefore limited in the amount of land available for waste treatment.
In Leesburg, Florida, a waste treatment plant had been in operation for
two years, combining the city sewage with the concentrated waste from
the Minute Maid citrus processing plant for treatment. It was not
feasible to combine domestic sewage with the waste from the processing
plant in Winter Garden because the city wanted nothing to do with the
citrus waste in their existing sewage treatment plant.
A waste treatment plant was under construction in Auburndale, Florida,
designed to treat the total flow from two citrus processing plants. This
*Environmental Engineering, Inc., 2324 S.W. 34th Street, Gainesville,
Florida 32601
177
-------
meant treating not only the concentrated wastewater but also the baro-
metric leg and cooling water. If this method had been attempted at
Winter Garden, it would have meant treating some 10 mgd of waste with
a BOD concentration of 400 mg/1 instead of treating just the concentrated
waste of 2 mgd with a BOD of 2000 mg/1. Again, there was not enough
land available; therefore, this type of treatment process could not be
used.
Neither the citrus waste treatment plant in Leesburg nor Auburndale has
operated successfully to date. The main problem with the plant at
Leesburg is overloading and lack of facilities to handle waste activated
sludge. At the Auburndale plant, the secondary clarifiers have never
functioned.
The decision was made to test on a laboratory basis the treatment of con-
centrated citrus waste alone by the complete mixed activated sludge
process. A continuous flow, bench-scale unit was operated for approxi-
mately four weeks. Results from this research showed that concentrated
citrus waste could be effectively treated in this manner. Further
laboratory tests showed that lime treatment of the biologically treated
effluent would give a final effluent low in suspended solids and phos-
phorus .
Based upon this short period of testing, a full-scale plant was designed.
Figure 1 shows a schematic of the Winter Garden waste treatment plant as
initially constructed. Plant design was based on a flow of 2 mgd and a
BOD of 2000 mg/1. Nutrients were added in the form of phosphoric acid
and anhydrous ammonia. The pH was controlled with NaOH as required.
Concentrated citrus wastewater was pumped into two concrete aeration
basins 144 feet in diameter. Side water depth was approximately 12 feet
and aeration was provided by three 75 hp floating aerators. The aera-
tion basins were designed so that operation in series or parallel was
possible.
Effluent from the aeration basins flowed to a clarifier from which
return sludge was pumped back into the influent pipe to the aeration
basins. Effluent from the clarifier flowed to the lime treatment tank,
then either back to the citrus plant for reuse or directly to Lake Apopka.
Waste sludge from the clarifier and lime treatment tank were discharged
to a wet well, then pumped to a surge tank back in the citrus plant.
The surge tank also served as a sludge thickening tank. Thickened
sludge from the bottom of the surge tank was pumped through a solid
bowl centrifuge. The supernatant was returned to the plant influent,
and the solids cake was mixed with pressed citrus peel for recovery as
cattle feed.
The FWQA awarded a research and development grant to Winter Garden Citrus
covering only the lime treatment unit, as a previous grant had been
awarded on activated sludge treatment of citrus waste at Leesburg and
Auburndale, Florida. However, Winter Garden Citrus Products Co-op, was
requested to provide complete chemical analyses on the activated sludge
portion of the treatment process and include the data in all reports.
The grant plan had several objectives:
178
-------
pH Adjustment
Nutrients
Influent.
Aeration
Tank
Return Sludge
Supernatant
Inplant Reuse
Effluent
Overflow
Sludge To Dewatering
Process
Sludge Thickening Tank
Centrifuge Dewatering
Process
Thickened Sludge To
Feed Mill
Figure 1. WASTEWATER TREATMENT SYSTEM
-------
a) to determine the design and operating parameters for lime treat-
ment of effluent from an activated sludge plant treating waste-
water from a citrus processing plant;
b) to reuse the treated effluent in the processing plant;
c) to determine the efficiency of lime treatment for removal of
residual organic matter including nutrients;
d) to determine the suitability of waste sludge as a cattle feed
additive.
WASTEWATER FLOW AND FRUIT PROCESSED
During the 1968-1969 citrus processing season, a daily record was kept
of tve number of boxes of fruit processed and the volume of wastewater
treated. A record was also kept of the type fruit processed. Many
factors influence the amount of fruit processed in any particular day.
A processing plant will operate one, two, or three shifts a day and up
to seven days a week,depending on the availability of fruit. Wastewater
is extremely variable throughout a processing season, and frequently on
weekends there is no wastewater at all.
Data collected showed that for every box of fruit processed (90 pounds
of fruit) approximately 25 gallons of concentrated wastewater were pro-
duced. This wastewater had an average BOD of 1170 mg/1; therefore,
approximately 0.25 pounds of BOD for each box of fruit processed required
treating.
NUTRIENT ADDITION AND CONTROL
The original waste treatment plant design called for nutrient addition
in the form of anhydrous ammonia and phosphoric acid. Control of pH
was to be by NaOH addition. Ammonia and phosphoric acid were added
manually. It soon became evident that there was no satisfactory method
to control nutrient additions due to the extreme fluctuations in hydrau-
lic and organic loading rates. After several weeks of operation it was
concluded that the most effective method of controlling nutrients would
be to measure the nutrient concentration in the effluent of the activated
sludge process.
Test kits were obtained which enabled the operators to determine concen-
trations of orthophosphate and ammonia every two hours as a means of
controlling nutrient addition. These tests were made a part of the
routine operation and an attempt was made to keep the ammonia and ortho-
phosphate concentrations in the activated sludge plant below 0.5 and
1.0 mg/1, respectively. As long as the aeration time in an activated
sludge process is not sufficient to convert ammonia to nitrate, then in
this manner the effluent nitrogen and phosphorus can be effectively
controlled.
180
-------
ACTIVATED SLUDGE PROCESS
As mentioned previously, the activated sludge portion of the waste treat-
ment plant was designed to treat 2.0 mgd with a BOD of 2000 mg/1. Aera-
tion was to be provided by four 75 hp mechanical surface aerators in
each of the two aeration tanks. Only three 75 hp aerators were provided
for each tank because of the expected reduced load. Hydraulically and
organically, the plant was loaded at approximately 60 percent capacity.
The maximum and minimum flows to the plant were 1.5 mgd and 0.0 mgd,
respectively. The maximum and minimum BOD levels applied were 3300 mg/1
and 270 mg/1, respectively.
Operational problems were caused by the release of peel press liquor and
water from the orange oil recovery system. Normally, all peel press
liquor is evaporated down to molasses, then recovered as cattle feed.
However, the evaporation capacity of the plant was stressed to a maximum,
and it was not unusual to receive 25,000 to 50,000 gallons of press
liquor over a short period of time. Peel press liquor has a BOD of
around 60,000 mg/1 and contains orange oil and d-Limonene, which are toxic
to the activated sludge process and also cause foaming. In general,
when a slug of peel press liquor was received, the aeration tank D.O. was
lowered drastically, the S.V.I, decreased, and foaming occurred. This
resulted in solids overflowing the weirs into the lime treatment tank
and significantly reducing the overall plant treatment efficiency.
Foaming was perhaps the major problem experienced with the activated
sludge process, occurring for a considerable portion of the citrus season
and causing a reduction in treatment efficiency for both the activated
sludge process as well as the lime treatment process. It was only during
the latter part of the season that the major cause of foaming was deter-
mined. Water from the orange oil recovery process was being discharged
into the treatment plant and this water contained a high concentration
of emulsified orange oil. As soon as this waste stream was diverted to
the evaporators and no excess press liquor was discharged, the activated
sludge process performed exceptionally, as did the lime treatment process.
ACTIVATED SLUDGE PROCESS LOADING RATES
Figure 2 shows the variation in process loading rates during the citrus
season. Loading rates varied from a minimum of 0.04 pounds BOD/pound
MLSS to a maximum of 0.378 pounds BOD/pound MLSS. The average during
the citrus season was 0.143 pounds BOD/pound MLSS. This loading rate
was within the limits of the loading rates of the extended aeration
process.
ACTIVATED SLUDGE PROCESS TREATMENT EFFICIENCY
Figures 3 and 4 show the influent BOD and the effluent BOD. As previously
mentioned, the maximum and minimum BOD concentrations were 3300 and 270
mg/1, respectively, in the influent. Maximum and minimum BOD's in the
effluent were 129 and 2 mg/1; however, the maximum values of 129 mg/1 is
probably incorrect as only two 24-hour composites of the effluent had
BOD concentrations greater than 100 mg/1.
181
-------
00
0.32-
0.28-
0.24-
to
* 020-
rH
£3
O
ca
w
£ 0.16-
00
c
•a
cfl
eS 0.12-
0.08
0.04-
0.00
Janua
• -
-
ry
-
-
-
-
-
-
-
February
-
-
-
-
-
March
-
April
-
-
May
-
-
-
1
June
Figure 2. ACTIVATED SLUDGE LOADING RATES
-------
oo
CO
3200 -
2800 •
2400 -
2000 -
^H
00
ry
g 1600 -
1200 -
800 -
400 -
0
Janua
ry
-
-
•
-
-
-
-
-
February
-
-
-
March
-
April
-
May
June July
Figure 3. INFLUENT BOD
-------
140 -
00
120
c
o
pa
100 -
80
60 -
40 --
20 -
0
January
February
March
April
May
I_L
I I i I I
June
July
Figure 4. BIOLOGICALLY TREATMENT EFFLUENT BOD
-------
The average BOD of the influent was 1170 mg/1 for the complete citrus
processing season, while the effluent BOD averaged 30 mg/1 during the
same period. This is an average BOD reduction of 97.5 percent for the
complete season. It should be kept in mind that many operational
problems were experienced during the season which actually decreased
the efficiency of the treatment plant. The actual capabilities of the
plant were demonstrated the latter part of the season when the foaming
problem had been solved by removing the wastewater containing orange oil.
Data collected during the period from June 4, 1969, through July 2, 1969,
showed BOD values of the influent averaged 1000 mg/1 during these four
weeks, and BOD values of the effluent averaged 7 mg/1 during the same
period. This is a treatment efficiency of 99.3 percent.
LIME TREATMENT PROCESS - OPERATION PROBLEMS
There was a variety of operation problems associated with the lime
treatment tank. The lime treatment unit was initially designed to remove
all waste activated sludge with the precipitated CaCOo sludge. The only
problem with this was the fact the lime sludge was much heavier than the
waste activated sludge; therefore, it settled to the bottom of the lime
tank. If solids were removed fast enough to keep the activated sludge
from going over the weirs, all of the lime sludge was quickly removed
leaving no lime sludge blanket. This problem was solved by wasting the
excess activated sludge directly to the wet well where the lime sludge
was wasted. The two sludges mixed and were then pumped to the citrus
processing plant for recovery as cattle feed.
Although activated sludge was not intentionally discharged to the lime
tank after the first few weeks of operation, unintentional discharge of
solids occurred during most of the citrus season, with the exception of
the last few weeks of operation. The main reason for this was the foam
carry over from the aeration tanks. The inability to produce a clear
effluent was the major result of this solids carry over. This signifi-
cantly affected the treatment efficiency of the lime treatment process.
Only during the last few weeks of operation, after the wastewater con-
taining orange oil was removed from the treatment plant, did the lime
tank produce an acceptable and clear effluent.
A series of coagulant aids was tested. Starch proved to be useless
because of the high concentration of activated sludge solids entering
the lime tank. Alum was used with similar results. Only after using
lime alone was any significant result obtained in settling wastewater
containing high concentrations of activated sludge. No problem was
experienced with obtaining a clear effluent as long as the effluent from
the sludge clarifier contained a low solids concentration.
TREATMENT EFFICIENCY OF THE LIME UNIT
FOR BOD, COD AND SOLIDS REMOVAL
Judgment of the efficiency in lime treatment of an activated sludge
effluent cannot be based on the chemical data obtained during the com-
plete citrus season. This is because the waste treatment plant was not
185
-------
Influent
3.80
0.49
3.14
0.04
0.01
0.02
3.20
1.23
1.20
0.75
Effluent
4.95
0.69
4.28
0.05
0.01
0.03
2.10
0.67
0.69
0.25
Influent
2.55
0.84
1.78
0.03
0.01
0.02
2.33
2.07
1.26
0.97
Effluent
1.88
0.25
1.60
0.01
0.00
0.01
0.46
0.30
0.16
0.11
operating properly on a continuous basis until the last few weeks of
the citrus season. Only after the foaming problem was solved and the
amount of activated sludge solids entering the lime treatment unit was
reduced could the data be considered as representative of a properly
operating unit.
Tabulated below are the average influent and effluent nutrient analyses
for the complete citrus season and for the period June 4 through June 12.
Total Citrus Season June 4-June 12
mg/1 as N or P
Total N
Ammonia N
Organic N
N03+N02
N02
N03
Total P
Ortho P
Total P (filtered)
Ortho P (filtered)
The effect of foaming and carry over of solids was most pronounced for
the nitrogen determinations. The fact that the effluent nitrogen con-
centrations were higher than the influent concentrations can be attributed
to the sampling method. All nutrient analyses on the influent were run
on a settled sample. This in effect removed most of the solids associated
with foam carry over. Nutrient analyses on the effluent were made on the
complete sample, which included all solids not effectively removed in
the lime treatment unit. Due to the quantity of activated solids going
into the lime unit during a considerable portion of the citrus processing
season, it is probable that a portion of these solids were placed into
fine colloidal form or into solution due to the mechanical interaction
of the lime solids. This would also account for higher nitrogen concen-
trations in the effluent than the influent.
Lime tank total influent phosphorus concentrations averaged 3.20 mg/1
while effluent concentrations averaged 2.10 mg/1 during the complete
citrus processing season. This is a removal efficiency of only 34 per-
cent. Again, the main reason for this lower removal efficiency was the
foaming problem. During the period June 4 through June 12, influent
total phosphorus concentrations averaged 2.33 mg/1 and effluent concentra-
tions averaged 0.46 mg/1. This is a treatment efficiency of 80 percent.
However, the use of lime treatment for phosphorus removal in an activated
sludge effluent containing only 2.5 mg/1 of phosphorus is not economically
justifiable.
186
-------
REUSE OF LIME TANK EFFLUENT IN-PLANT
Effluent from the treatment plant was successfully reused in-plant for
considerable periods of time. Reuse was limited to use as barometric
leg water for two reasons. Primarily, the pH from a lime treatment
unit will cause deposition of CaC03 in pipes and on equipment unless
the pH is lowered much closer to the pHs. Secondly, extensive piping
changes would be required to reuse the water for cooling purposes in
the citrus processing plant.
There is no question that the lime treated effluent can be reused as
barometric leg water and cooling water at Winter Garden Citrus Products
Co-op, by simply adjusting the pH with C02 and installing the required
piping. The only problem is that presently it is cheaper and easier to
pump well water from the abundant supply than reuse the waste treatment
plant effluent.
CENTRIFUGE OPERATION
Centrifuge operation was one of the weakest points in the overall treat-
ment efficiency of the plant. Overall efficiency of any treatment
plant depends greatly on removal of waste sludge from the system, in
this case, for recovery as cattle feed. Thickening the waste sludge was
necessary to prevent overloading of the rotary dryers. Therefore, a
centrifuge was installed for the thickening operation.
The centrifuge never operated to complete satisfaction during the entire
citrus season, the main problem being inability to keep it in continual
operation because of shear pin failure due to overloading. Early in the
season a major mechanical failure occurred in the centrifuge and it had
to be rebuilt.
One of the main factors contributing to the poor performance of the
centrifuge was that the entire treatment plant design was based on
parameters developed during a six-week laboratory study. No pilot plant
studies were performed; therefore, there were no firm parameters by
which to design the centrifuge. Another major problem was the variability
in the waste sludge entering the centrifuge. Waste sludge from the
treatment plant was pumped to a holding tank in the citrus processing
plant before it was centrifuged. The variation of the solids from this
surge and thickening tank was considerable. Lime solids tended to
separate from the activated sludge solids and settle to the bottom.
After a period of time when no solids had been withdrawn, the solids
concentration increased to as high as 10 to 12 percent. A sudden slug
of this material invariably sheared a pin in the centrifuge. Even after
a fairly consistent solids concentration was fed to the centrifuge, it
was difficult to keep it continuously operating.
The manufacturer spent considerable time trying to get the machine to
perform properly. Finally, toward the latter part of the season, the
centrifuge rpm was reduced, and this seemed to be the answer for continuous
operation.
187
-------
The inability of the centrifuge to operate properly meant that solids
could not be wasted as desired. When the centrifuge was out of opera-
tion, thickened solids from the bottom of the surge tank were put
directly on the citrus peel for drying. This increased the amount of
water going to the dryers, which were already loaded to a maximum. As
a result, solids built up in the waste treatment plant, compounding the
foaming problem.
Almost all of the waste sludge from the treatment plant was recovered
as cattle feed. This meant that several hundred tons of waste solids
were included with the citrus peel and recovered as cattle feed. All
of this feed has been sold and consumed by cattle. It should be kept in
mind that the waste sludge represented only about three tons out of a
total 350 tons of cattle feed produced each day. Therefore, it was
ultimately only a fraction of the entire cattle feed production.
CONCLUSIONS
1. The complete mixed activated sludge process will treat wastewater
from the citrus processing industry with an efficiency greater than 99
percent.
2. All waste sludge can be reclaimed along with the pressed citrus peel
as cattle feed.
3. Effluent from the waste treatment system can be reused in the citrus
processing plant.
4. Effluent nitrogen and phosphorus concentrations can be more effec-
tively controlled by controlling the addition of nitrogen and phosphorus
to the influent wastewater than by attempting to lime treat the activated
sludge effluent while adding an excess of nutrients to the influent.
5. Addition of large quantities of orange oil and peel press liquor
into the activated sludge process will completely disrupt the treatment
efficiency and cause heavy foaming.
6. High production of excess solids is one of the key factors to be
considered when treating citrus waste by the activated sludge process.
188
-------
SEAFOODS PROCESSING: POLLUTION PROBLEMS AND
GUIDELINES FOR IMPROVEMENT
by
. R. Soderquist, K. J. Williamson, and G. I. Blanton*
ABSTRACT
This paper contains discussions of the processing of the major
United States seafoods species, the resultant wastewater strengths
and flows, solid wastes magnitudes, current treatment and by-product
recovery methods, and current and recommended research in water
pollution abatement. The paper was based on a comprehensive
literature review and extensive on-site investigations of current
research, processing, and treatment activities in the major seafoods
centers of the U.S. The reader desiring an exhaustive review of the
entire industry, including all major species commercially harvested
in the United States, is directed to reference number 1 at the back
of this paper.
This study was supported by the Federal Water Quality Administration,
U.S.D.I., through Grant Number 12060ECF. Major assistance in the
national survey segment of the investigation was provided by the
National Canners Association and the U.S. Bureau of Commercial
Fisheries.
INTRODUCTION
The present world marine harvest stands at approximately 50 million
tons per year. Ninety percent of this catch is fish, the remainder
being whales, crustaceans and mollusks. From 1850 to 1950, this
harvest increased at an average rate of 2.5 percent per year. During
the last two decades this increase has jumped to 5 percent per year (2)
Some observers believe that even with present methods of fishing, the
yield can be increased to 5 to 10 times the present value. More
conservative analysts estimate a possible increase of 2 to 3 times the
present yield (3).
JL
Respectively, Instructor, Department of Food Science and Technology,
Oregon State University, Corvallis, Oregon 9733.1; Instructor, Department
of Food Science and Technology, Oregon State University, Corvallis,
Oregon, 97330; and Instructor, Green River Community College, Auburn,
Washington (formerly: Research Associate, Department of Food Science and
Technology, Oregon State University.)
189
-------
The increased catches of the 1950fs and 1960fs were due mainly to
the intensified fishing efforts of a few nations: Peru, Japan, and
the Soviet Union. The annual United States catch, meanwhile, has
been steadily declining; since 1962 U.S. fish harvests have decreased
20 percent. The most common reasons cited for this decrease are low
efficiencies, insufficient and expensive labor, and governmental
restrictions (4). Based on recent performance, the United States
seafoods industry is not expected to expand significantly in the near
future.
The annual U.S. catches average approximately 6 billion pounds (5).
The fish are utilized as follows: 35 percent rendered, 30 percent
marketed fresh, 20 percent canned, 10 percent frozen, and 1 percent
cured (6). Frozen fish products have been increasingly popular items;
a 150 percent increase in frozen fish sales over the next 15 years
has been predicted (7).
The U.S. consumption of fishery products has continued to rise, as
shown on Figure 1. These increasing demands, however, have been
satisfied by imported products. This increase in consumption has
been almost exclusively due to the population increase; the U.S. per
capita consumption of seafood products has remained at approximately
11 pounds per year over the last 20 year (5).
A significant portion of the fishes and shellfishes processed is wasted.
The percentage of each species wasted ranges from 0 percent for fish
which are completely rendered (e.g., menhaden) to 85 percent for some
crab (e.g., blue crab). The average wastage figure for all fish and
shellfish is about 30 percent. In addition to these large volumes of
solid wastes, significant wastewater flows result from the butchering,
washing and processing of the products. The volumes of solids and
wastewater vary widely with seafoods and processing methods.
Using the 30 percent figure, the total annual volume of solid wastes
generated in the seafoods industry in the United States is roughly
1.2 billion pounds. A large portion of these wastes is rendered for
animal feed. The remainder is taken to municipal or private disposal
sites or discharged directly to adjoining waters. The pollutional
strength per pound of fish waste has been estimated to be 0.2 pounds
of five-day biochemical oxygen demand (6005), or approximately one
daily population equivalent (8). As will be discussed later, this
figure is highly variable and probably represents only the carbon-
aceous demand, neglecting the nitrogenous demand which would be exerted
somewhat later. Using this figure and assuming that 50 percent of the
fish wastes, as an average, are rendered, the population equivalent of
this industry can be estimated to be 2 million people. The population
equivalent of solid and liquid fish processing wates has been estimated
190
-------
16 -
14
12
10
CO
CD
g
_i
CD
8
i
TOTAL SUPPLY
DOMESTIC CATCH
1959 '60 '61 '62 '63 '64 '65 '66 '67 1968
FIGURE I SEAFOODS CONSUMPTION IN THE U.S. (5)
191
-------
by one author to be from 66 to 1020 per ton of fish (9). For the U. S.,
this represents a population equivalent of 0.23 to 3.6 million. These
figures are deceptively conservative, for a major segment of the seafoods
production takes place during short seasons, intensifying the problem.
The industry is not typified by a constant output month after month.
The fish processing waste problem has become serious in certain areas.
Waste treatment will be necessary in the future to meet Federal and state
water pollution control regulations. The purpose of this study was to
evaluate the present state of the art of fish processing waste treatment
and by-product recovery and to suggest research necessary to advance this
technology to meet future needs. Only six major U. S. seafoods categories
are considered here.
THE INDUSTRIAL FISHERIES
Menhaden, herring, and alewives are oily fishes which comprise the bulk
of the "industrial fisheries" in this country: the rendering of whole
fish into meal, oils, and solubles. The meal is used primarily as
animal feed and fertilizer; the oil becomes an ingredient in paints,
varnishes, resins, and similar meterials, is added to animal feed, or
is used for human consumption abroad; and the solubles are either fed
directly to animals or are dried and processed into meal.
The menhaden fishery is the largest in the United States; the 1968 catch
totaled 1.4 billion pounds (5). The industry is located mainly in the
middle Atlantic and the Gulf states. Fishing takes place predominantly
during the summer and fall months.
In many cases, the menhaden rendering operations are highly mechanized.
The process, outlined schematically on Figure 2, involves first the har-
vesting of the fish in purse seines and storage in the holds of the
fishing boats. Ice or refrigeration is used for preservation if the
trip exceeds one day. At the plant, the fish are pumped from the holds,
washed, automatically sprayed and conveyed into the plant. Cooking is
done continuously by steam. The fish are then pressed to remove the oil
and most of the water. This pcesswater is screened to remove solids and
centrifuged to separate the oil. The remaining water, called stickwater,
is discharged or evaporated to produce condensed fish solubles. The solid
residual from which the water and oil have been pressed is known as press
cake. The press cake is dried to about 10 percent moisture and then ground
for fish meal.
Since the menhaden fishing areas are presently largely exploited to the
maximum, significant future growth in this industry is not anticipated.
Continuous production at levels near those of recent years is expected.
192
-------
PROCESS
WASTES
DISPOSAL
1 RECEIVE | ( SLIME, WATER )
( OIL, WATER )-—
STICKWATER V-
\ [ WASH DOWN K WATER, SOLIDS ^RECEIVING
FIGURE 2
MENHADEN RENDERING
193
-------
In a properly managed menhaden rendering plant, the quantities of
wastes produced are small. The only inherently troublesome waste
is the fish pumping water. The other wastes result from spills and
leakages, both of which can be minimized.
The wastewaters from the production of fish meal, solubles, and
oil from herring, menhaden and alewives can be divided into two
categories: high-volume, low-strength wastes and low-volume, high-
strength wastes.
The high-volume, low-strength wastes consist of the water used for
unloading, fluming, transporting, and handling the fish plus the
washdown water. One author (10) estimated the fluming flow to be
200 gallons per ton of fish with a suspended solids loading of
5,000 mg/1. The solids consisted of blood, flesh, oil and fat.
These figures vary widely; other authors listed herring pump water
flows of 250 gpm with total solids concentrations of 30,000 mg/1
and oil concentrations of 4000 mg/1 (11). The bilge water in the
boat was estimated to be 400 gallons per ton of fish with a
suspended solids level of 10,000 mg/1.
The strongest wastewater flows are the stickwaters. In the past,
stickwater was often discharged to the receiving waters, but now
this practice is generally forbidden by law. One reference listed
the average BOD^ of stickwater as ranging from 56,000 to 112,000
mg/1, with average solids concentrations, mainly proteinaceous,
ranging up to 6 percent (12). Fortunately the fish processing
industry has found the recovery of fish solubles from stickwater to
be at least marginally profitable. In most instances, stickwater
is now evaporated to produce condensed fish solubles. Volumes have
been estimated to be about 120 gallons per ton of fish processed (13).
TUNA
Tuna ranks as the number one seafood in the United States. Americans
consume over one billion cans of tuna per year (14). Tuna are large
migratory fish which feed on smaller macroscopic sea life. Their
distribution in the ocean is still largely unknown. The largest part
of the tuna harvest takes place off the Pacific Coast.
Most tuna canned in the United States are caught in relatively
distant waters. A modern tuna vessel can hold from 150 to 300 tons
of fish and has a range of 1,000 miles (15). Because of the length
of time in transport, the fish are normally frozen on board the
vessels. After the boat arrives at dockside, the tuna canning opera-
tion, as shown on Figure 3, begins with the mechanical unloading of
the fish, which are weighed and inspected while they are still frozen.
194
-------
1
r-1
Ir
at —
///
n ,-
,"//
I11 \
Un / —
( WATER ^
I SUPPLY )
\\\\
\\\\
\ \\ \—
^\
\\\._
\ \
\v~
\
V
FIGURE 3
PROCESS
/^RAWN
L PRODUCT/
I
STORE
I
THAW — (
I
RIITPHFR —I
ou i wrii_r\ ^^
|
WASH — (
I
DPF- f*r\r\it —/
rrxc. uuur\ ^
1
COOL — (
1
n FAN -/
OL_k./nlN ^^
1
PACK — (
1
CLINCH — (
1
SEAM
l
CAN WASH — (
1
RETORT -- <
1
COOL — (
1
CASE
i
WASH DOWN I— - <
^___L_^^
/FINISHEDN
L PRODUCT/
TUNA CANNING
WASTES DISPOSAL
(TREATMENT]
i
SOLIDS, WATER ) 1
\/IC?PFT?A ni nr\n V_ 1
WATER ' j
BLOOD, OIL, N 1
SLIME '
tTlfKWATFR \ ,.
WATER ) 1
rtFFAl V-
MCAT V 1 ,.
SOYBEAN OIL ) k^/^ ^^\
j /RENDERING]
SOYBEAN OIL ) ' \^ J
CONDENSATE ) '
WATER ) 1
I
1
1
WATER , SOLIDS ) 1
195
-------
Then the tuna are thawed. The next step, evisceration, is
normally conducted by hand in several phases. The body cavities
are flushed with water and all adhering viscera carefully removed.
The viscera are used for fish meal or pet food and the livers are
sometimes recovered for oil and vitamins. After butchering, the
fish are steam precooked. The time of cooking varies with the
fish size, but it is usually about three hours. This cooking
removes 22 to 26 percent of the moisture (15). Following cooking,
the fish are cooled for approximately 12 hours to firm the flesh.
The meat is then separated by hand from the head, bones, fins and
skin. All dark meat is removed and usually recovered for pet food.
The light meat for canning is placed on a conveyor belt and trans-
ferred to the "Pak-Shaper." Tuna slices are arranged lengthwise
in the Pak-Shaper which then molds the meat into a cylinder, fills
the cans and trims the meat after filling. Each machine has a
capacity of from 125 to 150 cans per minute (15). Prior to vacuum
sealing, salt and vegetable oils are added to the cans. After
sealing, the cans are retorted by standard procedures.
The annual tuna catch averages approximately 300 million pounds,
almost all of which is canned. The U.S. catch in recent years
has failed to meet the increased domestic market. Use of
scientific methods to determine fish migration should increase
future catches and enable the domestic market to expand. A slight
upward trend in tuna production has been evident now for the past
six years and is expected to continue (16).
It has been estimated that 65 percent of the tuna is wasted in the
canning process (17). Using this figure, the 1968 quantity of
wastes generated in the United States was calculated to be 190
million pounds. The degree of wastage probably varies somewhat
with species. A recent study examined in detail the waste from a
tuna canning and by-product rendering plant for a five-day period (18)
The average waste flow waa found to be 6800 gallons per ton of fish.
Organic loadings varied from 500 to 1550 mg/1 BOD5. The average
daily COD ranged from 1300 to 3250 mg/1 and the total solids averaged
17,900 mg/1 of which 40 percent was organic. The mean values and
their expressions in terms of pounds per ton of fish are listed
on Table 1.
196
-------
Table 1. Tuna Waste Characteristics (18)
Mean Concentration
Parameter
COD
BOD5
Total Solids
Suspended Solids
Grease
mg/1
2,273
895
17,900
1,081
287
Ibs/ton of
129
48
950
58
15
fish
These figures indicate that the five-day BOD of the tuna waste was
only approximately 40 percent of the COD value. Due to the high
proportion of particulate matter in the total solids (leading to low
surface-area-to-volume ratios), and to the organic nature of the
wastes, a considerable BOD is exerted after five days. In this case,
at 22 days, the BOD exerted was 3520 mg/1 and was noted to still be
rising. It is important to realize that the waste exerts a considerable
nitrogenous demand (that oxygen required to oxidize the less reduced
forms of nitrogen to nitrate) in excess of the five-day carbonaceous
value.
SALMON
The only significant commercial anadromous fishery in the United
States is the salmon fishery. The five main species harvested in
this country are chinook (king), sockeye (red), silver (Coho), pink
and chum. The major portion of the catch is canned.
The fish are caught fairly close to the canneries and are often stored
in the boats without refrigeration. Canning operations are conducted
for the most part employing standard cannery equipment in a conven-
tional manner as shown on Figure 4. After transfer from the holds of
the fishing boats, the fish are eviscerated, beheaded and the fins
removed. The raw meat, including the bones, is then placed into cans,
the cans are weighed, and their weights are adjusted (if necessary)
prior to sealing. The cans are then seamed, retorted, cooled and
cased prior to shipping.
The principal exception to this standard canning operation is the use,
predominantly in Alaska, of the "iron chink". The iron chink performs
several functions in one operation, removing heads, fins and viscera
mechanically.
197
-------
PROCESS
WASTES
DISPOSAL
STORE |~-^ BLOOD, SLIME )--j
EVISCERATE h -( VISCERA , WATER
I . .
BEHEAD H—(
1
SLIME
1
M FINS, WATER,
BLOOD, SLIME
PACK
I
WEIGH
I
MEAT
MEAT
>-
>-
>^
PATCH
I
SEAM
1
RETORT
1
COOL
I
CASE
}--(MEATt WATER )—-f
I
I
>-\
>--{
l
h-(
h-<
WATER
WATER
'--I WASH DOWN |—( WATER,SOLIDS)—I
FIGURE 4
SALMON CANNING
198
-------
The total 1968 salmon catch was 301 million pounds compared with
the average over the previous five years of 335 million pounds per
year (5). The harvesting is centered around Alaska with significant
contributions from Washington and Oregon.
The Pacific salmon fishery is now advancing after a general failure
which occurred in 1967. The future of this industry is largely
dependent on market conditions, pressure from foreign competitors
and on future conservation practices. A major expansion of the
salmon industry in this country is not anticipated. Continued pro-
duction at or near current levels is expected (5).
The quantities and possible uses of salmon wastes have been rather
thoroughly researched. One investigating team (19) found salmon
processing to produce 34 percent waste. Similar estimates were
reported by other sources: 33 percent C20), 27 percent (21), and
30-35 percent (17) . The Bureau of Commercial Fisheries (22) listed
waste fractions by species as: chinook, 30 percent; red, 33 per-
cent; Coho, 33 percent; pink, 35 percent and chum, 33 percent.
Using the BCF values of wastage and total catch (5), waste volumes
were calculated to be as shown on Table 2.
Table 2. Calculated Salmon Waste Quantities, 1968
Species
Chinook
Chum
Pink
Red
Coho
Total
Wastage
(%)
30
33
35
33
33
Quantity
(Ibs x 106)
7
26
36
18
12
100
Of these wastes, from 50 to 61 percent (depending on species) consists
of heads and collars. Another 11 to 16 percent is made up of tails
and fins. Other waste segments in decreasing order of magnitude are:
liver, roe, milt, digestive tract and heart (23).
Claggett and Wong (11) listed the flow from a salmon canning line at
300 gallons per minute with a total solids concentration of 5,000
mg/1 and oil concentration of 250 mg/1. Other investigators have
characterized the wastes from salmon canning operations as shown on
Table 3.
199
-------
Table 3. Salmon Processing Waste Strengths
Process
Canning (24)
Mild curing (24)
Mild curing (24)
and fresh
Mild curing (24)
and fresh
Caviar (24)
Canning (25)
Flow
(mgd)
0.043-0.046
0.015-0.066
0.011-0.036
0.014-0.046
0.33
COD
(mg/1)
5,920
—
—
—
—
BOD5
(mg/1)
3,780
173-1,320
206-2,218
397-3,082
270,000
3,680
BOD5/raw
product
(Ibs/ton)
65.2-178.2
10.2-80.0
3.3-36.0
3.8-18.6
—
Suspended
Solids
(mg/1)
508-4,780
44-456
112-820
40-1,824
92,600
2,470
Total
Solids
(mg/1)
1,188-7,444
258-2,712
484-2,940
88-3,422
386,000
Volatile
Solids
(mg/1)
1,048-7,278
98-2,508
84-1,756
67-2,866
292,000
NS
O
O
-------
The values for all parameters are quite variable, the wastewater
strength probably depending on the efficiency of solids removal.
The BOD5 concentrations ranged from 200 to 4,000 mg/1; suspended
solids, 40-5,000 mg/1; total solids, 80-8,000 mg/1; and volatile
solids 60-7,000 mg/1.
Caviar production often takes place with salmon canning. This
process results in extremely strong wastes, but the flows are small.
SHRIMP
The shrimp industry is the most important seafood industry of the
Gulf of Mexico and South Atlantic areas. It is also significant
along the Pacific Coast. The season runs from April to early June
and again from August to early October (26).
Shrimp are caught commercially in otter trawls in the coastal waters.
The shrimp are separated from the trash fish and stored by various
methods. When short storage times will suffice, no preservation
methods are used; the shrimp are taken directly to a processing
plant or to a wholesale marketing vessel. When longer storage times
are necessary, the shrimp are iced in the holds. In some places,
notably in the Gulf of Mexico area, the shrimp are beheaded at sea
and the heads dumped overboard. The heads contain most of the
active degradative enzymes, so this practice retards spoilage. If
the shrimp are beheaded within 30 minutes after being caught, the
intestinal veins are usually removed with the heads, which is
desirable from a quality standpoint.
The shrimp are unloaded from the vessels into a flotation tank at
dockside to remove the packing ice and then conveyed to a rotary
drum to remove surplus water and bits of debris. This is followed
by weighing. In Texas and the South Atlantic states particularly,
the shrimp are iced after the initial preparation to optimize
peeling conditions.
Next the shrimp are peeled (or picked) by hand or machine. Machine-
peeled shrimp are used mostly for canning (21). The machine-peeled
shrimp are paler in color, have a poorer flavor and have a texture
inferior to hand-picked shrimp. These disadvantages are offset
by the fact that an automatic peeler can handle 500 pounds of shrimp
per hour compared to average rates of 100 to 400 pounds of shrimp
of shrimp per day per man for hand picking (26).
After peeling, the meats are inspected and washed. They are then
blanched in a salt solution for about 10 minutes and dried by various
methods to remove surface water. Again the shrimp are inspected and
then canned. The mechanical peeling process is outlined on Figure 5.
201
-------
PROCESS
WASTES
DISPOSAL
WATER
SUPPLY
f RAW A
V, PRODUCT/
| STORE
1
H
/r-C
/// — c
In L
//
'// i
3 \ ,_
y 1
/-H
\\\ 1
\\\ i.
\\H
\\ _
\
\
PEEL
l
WASH
1
BLANCH
1
BLOWER
1
INSPECT
1
GRADE
1
PACK
1
CITRIC ACID
l
SEAM
i
RETORT
1
COOL
l
CASE
l
^-- fWASH DOWN
(TREATMENT)
— —/ 'iHE'Ll WATFR \—
_ ~J QMP*I 1 WATFO \^
~^ ontUU, WM 1 tr\ /^
--{ MEAT, WATER )-—
( HELL )
,/ MEAT \
\ MUAI /
/ MEAT \
\ MtA! ^
/ MEAT \
- -( WATER )
- -{ WATER )
__( WATER )--
( WATER , SOLIDS )
r
/^
RENDERING
O
FIGURE 5
MECHANICAL SHRIMP PEELING
202
-------
Shrimp are marketed fresh, frozen, breaded, canned, cured and
as specialty products. An increasing amount are sold breaded
or fresh-frozen, whereas the quantities of canned shrimp produced
in recent years have been relatively constant. About 40 percent
are sold frozen in the, shell (27).
The shrimp fishery in terms of total value is the most important
in the United States. In 1968, the catch exceeded 290 million
pounds with a value of approximately 110 million dollars. Currently
the most important finished products are frozen and breaded shrimp.
In 1968, these two products comprised 92 percent of the market.
Both of these products were successfully developed during the 1950's
and markets are apparently continuing to expand.
Except in Alaska, the catch areas appear to be fully exploited.
Yearly variation in catch seems to be dependent on the survival of
the population from previous year. The Alaskan catch, now less
than one-fourth of the national total, could expand substantially
with further developments (5). One writer predicted that the Alaskan
stocks are capable of producing a catch equal to or exceeding 250
million pounds annually, or 5 times the existing Alaskan catches (38).
Jensen (21) estimated that 78 to 85 percent of the shrimp is wasted
in mechanical peeling and 77 to 85 percent in hand picking. The
Oregon State Department of Environmental Quality (17) estimated 78
percent for hand picking and the Bureau of Commercial Fisheries
listed a cleaning loss of 55 percent (22). The low value from the
Bureau of Commercial Fisheries was apparently due to ignoring the
blanching loss, which ranges from 30 to 35 percent of the picked
wastes (26). Using a value of 80 percent wastage, the quantity of
shrimp wastes generated in 1968 was calculated to be 233 million
pounds.
Shrimp waste has been analyzed and been shown to be predominantly
protein, chitin (a complex polysaccharide, not readily biodegradable)
and calcium carbonate, as outlined on Table 4.
Table 4. Composition of Shrimp Wastes (29)
Source Composition
Protein Chitin Calcium Carbonate
Hand picking
Mechanical picking
27.2
22.0
57.5
42.3
15.3
35.7
203
-------
The protein concentration in shrimp waste has been judged satisfactory
for animal feed (30). Crawford (31) reported that mechanical shrimp
peeler effluents sampled averaged 29,000 mg/1 total solids.
CRAB
The blue crab, comprising 70 percent of the U.S. crab production, is
harvested on the Atlantic Coast, principally in the Chesapeake Bay area.
The remaining harvest takes place on the Pacific Coast where Dungeness
crab is the leading species followed by Alaskan king crab. Crabs are
harvested from shallow water in baited traps. Rapid and careful hand-
ling is necessary to keep the animals alive. Dead crabs must be dis-
carded because of rapid decomposition.
At most canning plants, the whole crabs are steam-cooked in retorts for
20 to 30 minutes (32). Pacific Coast Dungeness crab operations differ
in that they first butcher the crabs (i.e., remove the backs) and then
cook the crabs for 12 minutes or less. Cooked crabs are also marketed
in the shell, butchered, or the meats picked from the shell are marketed
fresh, frozen, or canned. The majority of the Atlantic blue crab meat
is marketed fresh or frozen, while the majority of the Pacific crab meat
is canned (33). A large quantity of Dungeness crab is sold in the shell
and a large quantity of king crab is butchered at sea (34) . Both practices
reduce the quantity of butchering waste to be handled at the processing
plant site.
Figure 6 depicts a frozen crab operation. Crab canning and crab freezing
operations are understandably similar. After cooking, the crabs are
water-cooled to facilitate handling. The backs are removed, if they
haven't been previously, and the remaining viscera are washed free. The
cooking, cooling, and washing waters contain considerable solids and
organic pollutants. After cooling, the meat is picked from the shell by
hand with a small knife. Mechanical methods have been recently developed
tb extract the meat from the shell but are not as yet widely employed (35) .
Crab meat quickly degrades in quality and must be chilled, frozen, or
canned. Chilled meat can be stored for only a few days and even frozen
meat looses texture and flavor qualities rapidly. Canning of crab meat
results in the additional wastewater flows of the retort and cooling waters.
The total crab catch in 1968 exceeded 238 million pounds (5) . The catches
of the three main species seem to have reached a plateau. Production
appears to be determined by the extent of previous years' hatches and the
extent of harvesting rather than by marketing conditions. Future harvests
of most species should continue at levels dependent on survival of off-
spring in the fishing grounds. Production of Alaskan king crab may event-
ually increase slightly due to stricter controls being imposed by the
Alaskan Board of Fish and Game (5) . The controls established a king crab
fishing season from 5 to 7 months long in Alaskan waters. In 1969 all areas
were closed on February 15 and not reopened until August.
204
-------
PROCESS
( RAW >>
\PRODUCTy
1
1 STORE
1
- A BUTCHER
/ 1
, — 1 COOK
// '
/VH COOL
/// 1
1 1 GILL
^. *jl/ \
S NX- H WASH
/ VVATER \ r— 1
( ^YippTv 1 — I TRIM
V J \ PACK
^***— "Cx* 1
X \ ^- H FREEZE
\ \ i
v ' — 1 GLAZE
^ i
\ | CASE
, I
^-H WASH DOWN
/FINISHEDN
^PRODUCT/
FIGURE 6 CRAB FREE2
WASTES DISPOSAL
TREATMENT]
i
/ VISCERA, SHELL A ,
\ WATER / '
- -( SOLIDS , WATER ) •
— ( WATER )--|
( Vl^rFRA S ...... ..
--( WATER ) 1
— —/ MPAT QMPI f \—
\r
--< WATER y-^ (RENDE
( WATER )--! \
1 V^
1
- -( WATER , SOLIDS ) j
[RECEIVING]
V WATERS I
:ING
:RING]
205
-------
The tanner crab harvest has been increasing in recent years due to the
decreased availability of king crab. Abundant stocks exist off the
Northern Pacific Coast and harvesting of this species should rapidly
increase (28).
The major portion of the crab is not edible and as a result is wasted
in processing. The waste consists of the shell and entrails and amounts
to approximately 80 percent of the crab by weight. Large quantities of
water are necessary for cooking, cooling, and washing of the entrails
from the body. The wastage of the total crab has been reported as:
blue crab, 86 percent (32); king crab, 80 percent (21); and Dungeness
crab, 73 percent (22). Using these figures, the solid waste load for
crabs for 1968 was calculated to be 194 million pounds, as shown on
Table 5. The actual wastes volumes at the processing plants were less,
Table 5. Calculated Quantities of Crab Waste, 1968
Species Waste Fraction Waste Quantity
(%) (Ibs x 106)
Blue 86
Dungeness 73
King 80
Total
94
32
68
194
because a significant proportion of the crabs harvested (especially Duigeness)
ace marketed whole or butchered whole to remove only the backs and entrails.
The composition of shellfish wastes is largely determined by the exoskeleton.
The exoskeleton is composed primarily of chitin, protein bound to the
chitin, and calcium carbonate. The major portion of the wastes consists of
exoskeleton materials with varying significant amounts of attached or
unrecovered flesh and visceral material included. The Ketchikan Tech-
nological Laboratory of the Bureau of Commercial Fisheries reported typical
compositions of these wastes as shown on Table 6.
206
-------
Table 6. Typical Crab Waste Compositions (29)
Tanner
Protein
Composition
Chitin Calcium Carbonate
Species
King
Tanner
Sources («) («)
Picking line 22.7 42.5
Leg and claw ..„ -, _, ,
(%)
34.8
n n
shelling
Body butchering
and shelling
21.2
30.0
48.8
The protein level is considered low compared to visceral fish wastes, thus,
this material is considered only marginally satisfactory as an animal
feed.
Crab processing wastes, like shrimp, include large volumes of processing
washwaters, solid wastes, canning waters (where applicable) and plant
clean-up water. No information was found in the literature describing
the organic strengths and liquid waste volumes from shellfish processing,
although currently on-going research may fill this void.
BOTTOM FISH
The most important bottom fish species are listed by Slavin and Peters (36)
as haddock, halibut, cod, ocean perch, whiting (silver hake), flounder,
hake, and pollock. Approximately 30 percent of the industry is located
in the North Atlantic region. The major halibut fishery is centered in
the Pacific Northwest where the commercial season extends from April
through October.
Bottom fish are usually caught in otter trawls. In a typical operation,
the fish are spread on the trawler decks, sorted and iced. Perch, flounder
and whiting are stored whole, whereas cod, haddock, halibut and pollock
are usually eviscerated on deck. The viscera and blood are washed over-
board .
Bottom fish are normally filleted. The typical filleting operation is
depicted on Figure 7. After unloading, the fish are weighed, washed,
and iced in tote boxes. In some larger plants, mechanized unloading
methods are used to maintain quality. In small plants, the fish are
processed by hand. The fillets are cut on a wooden board next to a sink,
washed and immediately iced in boxes for distribution.
207
-------
PROCESS
WASTES
DISPOSAL
FINISHED
PRODUCT
] ( SOLIDS )-
--( SOLIDS, WATER )
-( SOLIDS, WATER }—
I / SOI IDS, SKINS, \
I\ WATER /
( SOLIDS .WATER )
( WATER ) j
]~( WATER .SOLIDS ) j
RECEIVING
WATERS
FIGURE 7 BOTTOM FISH FILLETING
208
-------
Most plants processing fillets use mechanized equipment. First, the
fish are washed with sprays of water in large rotating tumblers; then
they pass through filleting machines or along hand filleting tables.
Filleting machines only operate on certain sizes and shapes, but they
are considerably more efficient and economical than hand filleting.
The skin is removed from the fillet by hand or machine. The solid
wastes from filleting and skinning operations are usually rendered for
pet food or animal meal.
The skinned fillets are transported by a conveyor belt to a washing and
at times a brining tank. After inspection, the fillets are packed in
containers or frozen and then packed. Fillets are marketed frozen
(fresh or breaded), chilled, or fresh.
In 1968 the bottom fish catch exceeded 490 million pounds, of which
approximately 190 million pounds was harvested on the Atlantic Coast
(5). The total domestic supply of bottom fish fillets and steaks has
been steadily declining since the 1940's (5). The major reason for the
drop in domestic production seems to have been lower yields in Atlantic
Coast waters. The halibut production, on the other hand, has been
predicted to remain, in the near future, at approximately the 1968 level
of 26 million pounds (21). This estimate was based.on consumer demand,
requirements for growth and limits imposed by the International Pacific
Halibut Commission.
About 35 to 40 percent of the halibut is wasted in processing (21).
The viscera are usually disposed of at sea. The remaining wastes
(heads, skins and fins) have been estimated to amount to approximately
12 percent of the total wastes (37). Using this 12 percent figure, the
halibut wastes in 1968 were calculated to total 3 million pounds. In
most other filleting operations, the fish are not eviscerated. The
unfilleted portions are discarded or recovered for by-products. Water
is run continuously in the spray-washers and during filleting and
skinning for bacteriological control. Blood and small pieces of fish
flesh are entrapped in this flow. Other waste flows include the packing
ice and the cooling waters.
The Oregon State Department of Environmental Quality estimated the solid
waste fraction from bottom fish processing to range from 35 to 40 percent
(17). Using the 40 percent value, the total waste quantity for bottom
fish in 1968 was calculated to be 189 million pounds. As noted earlier,
cod, haddock and pollock are eviscerated as caught and the wastes dumped
at sea. This decreased the estimate of total processing plant wastes
to 139 million pounds per year.
209
-------
Thurston (38) determined the composition of wastes from sole and
flounder processing. Composite samples were prepared from the non-
edible parts of 214 fish. The average composition was: moisture,
77.4 percent; oil, 5.68 percent; protein, 13.6 percent; ash, 3.84
percent; sodium, 0.16 percent; and potassium, 0.22 percent. Although
the non-edible parts of sole and flounder had lower values for protein
and ash than did those of other saltwater species, they were judged
to be of high enough quality for by-product utilization. The fish
averaged 72 percent waste. The waste flows from bottom fish processing
plants include large volumes of wastewater which contains blood, small
pieces of flesh, the body portion of the fish after filleting and the
skin. Claggett and Wong (11) measured the waste flow from a bottom
fish plant to be 450 gpm, with a solids loading of 750 mg/1. Other
investigators (24) have reported organic loadings varying from 192 to
640 mg/1 BOD5 with the average being 74 pounds of BOD^ per ton of fish.
Reported bottom fish processing wastewater characteristics are
summarized on Table 7.
Table 7. Bottom Fish Processing Wastewater Characteristics
Flow BOD5 Suspended Reference
Solids
(mg/1) _ (mg/1)
105 640 300 (25)
320-410 192-640 --- (24)
132 1726 --- (39)
450 --- 750 (11)
BY-PRODUCT UTILIZATION
Much effort has been devoted to the development of saleable by-products
from seafood processing wastes. From a standpoint of water pollution
abatement, however, only one by-product group has contributed signifi-
cantly to reducing the magnitude of the problem: fish meal and other
animal feeds. The other processes (discussed later) consume only a small
portion of the wastes and new wastes generated in the processes often are
more noxious and less biodegradable than the original waste. If any of
these methods were developed into profitable enterprises, however, the
revenues realized from these operations could conceivably be used to
partially defray the expense of disposing of the remaining wastes.
210
-------
The use of whole fish and fish scraps for animal feed has been studied
thoroughly. The fresh wastes or whole fish are usually processed to a
fish meal, fish oil or animal feed (cooked and canned). In addition to
meal and oil plants, there were in the United States in 1968 ten plants
processing crushed shell for poultry feed, four plants producing animal
feeds and two plants producing pelletized fish hatchery feed, all from
fish wastes and whole fish (40). Jones (41) discussed the various
species and their wastes that could be used for pet food for several
geographic areas of the United States. He concluded that now-discarded
fish and fish wastes will be needed in the future to satisfy increased
raw material demands. Several species of under-utilized fishes were
listed. However, if all amenable wastes were processed into fish meal
or other animal feed at this time, it is questionable whether the
market could absorb the increased supply.
Besides fish meal, oils, solubles, and other animal feeds, many products
based on fish wastes have been developed and/or promoted in the past.
These include protein hydrolysates, fats and lipids, enzymes, homones,
vitamins, chitin, glucosamine, fertilizers, lime, limestone, pearl
essence, glue, caviar and other roe products, shell products and fish
protein concentrate. The processes resulting in these products in
general fail to reduce significantly the magnitu des of the wastes from
which they are made. In most cases, only a very small fraction of the
waste is utilized and the remainder must still be handled.
STANDARD WASTE TREATMENT METHODS
The liquid wastes resulting from fish processing are most commonly
discharged to the adjoining waters. This practice has been restricted
in many areas recently because of the consolidation of plants and
intensified enforcement of long-standing and new water pollution
regulations. The alternative in many cases has been to discharge the
wastes to the municipal sewerage systems. In only one instance did the
literature describe a United States processor having on-site treatment
of fish processing wastes before discharge to a water body (42).
The specific difficulties encountered in the treatment of fish processing
wastes are, in large part, attributable to the characteristics of the
wastes. These are usually: high flows, medium to high 6005 and
suspended solids, and high grease and protein levels, compared to domestic
sewage. The frequently short processing season, high peak loadings, and
rapid biodegradability of the wastes also contribute to the problem.
Claggett and Wong (43) studied the effectiveness of screening the wastes.
Two specific screen types were tested: rotary and tangential. A 34-
mesh rotary screen made of stainless steel was first investigated. The
4 foot-long barrel section was rated at 100 gallons per minute. Solids
211
-------
were removed on a screw conveyor and blinding was prevented through
the use of high pressure nozzles. The tangential screen employed
two screening surfaces, each one square foot in area, sized at 20 and
40 mesh. The resulting operating capacities were 35 and 20 gallons
per minute, respectively. Both screens were judged successful for
salmon canning wastes (43). The results indicated that with the low
investment and operating costs associated with screening, a processor
could expect removals of over one-half of the total solids from his
wastewaters.
Jaegers and Haschke (44) stated that centrifuges can be effectively
used to remove fish pulp from the waste streams. Fats and proteins
can also be recovered with this method. However, centrifuging entire
waste streams would be very expensive when compared to other methods.
In the processing of oily fishes (sardines, herring, etc.), large
quantities of fats and greases are present in the wastewaters. Knowlton
(45) reported the fat and grease content of sardine canning wastewaters
to be from 1,000 to 30,000 mg/1 compared to 50 to 200 mg/1 for domestic
sewage. Fats and grease are present as flotables or as emulsions. When
the wastes are discharged untreated, serious problems can result if the
emulsified grease coalesces and rises to the surface of the receiving
waters.
Grease can be removed in clarifiers by two methods: flotation and
sedimentation. Limprich (39) reported that the application of 2.5 grams
of clay, 2.5 grams of lime and 100 milligrams of ferric chloride per
liter of waste gave an optimal precipitation with a resulting 75 per-
cent decrease in 6005. A similar procedure was described by Schulz (46)
using A1203, lime, and ferric chloride. Griffen (47) mentioned that
high fat and protein wastes can be treated with lime. Chlorination
before sedimentation was recommended to prevent the serious odor problems
that can result from rapid degradation.
Sedimentation tests of fish processing wastewaters were reported by
Buczowska and Dabaska (48). In two hours of quiescent settling, 32 per-
cent of the suspended solids were removed with a concomitant 6005 re-
moval of 25 percent. About 58 percent of the organic matter in the
wastewaters was in solution or colloidal suspension. Limprich (39)
stated that 58 percent of the suspended matter in fish wastes settled
out in two hours. Large volumes of sludge resulted.
A partially successful gravity clarification system was developed using
large quantities of a commercial coagulant called F-Flok (43). F-Flok
is marketed by the Georgia-Pacific Corporation and is derived from
]ignosulfonic acid. The floe formed slowly, but after formation
sedimentation rates of 4 feet per hour could be achieved. The summary for
a large scale test on salmon wastewater (Table 8) shows a maximum solids
removal of about 70 percent •
212
-------
Table 8. Gravity Clarification Using F-Flok Coagulant (43)
Coagulant Total Solids Protein
Concentration Recovery Recovery
(mg/1) (%) (%)
5020 68 92
4710 60 80
2390 47 69
An alternative to the normal sedimentation clarifier is the flotation cell.
The flotation technique relies on the entrainment of minute air bubbles
which float particles to the water surface. The resulting sludge blanket
is continuously skimmed from the surface. Two methods are used to entrain
the air bubbles in the flow, each method having Advantages over the other.
The first method uses mechanical aerators to "whip" air bubbles into
solution. Dreosti (49) reported that good laboratory results were obtained
using fish wastes with suspended solids levels up to 8,000 mg/1. Higher
suspended solids concentrations produced sludges that did not consolidate
well on the surface. For optimum results, Dreosti recommended a minimum
of air flotation and short agitation times. Coagulants improved the
removal efficiency; however, no mention was made of types or quantities
used. The minimum detention time was estimated to be five minutes.
Hopkins and Einarsson (42) reported on the results of a similar flotation
unit operating on fish wastes at a flow of 0.065 mgd. The resulting
sludge contained 15.5 pounds of grease and 35 pounds of fish solids per day.
The second flotation method involves flow pressurization. The total flow
or a part of the flow is pressurized and then passed into the flotation
unit which is at ambient pressure. The now supersaturated solution begins
to release air, forming many tiny bubbles. These bubbles then float the
suspended solids to the surface. This method requires pressure pumps and
containers and necessitates the recirculation of a portion of the flow,
however, greater efficiency is usually obtained than with the "whipping"
method. Three papers (11, 43, 50) describe in detail pilot-scale tests of
the flow pressurization method of flotation clarification of fish processing
wastes. Water was pressurized by a centrifugal pump at about 40 psig. Air
was added at the rate of about 2 percent by volume. The pressurization
tank had a one-minute detention time. The recovered sludge was heated and
then the protein and oil fractions were removed by centrifugation. All
tests were conducted using coagulant aids. The specific aids tested were
alum, ferric chloride, F-Flok, aluminum hydroxide, Zetol-A (trade name for
an animal glue), and lime. In the first tests on salmon processing waste-
213
-------
waters, alum, ferric chloride, and F-Flok were compared. Alum and
ferric chloride performed well as coagulants but large doses of F-Flok
were necessary to achieve comparable results. The iron in the ferric
chloride tended, however, to catalyze lipid oxidation and this coagulant
aid was thus judged unsatisfactory. Further tests were conducted and the
authors concluded that flotation cells could be used effectively on fish
processing wastes. Alum treatment was judged the most promising of the
methods used. Feeding tests showed that alum could be included in the
recovered solids up to the 1 percent level and the resultant sludge fed
to chickens without altering their growth rate.
A method has been developed (51) to remove fish oils down to the 0.008
percent level by acidifying the waste stream followed by flotation. This
method requires neutralization after treatment. Specially-coated treat-
ment equipment is needed to reduce corrosion.
Little work has been done on biological treatment of seafoods processing
wastes. Buczowska and Dabaska (48) did report that the carbon-to-nitrogen
ratio of fish processing wastewaters is satisfactory for biological
treatment. The biochemical oxidation rate was said to be similar to
sewage, but nitrification begins sooner and is more significant. Assuming
primary stage removal of reasonable levels of solids, greases and oils,
the authors concluded that no special treatment problems should be
encountered. Without this pretreatment, several problems can develop. For
example, the oil and grease can interfere with oxygen transfer in an
activated sludge system (52). Czapik (53) reported on a trickling filter
that clogged due to high levels of solids and oil from a fish processing
plant.
Anaerobic treatment of fish wastes has been investigated (9). Liquid
fish wastes were judged to produce no unusual problems in a digester
operation when in-plant screening was employed. Matucky, et al. (52)
stated that fish solids and oils digested readily and the resultant
sludge dewatered easily. The digester loading rates used by these
investigators varied from 0.1 to 0.35 pounds of volatile solids per cubic
foot per day.
CURRENT RESEARCH
A variety of research projects on subjects relating to fish processing
wastes are presently in progress or recently completed. These projects
are briefly summarized below to describe the general trends in research
efforts and to indicate specific individuals who can provide up-to-date
information.
214
-------
Harvesting and Processing Modifications
The Oregon State University Seafoods Laboratory (54) is presently
studying the efficiency of the Yaaagiya Flesh Separator. The device
consists of a revolving stainless steel drum perforated with numerous
1/16" diameter holes. A continuous belt is held against a portion of
the drum. Whole fish or filleted fish bodies are placed between the
revolving drum and the belt and the soft flesh portions are continuously
pressed through the drum and thereby extracted. Present data show
excellent recovery of the flesh portions from the bone structures of
the fish. A larger model of this device is presently being used by one
Oregon processor to remove cooked tuna flesh from bone scraps. The
recovered flesh is processed as dog and cat food.
Richardson and Amundson (55) have undertaken a 5-year study of rendering
processing of Great Lake alewives. Microbial activity is used to
separate the oil and scrap. Proposed uses are fish protein concentrates,
fish oils and various oil-based products.
The College of Fisheries of the University of Washington (56) has
concluded research on the enzyme digestion of shrimp wastes. An effort
was made to develop an active digestive system that could operate at
high temperatures.
Law (57) currently has a U.S.D.A. grant to study the utilization of
marine waste products.
A new rapid method of ship-board fish meal production has been developed
by a Mexican firm (58) . Fresh fish are ground and dried simultaneously
In a 240°C gas stream. The meal is then cooled and packaged. The
complete process takes from 6 to 8 seconds as compared to 22 minutes in
a;n alcohol extraction process. One ton of fish meal is recovered from
five tons of fish. No reference was made to the applicability of this
method to fish wastes, but there is no obvious reason to discount it
as a possibility.
A packaged on-board freezer has been recently marketed by a Pennsylvania
firm (59). This unit freezes up to 300 pounds of shrimp per hour and
maintains freezing temperatures in the storage hold. Utilization of this
apparatus could eliminate the use of ice and its resultant wastewater.
Two new American ocean vessels have recently been active in the harvesting
and processing of fishery products (60). Named the Seafreeze Atlantic
and the Seafreeze Pacific, these two ships cost over $5 million each, and
each can handle 50 tons of fish per day. Processing is so complete that
''only the skins are wasted". If this venture proves successful,
terrestrial accumulation of fish processing wastes could be substantially
reduced in the future.
215
-------
The Bureau of Commercial Fisheries (61) has developed a trap to
harvest the sablefish population off the Pacific Coast. The trap
has been judged to be moderately successful and further development
is planned.
Waste Strengths and Volumes
The National Canners Association (62) is presently conducting research
en wastewater characteristics from sardines, shrimp, salmon, and tuna
processing plants. The wastewater parameters to be measured are COD,
BOD5, total solids, dissolved solids, suspended solids, oil, grease,
nitrogen and chlorides. A study on Maine sardine plants has been
completed.
Waste Treatment
A Northern California firm (30) has developed a direct-fired gas drier
to economically dry fish meal. The drier jet exhausts upward with an
adequate velocity to "fluidize" the drying bed of meal. High heat
efficiencies have been obtained with this machine (i.e., greater than
95 percent recovery).
Kempe (63) has proposed research on the efficiency of spray-evaporation
of stickwater. This method is considered to be superior to other
evaporation methods due to lower cost, simplicity of operation, and
faster start-up. These factors are especially important to the smaller
rendering plants with limited capital.
Johnson and Hayes (64) have proposed a pilot plant study of the
utilization of the chitin fraction of king crab wastes. Matthews (64) ,
of the University of Alaska, is presently studying the utilization of
king crab wastes.
Deyoe (65), at Kansas State University, has proposed research on the
nutritive value and economic utilization of catfish processing wastes.
Meals produced by various methods would be chemically analysed and
animal feeding tests performed.
RECOMMENDATIONS FOR FURTHER RESEARCH
Factors Influencing Research Recommendations
The survey of the seafood industry demonstrated that the water pollution
problems generated therein are, with a few isolated exceptions, not as
critical as those of many other industries. There are two basic reasons
for this conclusion. First, seafood processing plants generally dispose
of their wastes into estuaries, which results in dispersion and dilution.
Soluble pollutant loads are quickly reduced in well-mixed estuarine
environments. Secondly-, in many cases the processing plants are located
216
-------
in sparsely populated areas where other industrial wastes and domestic
wastes are of limited magnitudes, minimizing the competition for the
assimilative capacities of the estuaries. This is not to say that
seafoods processing pollution problems can be justifiably ignored;
rather that it is advisable to attack the problems systematically and
develop reasonable solutions in a rational manner.
In the opinion of the authors, based on personal experience and the
consensus expressed in the literature, seafoods wastewaters are
readily amenable to biological treatment and should present no special
difficulties from the standpoint of toxicity. The problem with the
wastewaters, therefore, remains basically one of economics, not of
technology.
Economics also seems to be the major concern in the disposal of solid
wastes. Solid wastes, unlike liquid wastes, are of potentially
significant economic value in the form of by-products. This potential
should be recognized and utilized wherever possible in future research
and development efforts.
The seafoods industry consists of a myriad of processing centers located
along our coastlines. The plants are frequently autonomous, intensely
competitive, and notably lacking in cooperative spirit. Common problems
are seldom handled jointly. Organizations such as the National Canners
Association, National Fisheries Institute, Pacific Fisheries Technolog-
ists and others are striving to reverse this trend, but without much
success to date. One outstanding exception to this pattern is the current
cooperative effort being mounted by the crab processors of Kodiak, Alaska.
This undertaking involves the common collection of solid wastes followed
by disposal at a single sanitary landfill. Hopefully, this activity
is indicative of a developing awareness within the industry of the
advantages of attacking in concert the water pollution problems common
to all.
The lack of geographic concentration of the industry will tend to
influence the type of research undertaken. Solutions which rely on
combining the effluents (or solid wastes) from several plants or, the
effluent from a single plant with that of a sizeable municipality, will
not always be appropriate. Many of the major offenders are remotely
located, with few, if any, other industries near at hand, and with only
a handful of nearby residents who are most likely employed by the cannery.
This situation, of course, is not usually the case, but, nonetheless, is
common enough to warrant consideration.
ITie diversity of the industry is an added factor which must be considered
when planning waste utilization and treatment research. Unlike some of
the single-commodity food processing industries, the seafoods processors
produce wastes which, while all highly organic and nitrogen-rich
(excluding cooling waters), vary from negligible to staggering volumes.
217
-------
Funding alternatives, both for research and for the ultimate full-
scale utilization of the research findings, are an especially important
consideration in this case. The industry, as most of the food industries,
is typically a low profit enterprise. This fact, compounded by the current
stationary production posture and increasing pressures from foreign
competitors has already forced many small plants to discontinue operation.
Significant increases in expenses, whether for in-house research or water
pollution control, are likely to be untenable to many of the remaining
smaller processors. Public research and demonstration project funding
and treatment facility subsidies (in the form of tax credits or similar
arrangements) will probably be necessary to permit the industry to
survive in its present form.
Recommendations
It appears that the most urgent research needs in the seafoods field lies
in three main areas: (a) demonstration scale (solid and aqueous) waste
treatment and/or disposal projects; (b) development of in-plant processing
modifications (including by-product development) designed to reduce
wastage; and (c) production-Bcale projects demonstrating the technical
and economic feasibility of the solutions developed in (b).
To provide the bases for abating the most critical problems now facing
a few plants, item (a) should be given highest priority. To make
allowance for longer term considerations, items (b) and (c) should also
be afforded increased emphasis. In many cases, the basic work out-
lined in item (b) has already been accomplished (or nearly so), and
it only remains to demonstrate the new approach. Work with flesh
separators, for instance, has indicated, at least in a preliminary
fashion (54), that significant solids recoveries can be realized. Perhaps
full-scale demonstration of this concept should be encouraged.
Similarly, the previously-mentioned high-speed shipboard meal plant (57)
could be applied to fish waste utilization and its effectiveness evaluated.
Floating canneries are just now being placed into operation (59). Their
advantages from the standpoint of wastes reduction should not be over-
looked as their performances are studied.
Waste Treatment and Disposal
Before demonstration-scale projects can be intelligently developed the
designer must be familiar with the characteristics of the wastewaters
with which he is dealing. Definitive studies of seafoods processing
wastes are scarce, especially for shrimp and crab processing. Further
work in this area should be supported.
218
-------
Most states' water quality standards are phrased in terms such as
"....water quality shall not be impaired to the detriment of
legitimate existing or foreseen water uses...." Since a treatment
facility design is based on anticipated efficiencies (in terms of
BOD5 and suspended solids), the level of treatment must be pre-defined.
This requires a thorough knowledge of the effects of the wastes on the
aquatic environment, therefore, studies of the effects of seafood plant
wastes on marine and estuarine environments should be conducted. In-
depth investigations of dissolved oxygen depletion, temperature effects,
benthic disturbances, tidal effects, effects on primary and secondary
productivity, effects of highly variable and shock loadings, degree of
and rate of off-season recovery and many other variables should also
be conducted.
The applicability of standard treatment methods is generally well
accepted, but has not been sufficiently demonstrated; nor have the
optimum operational characteristics been defined for each major type
of primary and secondary process. This should be done at full
(demonstration) scale for sedimentation, flotation, biological
filtration, perhaps activated sludge and ultimately aerobic and
anaerobic digestion.
Joint municipal-industrial waste treatment should be utilized whenever
practical, since the same advantages inherent in joint treatment of
other industrial wastes also apply here: dilution, equilization, the
economics of size, etc.
Innovative techniques and new treatment methods, while not critical
to the immediate solution of the problem, should, nonethelesss
be encouraged.
It is the opinion of the authors that solid waste should be considered
a resource rather than useless refuse. Direct disposal, either in a
landfill or at sea, should only be considered as a last resort, but in
instances where deep sea disposal is the only acceptable alternative,
perhaps investigations of methods, economics and consequences of deep
sea disposal should be carried out.
Another alternative is to avoid concentrating the waste at the plant;
i.e., pre-process at sea. Placement of an "iron chink" in a "mother-
ship" which would accompany the salmon fishing fleet is one example.
The wastes would be returned immediately to the fishing grounds. The
effects of this practice would, of course, require attention.
219
-------
By-Product Development
The manufacture of by-products from solid residues has been extensively
researched, especially with regard to salmon. An evaluation of these
methods as potential waste reduction techniques leads to the conclusion
that only those which utilize all or most of the solids are helpful.
Animal feeding and other whole-waste utilization methods should be
stressed.
Perhaps more basic by-product development work is needed in the crab
and shrimp industries, but, in general, the economic aspects of the
operations should be emphasized. Market surveys are needed and
transportation alternatives evaluated to determine the economic
feasibilities of various approaches.
The seafoods industry has a problem: water pollution. It has not
generally reached crisis proportions yet. It does, however, require
immediate, but rational attention. The various avenues of approach
should be investigated for each general geographical area and each
commodity subsection and the best combination of solutions delineated
and implemented in each instance.
220
-------
REFERENCES
1. Soderquist, M. R. , K. J. Williamson, G. I. Blanton, D. C. Phillips,
D. L. Crawford, and D. K. Law. 1970. Water Pollution Abatement
in the U.S. Seafoods Industry. Pacific Northwest Water Laboratory,
F.W.P.C.A., U.S.D.I., Corvallis, Oregon (in press).
2. Holt, S. J. 1969. The Food Resources of the Ocean. Scientific
American. 221, 3, 178-194.
3. . 1961. Report of the International Conference on Fish
in Nutrition. FAO Fisheries Report No. 1. Washington, D.C. 91 pp.
4. Wilcke, H. L. 1969. Potential of Animal, Fish, and Certain Plant
Protein Sources. Journal of Dairy Science, 52, 409-418.
5. Lyles, C. H. 1969. Fisheries of the United States...1968. C.F.S.
No. 5000, Bureau of Commercial Fisheries, U.S. Fish and Wildlife
Service, U.S.D.I. Washington, D.C. 83 pp.
6. Borgstrom, G. and C. D. Paris. 1965. The Regional Development of
Fisheries and Fish Processing. In: Fish as Food, Volume III
(G. Borgstrom, ed.) Academic Press, New York. 301-409 pp.
7. . 1969. Computer Projects 73% Frozen Food Poundage Rise
During Next Decade. Quick Frozen Foods, 31, 11, 45-46.
8. Clark and Groff Engineers. 1967. Evaluation of Seafood Processing
Plants in South Central Alaska. Contract No. EH-EDA-2. Branch of
Environmental Health, Division of Public Health, Department of
Health and Welfare, State of Alaska, Juneau, Alaska, 28 pp.
9. Nunnalee, D., and B. Mar. 1969. A Quantitative Compilation of
Industrial and Commercial Wastes in the State of Washington. 1967-69
Research. Project Study 3F, State of Washington Water Research
Center. Washington State Water Pollution Control Commission. Olympia,
Washington. 126 pp.
10. Davis, H. C. 1944. Disposal of Liquid Waste from Fish Canneries,
from the Viewpoint of the Fishing Industry. Sewage Works Journal,
16^ 947-948.
11. Claggett, F. G. and J. Wong. 1968. Salmon Canning Waste Water
Clarification. Part I. Flotation by Total Flow Pressurization
Circular No. 38. Fisheries Research Board of Canada. Vancouver,
B.C., Canada. 9 pp.
221
-------
12. Paessler, A. H. and R. V. Davis. 1956. Waste Waters from
Menhaden Fish Oil and Meal Processing Plants. Proceedings, llth
Industrial Waste Conference, Purdue University, Engineering
Extension Series No. 91, 371-388.
13. Kempe, L. L., N. E. Lake and R. C. Scherr. 1968. Disposal of
Fish Processing Wastes, Final Report, ORA Project 01431. Department
of Chemical and Metallurgical Engineering, University of Michigan,
Ann Arbor, Michigan. 46 pp.
14. . 1963. The U.S. Tuna Industry. Good Packaging, 24,
7, 56-76.
15. Dewberry, E. B. 1969. Tuna Canning in the United States. Food
Trade Review, 39. 11, 37-42
16. . 1970. Tuna and Tuna-Like Fish Received by California
Canneries, 1969. Western Packing News Service, 65, 9, 5.
17. . 1969. A Survey of Waste Disposal Methods for the Fish
Processing Plants on the Oregon Coast. Oregon State Sanitary
Authority. Portland, Oregon. 77 pp.
18. Chun, M. J., R.H. F. Young, and N. C. Burbank, Jr. 1968. A
Characterization of Tuna Packing Waste. Proceedings, 23rd
Industrial Waste Conference, Purdue University Engineering Extension
Series No. 132, 786-805.
19. Magnusson, H. W. and W. H. Hagevig. 1950. Salmon Cannery
Trimmings, Part I. Relative Amounts of Separated Parts.
Commercial Fisheries Review, 12, 9, 9-12.
20. Brody, J. 1965. Fishery By-Products Technology. AVI Publishing
Co., Westport, Conn. 232 pp.
21. Jensen, C. L. 1965. Industrial Wastes from Seafood Plants in the
State of Alaska. Proceedings, 20th Industrial Waste Conf., Purdue
University Engineering Extension Series No. 118, 329-350.
22. . 1944. Operations Involved in Canning, Fisheries Leaflet
No. 79. In: Research Report No. 7. Bureau of Commercial Fisheries,
Fish and Wildlife Service, U.S.D.I., Washington, D.C. 93-111 pp.
23. Jones, G. I. and E. J. Carrigan. 1953. Possibility of Development
of New Products from Salmon Cannery Waste: Literature Survey. In:
Utilization of Alaskan Salmon Cannery Waste, Special Scientific .Report:
Fisheries No. 109 (M.E. Stansby, e_t al., ed.). Fish and Wildlife
Service, U.S.D.I., Washington, D.C. 7-35 pp.
222
-------
24. . 1966. Industrial Waste Survey for Port of Bellingham
(unpublished). Stevens, Thompson, Runyan and Ries, Inc. 38 pp.
25. Foess, Jerry. 1969. Industrial and Domestic Waste Testing Program
for the City of Bellingham (unpublished). Cornell, Rowland, Hayes
and Merryfield, Inc. Seattle, Washington. 17 pp.
26. Dewberry, E. B. 1964. How Shrimps are Canned at a New Orleans
Factory. Food Manufacture, 39, 7, 35-39.
27. Idyll, C. P. 1963. The Shrimp Fishery. In: Industrial Fishery
Technology (M.E. Stansby, ed.). Reinhold Publishing Corp.,
New York. 160-182 pp.
28. Alverson, D. L. 1968. Fishery Resources in the Northeastern Pacific
Ocean. In: The Future of the Fishing Industry of the United States
(D. Gilbert, ed.), Publications in Fisheries, New Series, Volume IV.
University of Washington, Seattle, Washington. 86-101 pp.
29. . 1968. Preservation and Processing of Fish and Shell-
Fish - Quarterly Progress Report. Technological Laboratory, Bureau
of Commercial Fisheries, Fish and Wildlife Service, U.S.D.I.
Ketchikan, Alaska 2 pp.
30. Grimes, E. 1969. personal communication.
31. Crawford, D. L. 1968. personal communication.
32. Stansby, M. E. 1963. Processing of Seafoods. In: Food Processing
Operations (M.A. Joslyn and J.L. Heid, eds.). AVI Publishing Co.,
Westport, Conn. 569-599 pp.
33. Dassow, J. A. 1963. The Crab and Lobster Fisheries. In: Industrial
Fisheries Technology (M.E. Stansby, ed.). Reinhold Publishing Corp.,
New York. 193-208 pp.
34. Crawford, D. L. 1969. personal communication.
35. . 1970. Extracts Crabmeat Automatically. FoodL Engineering,
42., 2, 36.
36. Slavin, J. W. and J. A. Peters. 1965. Fish and Fish Products, In:
Industrial Wastewater Control (C.F. Gurnham). Academic Press.
New York. 55-67 pp.
37. Dassow, John A. 1963. The Halibut Fisheries. In: Industrial
Fishery Technology (M.E. Stansby, ed.). Reinhold Publishing Corp.
New York. 120-130 pp.
223
-------
38. Thurston, C. E. 1961. Proximate Composition of Nine Species of
Sole and Flounder. Agricultural and Food Chemistry, 9^, 313-316.
39. Limprich, H. 1966. Abwasserprobleme der Fischindustrie [Problems
Arising from Waste Waters from the Fish Industry]. IWL-Forum
66/IV. 36 pp.
40. . 1969. Industrial Fishery Products, 1968 Annual Summary.
C.F.S. No. 4950. Bureau of Commercial Fisheries, U.S. Fish and
Wildlife Service, U.S.D.I. 9pp.
41. Jones, W. G. 1960. Fishery Resources for Animal Food. Fishery
Leaflet 501. Bureau of Commercial Fisheries, Fiah and Wildlife
Service, U.S.D.I. 22 pp.
42. ^opkins, E. S. and J. Einarsson. 1961. Water Supply and Waste
Disposal at a Food Processing Plant. Industrial Water and Wastes,
6., 152-154.
43. Claggett, F. G. and J. Wong. 1969. Salmon Canning Waste-Water
Clarification. Part II. A Comparison of Various Arrangements for
Flotation and Some Observations Concerning Sedimentation and
Herring Pump Water Clarification. Circular No. 42. Fisheries
Research Board of Canada. Vancouver, B.C., Canada. 25 pp.
44. Jaegers, K. and J. Haschke. 1956. [Waste Waters from the Fish
Processing Industry]. Wasserw.-Wassertech. j6_, 311.
45. Knowlton, W. T. 1945. Effects of Industrial Wastes from Fish
Canneries on Sewage Treatment Plants. Sewage Works Journal, 17,
514-515.
46. Schulz, G. 1956 [Purification of Waste Waters from Fish Ponds].
Wasserwirtsch. Wassertech., 6^, 314-316.
47. Griffen, A. E. 1950. Treatment with Chlorine of Industrial Wastes.
Engineering Contract Rec., 63, 11, 74-80.
48. Buczowska, Z., and J. Dabaska. 1956. [Characteristics of the Wastes
of the Fish Industry]. Byul. Inst. Med. Morsk. Gdansk. 7, 204-212.
49. Dreosti, G. M. 1967. Fish Solids from Factory Effluents. Fishing
News International, (>_, 11, 53, 54.
50. Claggett, F. G. 1967. Clarification of Waste Water Other than
Stickwater from British Columbia Fishing Plants. Technical Report
No. 14. Fisheries Research Board of Canada. Vancouver, B.C.,
Canada. 8 pp.
224
-------
51. Aktieselskabet, Aminodan. 1968. Recovery of Oil and Protein from
Waste Water. British Patent #1,098,716. 4 pp.
52. Matusky, F. E., J. P. Lawler, T. P. Quirk and E. J. Genetelli. 1965.
Preliminary Process Design and Treatability Studies of Fish Processing
Wastes. Proceedings, 2Qth Industrial Waste Conference, Purdue
University Engineering Extension Series No. 118, 60-74.
53. Czapik, A. 1961. Fauna of the Experimental Sewage Works in
Krakow. Acta Hydrobiol., _3_, 63-67.
54. Law, D. K. 1969. personal communication.
55. . 1970. Converting the Alewife. The Sciences, 10, 2, 35,36.
56. Listen, J. 1969. personal communication.
57. Law, D. K. 1969. personal communication.
58. Lopez, J. L. G. 1967. [Fish Meal Manufacturing Machines on Board
Shrimpers - Economic Study]. Report No. 35, Mexican Fisheries
Bureau. 11 pp.
59. . 1969. Lightweight Packaged On-Board Freezer for Shrimp
Trawlers Now on Market. Quick Frozen Foods, 32, 4, 93.
60. . 1969. U.S. Tests Fish Factory Vessels. Canner/Packer.
138, 2, 12.
61. . 1969. Sablefish Off West Coast Sought as Resource for
Frozen Packers. Quick Frozen Foods, 32, 5, 107, 108.
62. Sternberg, R., and G. Brauner. 1969. personal communication.
63. Kempe, L. L. 1969. Spray Evaporation of Stickwater from Fish
Rendering (unpublished). Department of Chemical Engineering, University
of Michigan, Ann Arbor, Michigan. 6 pp.
64. Simon, R. 1969. personal communication.
65. Deyoe, C. 1969. The Nutritive Value and Economic Utilization of
Catfish Processing Waste in Animal and Fish Diets (unpublished).
The Food and Feed Grain Institue, Kansas State University, Manhatten,
Kansas. 4 pp.
225
-------
CANNERY WASTE TREATMENT BY A
HIGH SOLIDS ACTIVATED SLUDGE PROCESS
by
Warren G. Palmer*
INTRODUCTION
The activated sludge process is used for treating a variety of
industrial wastes; however, these wastes often cause problems because of
the seasonal nature of their shock loading effect on a conventional
treatment plant by way of sudden high strengths or flows. Examples
of such wastes are cannery, meat packing or processing, dairy, paper
mill, and textile plant effluents. Adequate processing is needed to
overcome such treatment difficulties.
An adaptation of the activated sludge process that appears to be
capable of processing high strength wastes has been studied in Europe
and uses high concentrations of mixed liquor suspended solids with
little sludge wasting. This adaptation, which was a parallel
development to extended aeration, is the Totalklaranlage developed by
Professor Dietrich Kehr of the Technishe Hochschule in Hanover, Germany
(2) and consists of an activated sludge with a concentration of mixed
liquor suspended solids of 10,000 and 14,000 mg/liter and an aeration
time of 6 hours. Kehr has reported high removals of BODj, nitrogen,
and phosphorus with a loading of about 94 pounds of BOD5/1000 ft' of
aeration tank. The high solids concentration of the process when
coupled with physical conditions capable of maintaining a proper
environment comprises an inherent buffering capability for high strength
waste.
The basic objectives of the investigation reported herein are to
demonstrate reliable performance of an adaptation of the activated sludge
process using high concentrations of mixed liquor solids. The FMC
Corporation has termed this the "Kehr Activated Sludge Process" (KASP).
This system is to treat both domestic sewage and cannery wastes having
BOD5 strengths of 200 to 2000 mg/liter.
This study was financed under the United States Department of the Interior
FWPCA Research and Development Grant 12060 EZP. The work was performed
at FMC Corporation's Central Engineering Laboratories (GEL) in Santa Clara,
California.
* FMC Corporation, 1185 Coleman Ave., Santa Clara, California 95052,
22?
-------
This report describes the tests conducted to determine the effectiveness
of the KASP for treating domestic sewage and cannery wastes.
THEORY
The activated sludge type of treatment plant has been widely used for
stabilizing organic materials in both industrial wastes and domestic
sewage. The microorganisms present in activated sludge floe accomplish
stabilization of the organic materials by the mechanism shown in Figure 1.
As illustrated, the decomposable organic substrate is used as food by the
microorganisms to derive new cell mass by synthesis and energy for their
metabolic functions by oxidation of the organic substrate to C02 and water.
As well as using sewage organic wastes for food, the microorganisms
ut ^.ize the degradable portion of the cell mass that is released on death
of the cell. However, a portion of the cell mass is resistant to biological
attack and cannot be readily metabolized. The final products of complete
biological stabilization are then CC>2, water, and non-biodegradable
organic matter.
The growth of microorganisms during the batch stabilization of an organic
waste can be characterized by four phases. First a "lag phase" occurs
where the microorganism population adjusts to the new environment. This
is followed by the "log-growth phase" where there is sufficient food for
unrestricted growth of the acclimated organisms. During this period of
"log growth" the rate of BOD removal and cell growth is proportional to
the number of organisms present. As the food concentration decreases it
begins to limit the rates of cell growth and BOD removal. In this
"declining growth phase" the rates of cell growth and BOD removal are
proportional to the food present. Finally, "the endogenous phase" begins.
In this phase there is insufficient food present to support cell growth
at a rate greater than loss from death and lysing and active cell mass
decreases. The degradable components of the lysing cells are metabolized
by the remaining population.
Within an activated sludge system the organisms and waste present are
quite heterogeneous causing these phases in actual operation systems to
be the composite result of the reactions of the various organism- populations
with the food components. Most activated sludge systems operate in the
"declining growth phase."
One mathematical model used to describe the rates of BOD removal and of
sludge growth in the "declining growth phase" is as follows: (3)
Sludge Growth Rate - ~ = ksF (1)
BOD Removal Rate « dF_ = kRF (2)
dt
228
-------
Oxidation
Degradable
Organic
Waste
t-o
S3
Micro-Organisms
0,
Energy
New
Cellular
Material
Degradable
Organic
Material
Death and Lysing
Inert
Organic
Residue
Figure 1. MECHANISM OF MICROBIAL STABILIZATION OF AN ORGANIC WASTE
-------
Where:
M = Active sludge mass concentration.
F = Total food or ultimate BOD concentration.
kg = Rate constant for sludge growth in the declining growth phase.
kjj = Rate constant for BOD removal in the declining growth phase.
The active portion of the activated sludge floe is that portion of the
sludge mass that enters into the dynamics of the system and it comprises
only a fraction of the total sludge mass . Volatile suspended solids have
often been used as a measure of the active mass. This procedure yields
a much greater value than the real active mass because the 600 °C ignition
temperature used in the procedure volatilizes more of the total mass than
just the active portion. For this report the active mass will be assumed
to be some constant fraction of the volatile suspended solids.
The loss of cellular mass by endogenous respiration occurs during all
the growth phases and is proportional to the active mass present. During
the log-growth phase the loss by endogenous respiration is masked by the
rapid growth and only during the declining growth and endogenous phases
does endogenous respiration become important. The rate of cellular loss
by endogenous respiration can be expressed mathematically as :
- M (3)
Where:
k = Rate constant for endogenous respiration.
The use of the previous equations to describe an activated sludge system
is greatly simplified by using a completely mixed aeration tank. The
contents of a completely mixed aeration tank are homogeneous so that the
description of one portion of the tank describes the entire tank contents
and the tank effluent. Smith (3) reported that all measurable BOD
reduction in a completely mixed activated sludge plant takes place in the
aeration tank. This further simplifies the use of the previous equations
because it means that the composition of the solution phase of the plant
effluent is essentially the same as that throughout the aeration tank.
The total BOD of the plant effluent, being both solution and solid phase
is affected by the efficiency of the solids separation process.
The solution phase BOD in effluent of a completely mixed activated sludge
system operated in the "declining growth phase" is given as (3) :
(4)
230
-------
Where:
kj, = Rate constant for BOD removal in the declining growth phase.
F = Concentration of solution phase BOD in the effluent.
F0 = Concentration of BOD in the feed.
t = Aeration time of the sewage alone.
A material balance for active cellular mass can be made around an
activated sludge plant. Such a balance sums cellular synthesis and
cellular loss by endogenous respiration in the aeration tank as well
as the amounts carried into and out of the system in the various
liquid streams. Figure 2 shows a material balance for active cell mass
around an activated sludge plant.
In a conventional activated sludge plant, part of the influent organic
waste is disposed of as waste activated sludge solids which are composed
of new cellular material and non-biodegradable matter. Because this
waste activated sludge is difficult to handle, the extended aeration
type of treatment plant was developed. Also known as the complete
oxidation type of treatment plant, extended aeration attempts to obtain
complete stabilization of the organic matter as illustrated in Figure 1.
Extended aeration produced aerobic digestion within the aeration tank
rather than in a separate digester. Long aeration periods are used and
the process attempts to operate such that the rate of synthesis of cellular
material is equal to the loss by endogenous respiration. If the amount
of cellular material flowing into and out of the treatment plant in the
sewage and effluent are ignored, and no sludge is wasted, the material
balance presented in Figure 2 can be rewritten as follows:
= ksF - keMa (5)
Where:
Ma = Concentration of active mass in aeration tank.
When extended aeration operates with no net cellular growth, —3— is equal
to zero. For such a system equation 5 yields the following relationship
of the concentration of active cellular mass to effluent BOD.
231
-------
Accumulation = In - Out + Synthesis-Endogenous Loss
dt
k FV - k M V
s a e a a
K3
bJ
to
Sewage
Aeration Tank
va
a
Settling
Tank
Q = Flow
M = Concentration of Active Mass
V = Volume
Subscripts
I = Influent
a = Aeration Tank
E = Effluent
W = Waste Sludge
Effluent
Waste
Sludge
Figure 2. MATERIAL BALANCE FOR ACTIVE MASS AROUND AN ACTIVATED SLUDGE TREATMENT PLANT
-------
Equation 6 shows that the steady state concentration of active mass
is a function of the effluent BOD. Equation 6 may be combined with
equation 4 to express the steady state active mass concentration in
terms of the influent BOD and aeration time.
Traditionally extended aeration plants have used long aeration periods
resulting in low effluent BOD and a correspondingly low steady state
active mass concentration. The KASP as practiced at FMC Corporation's
Central Engineering Laboratories (CEL) operates on the principal of
steady state concentration active mass but at shorter aeration periods
with subsequently higher concentrations of effluent BOD and mixed liquor
active mass. It was attempted to operate the system such that each
aeration period at the existing sewage strength produced a steady state
concentration of suspended solids in the aeration tank of about
10,000 mg/liter. Fluctuations in sewage strength prevented the aeration
period to be matched exactly to the sewage so that the concentration of
mixed liquor suspended solids varied from about 5,000 to 12,000 mg/liter.
As shown in Figure 1, the microbiological stabilization of the organic
waste produces non-biodegradable sludge causing an increase in the
total sludge mass. In any system such as the KASP with no intentional
sludge wasting the production of the non-biodegradable solids will
cause the sludge mass to increase to where the solids liquid separation
systems capacity is overtaxed and solids will wash out in the effluent
to prevent an unlimited increase in the mixed liquor solids. Both inert
organic solids and active biological solids will be carried out in the
effluent. Thus in a working system there will be uncontrolled wasting
of active mass in the plant effluent, and the active mass in the mixed
liquor will be less than that for true steady state operation.
TEST FACILITY AND EQUIPMENT
Permanent Facility - The permanent facility at FMC's Central Engineering
Laboratories (CEL) Environmental Engineering site includes a 12" diameter
inlet sewer from an 18" diameter sanitary sewer serving part of the City
of Santa Clara. An adjustable hinged gate has been provided in the 18"
line to divert sewage into the test site. A pumping station which receives
the sewage from the 12" line is provided with two parallel open channels
in which various forms of screening and comminution are available. Any
one or all of the two channels and the 12" line may be closed off and
emptied if desired. At the present time a standard Chicago Pump comminutor
serves one channel and a production prototype Barminutor machine the other.
233
-------
All sewage, whether directly from the sewer, screened, or comminuted,
flows into a wet well located beneath the channels and comminuting
equipment. In an adjacent dry well are two centrifugal pumps for
delivering the sewage from the wet well through a six inch pressure
line to an elevated flow splitter located in an outdoor test area.
Valving in the dry well permits direct pumping back to the city sewer
downstream from the point from which it was withdrawn. Maximum
delivery to the elevated splitter is 750 gpm with both pumps running.
A portion of the total flow to the splitter is diverted to a settling
cone where grit and a portion of the suspended solids are removed. The
overflow from this settling tank flows to another splitter box where it
can be directed to sites throughout the test facility. A schematic
diagram illustrating the CEL facilities is shown in Figure 3.
Experimental Facilities - A schematic drawing of the activated sludge
system used for the KASP study is shown in Figure 4. The primary
components of the system consisted of an aeration tank with a liquid
volume of about 120 gallons and a circular sedimentation tank four feet
in diameter.
Mixed liquor was pumped from the aeration tank to the final clarifier
by a positive displacement pump driven by a variable speed motor. The
flow rate was controlled by the speed of the motor. The return sludge
was also pumped by a positive displacement pump and the flow rate was
set equal to the settled sewage flow rate by a timer controlling the
percentage of the time the pump was on. A float switch controlled the
settled sewage flow.
A diurnal flow variation was produced by a 24-hour program timer, which
controlled the power to the mixed liquor and return sludge pumps for
each 5-minute interval during the day. A flow pattern similar to that
at the San Jose - Santa Clara Water Pollution Control Plant could be
produced by controlling the number of 5-minute cycles during each hour
that the pumps were on and off.
Oxygen was supplied by a blower which pumped air through 4 carborundum
diffusers 3 inches 0. D. and 24 inches long. The air flow was varied
from 16 CFM to 30 CFM to maintain the concentration of dissolved oxygen
in the mixed liquor greater than 2 mg/liter.
A schematic flow diagram showing the arrangement of the experimental
system is also presented in Figure 3.
EXPERIMENTAL PROCEDURE
Continuous operation of the KASP was performed by adjusting the flow rate
of the return sludge to one half of the flow rate of mixed liquor to the
final tank and the sewage flow, which was controlled by a float switch,
made up the balance. Since no sludge was intentionally wasted, the only
234
-------
d) 18" Santa Clara Sewer
(g) 12" Supply Sewer
Open Channels (Gates - Screens)
Barminutor Machine
Conuninutor Machine
Sewage Pumps
Flow Splitter (Elevated)
i) Settling Tank
oj Splitter Box (Elevated)
jj) Storage
l) Aeration Tank
2) Final Tank
Tl Electroflotation Cell
KF.HR Prnrpga F.ffliipnt-
0-E
G>
Figure 3. FLOW DIAGRAM, PERMANENT AND EXPERIMENTAL FACILITY
0
-------
Settled
Sewage
(feed)
Settling Tank
Aeration Tank
Figure 4. KEHR ACTIVATED SLUDGE PLANT
-------
other control required was to maintain the flow of air in sufficient
quantity to maintain the mixed liquor dissolved oxygen above 2 mg/liter.
Both the sewage and effluent were sampled every hour during the 8:00 AM
to 5:00 PM working day for an eight-hour composite. The mixed liquor
and return sludge were sampled on a grab basis.
A programmed study was conducted to determine the performance of the high
solids activated sludge system under shock loading conditions. This was
designed to simulate shocking caused by a sudden decrease followed by a
sudden increase in sewage strength. The underloading shock was simulated
by changing the feed to the aeration tank from sewage to fresh water.
After 48 hours the feed to the aeration tank was changed back to sewage
so that an overloading shock occurred. The system was operated at a
constant feed flow during this study. The constant feed was begun for
one week prior to the test to insure that a steady state condition
existed at the start of the test. Grab samples were taken of mixed liquor,
return sludge, final effluent, and the sewage feed.
To determine if the KASP aeration tank was completely mixed, a tracer
washout study was performed by first filling the KASP aeration tank with
fresh water. Then a concentrated sodium chloride solution, simulating
both sewage and return sludge, was pumped into the aeration tank. The
effluent was periodically sampled and the concentration of the chloride
was determined as a function of time. The course of the chloride
concentration in the effluent from a completely mixed tank can be
expressed mathematically by the following equation:
log °t - CE
8
Where: Cp = Feed concentration.
Cg = Effluent coneentration.
Cj = Initial concentration in the system.
D = Tank displacement per unit time.
RESULTS
Tracer Washout - The degree of complete mixing of the aeration tank was
determined by adding a sodium chloride solution (3400 mg/liter as Cl~) to
the aeration tank at a flow rate of 2.0 gpm, which resulted in a liquid
detention time of 59 minutes . Effluent samples were analyzed for
chloride and these data as well as the theoretical washout curve are
plotted in Figure 5 . The observed data points follow the theoretical
curve very closely indicating that the aeration tank is completely mixed.
237
-------
L
-
-
C
i
-
M
CJ
C
:
U
-
—
•-
.0
.9
.8
.7
.09
.08
\b
\^
\
1
)
\
\
\
i
\
C = Feed Concentration
C = Effluent Concentration
C = Initial Concentration
V
1
heoretical
Washout
\
\
TD
30 60 90
Time Of Washout, Minutes
120
150
Figure 5. TRACER WASHOUT CURVE FOR THE KEHR PROCESS AERATION TANK
238
-------
The data from the tracer washout also demonstrated the ability of
a completely mixed aeration tank to buffer shock loadings as compared
to a plug flow tank. The salt solution fed to the KASP aeration
tank was a hundred times as concentrated as the fresh water initially
in the tank. As the strong salt solution was added, it was immediately
dispersed throughout the aeration tank, which resulted in a gradual
increase in the salt concentration in the aeration tank. The washout
data demonstrating this buffering phenomena are shown graphically
in Figure 6. In a plug flow tank, the hundred fold increase would not
be dispersed throughout the tank and would move through the tank as
a highly concentrated core of salt water. If a hundred fold increase
occurred in sewage strength, this same buffering phenomena of the
completely mixed aeration tank would allow the microorganisms in the
mixed liquor to react to a slowly changing environment rather than
any sudden shock that would be experienced in a plug flow tank.
Continuous Operation of KASP - In the KASP as operated at CEL, no solids
were removed from the system as a waste sludge stream. Therefore, the
KASP could only remove carbonaceous matter and nutrients from the liquid
waste by converting them to a gaseous form. Carbon, nitrogen, and
phosphorus were the materials measured in this study and the removal
of each of these by the KASP is presented in the following results.
The KASP was operated for 28 days of continuous operation during the
Fall of 1967 and for 12 days during the Fall of 1968. Both industrial
waste and domestic sewage were treated by the KASP during this
continuous operation, and the dividing line between domestic and
cannery waste was arbitrarily set at a 8005 concentration of 300 mg/liter.
Because the canneries did not usually pack on the weekends, the sewage
during the canning season was characterized by high concentrations of
BOD5 from Monday to Friday and low BOD5 domestic waste on the weekend.
After the canning season the sewage, which primarily consisted of
domestic waste, had a fairly constant concentration of BOD^ throughout
the week.
Data showing the performance of the KASP stabilizing organic waste are
shown in Figures 7 and 8. Figure 7 shows the TOG data while Figure 8
shows the data for BOD5. On weekdays during the canning season, the
KASP treated cannery wastes with concentrations of BOD5 ranging up to
1550 mg/liter and achieved over 90 percent BOD5 removal and 80 percent
TOC removal. However, on the weekends the sewage strength decreased
while the concentration in the effluent of both BOD5 and TOC remained
at about the same level as during the week. This caused a reduction
in the percent removals of both TOC and BOD5 during the weekends. The
most illustrative example of this phenomena was Sunday, September 8, 1968
where the percent removal of BOD5 dropped to 28.4 percent and the percent
removal of TOC dropped to -48.4 percent.
239
-------
3,500
3,000
2,500
„ 2,000
B
.
—
E
1,500
1,000
500
Feed Chloride Concentration
V
Washout
Curve
Air Flow = 20 CFM
Liquid Detention Time = 59 Min.
30
60 90
Time Of Washout, Minutes
120
150
Figure 6. TRACER WASHOUT CURVE FOR THE KEHR PROCESS AERATION TANK
240
-------
CJ
o
H
100 --
50 --
0 --
50
1000
• 1968
1967.
00
e
o
H
14-1
O
c
o
c
0)
u
*Aeration time
800 --
600 --
400
200 --
30 5
October
-i—i—r
September
Figure 7 . TOC DATA FOR THE KEHR PROCESS FOR CONTINUOUS OPERATION
10
15
20
-------
i
0)
o
o
PQ
NJ
*-
N3
60
•t
c
o
C
0)
O
C
o
O
100
50
0
1600
1400
1200
1000
800
600
400
200
*Aeratlon time, hr
fc' ' '
August September October
Figure 8. KEHR PROCESS BOD DATA FOR CONTINUOUS OPERATION
-------
The nearly constant effluent quality is due in part to the inverse
of the washout phenomena shown in Figure 6. Another factor that
probably contributed to the nearly constant effluent quality is that
the BOD5 in the effluent is affected by the efficiency of the solids
separation system which is somewhat independent of the influent BOD.
During the period when the KASP treated domestic sewage, the influent
and effluent qualities were fairly constant yielding constant percent
removals of carbonaceous materials measured as either BODtj or TDC.
Because of the consistent sewage quality, the only shocking of the KASP
came from the diurnal flow variation programmed into the system.
During one 24-hour period the biological system was sampled every hour.
Four-hour composites of the feed and effluent were made up and analyzed
for TOC. The results of this test are shown in Figure 9. These data
show that the KASP produces a very consistent effluent thereby damping
out the variation in loading due to flow in a manner similar to that
observed for the organic shocking during the canning season.
Based upon the 8-hour composite samples, the BOD^ and TOC removals for
each aeration period tested were calculated. The percent removals were
calculated by summing the total amounts of 6005"and TOC in both the
feed and effluent for each aeration period rather than on a daily
basis. These percent removals are presented in Table 1 along with the
number of days of operation at each aeration period.
Table 1. Cumulative BOD^ and TOC Removals
Aeration
Period
8 hours
6 hours
4 hours
2 hours
Cumulative BOD5
Removal (%)
88.9
84.4
85.7
95.3
Cumulative TOC
Removal (%)
80.2
77.2
77.1
87.4
Period of
Operation
11 days
7 days
9 days
11 days
AVERAGE 88.5 80.4
The operation of the KASP was such that the changes in aeration period
coincided with changes in sewage strength and character resulting from
seasonal fluctuations in cannery operation. For this reason as well
as the short period of operation at each aeration period, the results
at different aeration periods cannot be compared directly. However,
the average removals of about 80 and 90 percent for TOC and BOD^ ,
respectively, do give some indication of the treatment efficiency of
the KASP.
243
-------
300
250
200
u
-
-
-:
-
.
:
_
.
c
:
j
150
100
x- X
x Effluent
M
Hour Of The Day
Figure 9 . DIURNAL VARIATION OF KEHR PROCESS TOC
244
-------
Using the average percent removals of BOD5 based upon unfiltered
samples, the rate constant for BOD removal for each aeration period
can be calculated from equation 4. These data yield rate constants
of 1.0, 0.9, 1.5, and 10.2 per hour for the 8, 6, 4, and 2-hour
aeration periods, respectively. The weekday concentration of BOD5
during the 8, 6, and 4-hour aeration periods was always greater than
300 mg/liter while for the 2-hour aeration period it was usually about
200 mg/liter. The fairly consistent values for the BOD removal rate
constant of 0.9 to 1.5 per hour at the 8, 6, and 4-hour aeration
periods were for sewage that contained industrial waste. The BOD
removal rate constant of 10.2 per hour at the 2-hour aeration period
was for domestic sewage and agrees well with the value of 15 per hour
for domestic sewage reported by Smith (3).
It is interesting to note the nearly 10 fold difference in the
calculated rate constants for the two wastes.
When designing a particular activated sludge treatment plant, one of
the most important design parameters is the loading that the type of
activated sludge being used operates under. Three different loading
parameters were calculated from the data for operation of the KASP.
The three are: the volumetric loading measured as the pounds of BOD5
per day per thousand cubic feet of aeration tank and two forms of the
organic loading which are measured as the pounds of 6005 per day per
pound of mixed liquor suspended solids and as the pounds of BOD^ per
day per pound of mixed liquor volatile suspended solids.
The various loading parameters were not constant due to the lack of
steady state conditions. For the KASP with no intentional sludge
wasting the observed average loadings, excluding weekends, were 200
pounds of BOD^ per day per thousand cubic feet of aeration tank
capacity, 0.473 pounds of BOD^ per day per pound of mixed liquor
volatile suspended solids, and 0.418 pounds of 6005 per day per
pound of mixed liquor suspended solids. It should be noted that the
mixed liquor suspended solids were high in volatile matter. They were
85 to 90 percent volatile matter which is higher than the 75 percent
found for many wastes.
The phosphate analyses on the sewage and effluent streams during the
continuous operation are shown in Table 2. Because the system was
sampled only during the eight-hour working day, a phosphate balance
around the system could not be made due to diurnal variations in flow
and waste strength. These variations caused alternate periods of
phosphate accumulation and loss by the bio-mass which are unknown.
However, it would be expected that an activated sludge system with no
waste sludge would not remove phosphate from the liquid waste. This is
shown in Table 2 where little or no phosphate removal is recorded.
245
-------
Table 2. Kehr Process Influent and Effluent Phosphate Concentrations
Total Concentration of Phosphate in mg/liter*
Date
8/29/68
8/30/68
8/31/68**
9/1/68***
9/2/68****
9/3/68
9/4/68
9/5/68
9/6/68
9/7/68**
9/8/68***
9/9/68
9/22/67
9/23/67**
9/24/67***
9/25/67
9/26/67
9/27/67
9/28/67
9/29/67
9/30/67**
10/1/67***
10/2/67
10/3/67
10/4/67
10/5/67
10/6/67
10/7/67**
10/8/67***
10/9/67
10/10/67
10/11/67
10/12/67
10/13/67
10/14/67**
10/15/67***
10/16/67
10/17/67
10/18/67
10/19/67
Aeration
Period
8 hours
n
n
n
n
n
n
n
n
"
n
"
6 hours
n
n
11
"
n
n
4 hours
"
ii
"
"
n
n
"
n
n
it
2 hours
11
n
ii
n
n
n
it
"
it
Sewage
19.2
58.0
41.8
50.7
44.4
31.1
32.0
26.0
20.5
58.9
41.1
28.6
35.8
26.6
24.8
20.6
23.4
42.4
50.0
57.2
23.8
54.2
51.2
30.2
33.4
40.4
77.6
22.6
38.0
28.6
23.4
29.4
27.4
28.6
29.
29.
28.6
28.8
27.2
.2
.2
.3
.1
.9
Effluent
14.2
18.5
18.5
23.0
36.4
35.
26.
19.
27.0
17.5
29.7
24.4
14.2
11.6
21.8
23.8
17.6
18.4
28.6
32.6
29.4
26.4
51.4
10.8
36.6
21.6
25.2
32.6
31.0
30.6
25.6
23.6
22.4
30.2
27.2
26.2
29.4
25.2
28.8
23.8
* As P04
** Saturday
*** Sunday
**** Labor Day
246
-------
The results for the nitrogen analyses are shown in Table 3 and 4.
During the period of continuous operation, the nitrogen present
in the effluent from the KASP was primarily in the ammonia or
organic form with little or no nitrite or nitrate present. The
general trend was that as the BOD5 of the sewage increased, more
of the effluent nitrogen was in the organic form. With high
strength cannery waste nearly 100 percent of the effluent nitrogen
was in the organic form while the domestic sewage only about 15
percent of the nitrogen was organic with most of the remainder
being ammonia.
Because no waste sludge was removed,the removal of nitrogen from
the liquid waste would be expected to be a minimum. However, the
data show apparent nitrogen removals of 62, 66, 53, and 48 percent
for the intervals operated at aeration periods of 8, 6, 4, and 2-hours>
respectively. These removals are based upon 8-hour composite
samples and may be explained in part by the diurnal variation in
sewage flow and strength. The eight-hour working day follows a
period of low flow and low sewage strength and therefore would be a
period of synthesis, which causes accumulation of nitrogen by the
bio-mass. This accumulation would tend to be released during the
low flow period when no samples were taken. Also complete balance
for nitrogen around the KASP system is not possible without
analyzing the composition of the off gases.
A 28-hour sampling program was conducted on the KASP effluent and
sewage in an attempt to provide more complete phosphate and
nitrogen analyses for the KASP. Grab samples of sewage and effluent
were taken every hour and made up into four-hour composites. The
nitrogen and phosphate analyses on these composites are shown in
Figures 10 and 11.
For both the nitrogen and phosphate, the difference between the
concentration in the sewage and that in the effluent, decreased
through the first 24 hours. Then during the first 4 hours of the
second day this difference increased again. The sludge was not
analyzed and during this 28 hours it was not possible to c6mplete
a nutrient balance using the feed and effluent concentration alone.
Artificial Shock Loading - To better study the effect of shock loading
conditions on the KASP, an artificial shock loading was imposed on the
system. For 48 hours fresh water was fed to the bio-mass and then
sewage (BOD5 of 230 and 180 mg/liter) was fed for the following 48 hours.
The aeration time for the study was 2 hours. The response of the
biological system to the artificial shock loading as measured by the
effluent concentrations of 6005, TOC, and suspended solids is shown
in Figures 12 through 14. During the water feed, there was a washout
of nutrients from the aeration tank. The concentrations decreased
rapidly and then appeared to level off. Therefore due to the washout
phenomena, the fresh water feed improved the effluent quality over
that prior to the study.
247
-------
Table 3. Nitrogen Data For Kehr Process Sewage
Date
8/29/6S
8/30/68
8/31/68**
9/1/68***
9/2/68****
9/3/68
9/4/68
9/5/68
9/6/68
9/7/68**
9/8/68***
9/9/68
9/22/67
9/23/67**
9/24/67***
9/25/67
9/26/67
9/27/67
9/28/67
9/29/67
9/30/67**
10/1/67***
10/2/67
10/3/67
10/4/67
10/5/67
10/6/67
10/7/67**
10/8/67***
10/9/67
10/10/67
10/11/67
10/12/67
10/13/67
10/14/67**
10/15/67***
10/16/67
10/17/67
10/18/67
10/19/67
Aeration
Period NH3* Organic-N* N02* NO-}*
8 hours 6.65
4.60
5.36
5.65
" 28.00
11 9.30
7.57
6.41
6.70
" 6.99
11 6.99
" 8.30
6 hours 7.0
" 9.3
" 29.9
14.1
" 8.6
6.9
" 10.0
4 hours 12.0
24.4
27.5
" 18.1
14.6
15.6
16.0
17.2
" 23.0
32.2
34.4
2 hours 21.9
" 20.3
20.6
15.6
" 24.0
108.2
21.6
15.3
14.3
" 19.7
18.60
16.80
15.10
10.90
15.10
20.70
19.50
—
21.30
13.10
6.41
17.80
13.1
12.1
11.1
13.6
16.6
16.2
10.1
10.1
10.8
10.2
8.9
11.1
7.5
9.8
11.3
10.0
10.7
9.1
11.3
10.3
13.0
3.8
10.3
9.6
6.4
11.1
10.5
11.1
__
—
—
—
—
—
—
—
—
—
—
—
__
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
.01
.01
Trace
—
Trace
Trace
—
Trace
—
—
—
1.35
0.88
1.10
0.72
0.38
0.50
0.28
2.36
1.54
1.55
1.15
0.95
1.00
—
—
0.98
0.96
0.82
0.54
0.85
0.18
0.08
—
—
—
—
—
—
—
—
0.16
—
0.19
0.78
0.13
—
1.36
0.92
0.92
0.92
Total N*
26.60
22.28
21.56
17.27
43.48
30.50
27.35
—
29.54
21.64
13.09
27.05
21.10
21.40
41.00
26.68
28.68
23.92
20.64
22.95
35.38
37.78
27.00
25.70
23.10
25.80
28.50
33.00
42.90
43.51
33.37
30.60
33.79
20.18
34.43
117.80
33.62
27.32
25.72
31.72
**
***
****
Values given are in mg/liter as Nitrogen
Saturday
Sunday
Labor Day
248
-------
Table 4. Nitrogen Data For Kehr Process Effluent
Aeration
Date Period
8/29/68 8 hours
8/30/68 "
8/31/68**
9/1/68*** "
9/2/68****
9/3/68 "
9/4/68
9/5/68
9/6/68
9/7/68**
9/8/68*** "
9/9/68
9/22/67 6 hours
9/23/67** "
9/24/67***
9/25/67
9/26/67
9/27/67
9/28/67
9/29/67 4 hours
9/30/67**
10/1/67*** "
10/2/67
10/3/67
10/4/67
10/5/67 "
10/6/67
10/7/67**
10/8/67***
10/9/67
10/10/67 2 hours
10/11/67
10/12/67
10/13/67
10/14/67** "
10/15/67*** "
10/16/67
10/17/67 "
10/18/67
10/19/67
_NH3*
__
—
—
—
—
—
—
—
—
—
—
0.5
1.2
—
2.3
9.8
4.8
2.8
3.1
2.0
1.2
11.7
15.7
7.0
4.4
10:3
11.6
6.5
23.7
27.0
16.8
15.0
16.8
16.3
14.5
26.8
23.8
10.1
13.4
13.7
Organic-N*
9.04
7.05
10.00
5.45
12.20
8.40
10.50
—
11.07
10.77
9.61
8.90
7.2
5.4
4.3
4.0
5.2
5.2
5.9
6.4
4.2
3.5
5.2
2.3
4.2
2.0
5.4
3.2
2.7
3.2
11.8
3.3
3.3
2.3
2.4
2.0
3.2
3.5
2.0
1.7
N02*
__
—
— —
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
Trace
Trace
Trace
Trace
Trace
Trace
Trace
0.01
—
Trace
Trace
—
Trace
—
—
—
—
—
0.20
0.18
0.10
0.10
0.10
0.06
0.35
0.60
0.29
0.31
0.29
0.18
0.36
0.22
0.40
0.22
Total N*
9.24
7.23
10.10
5.55
12.30
8.46
10.85
11.36
10.77
9.91
9.69
8.40
5.40
6.60
13.98
10.36
8.22
9.40
8.40
5.62
—
—
—
—
—
—
—
—
—
__
—
0.12
—
—
—
0.06
—
0.07
0.06
15.20
20.90
9.30
8.60
12.30
17.00
9.70
26.40
30.21
28.60
18.30
20.22
18.60
16.90
28.80
27.06
13.60
15.47
15.46
**
***
****
Values given are in mg/liter as Nitrogen
Saturday
Sunday
Labor Day
249
-------
.
Hour Of The Day
Figure- 10. DIURNAL VARIATION OF KEHR PROCESS NITROGEN
250
-------
8
N
Figure 11,
4 8 M 4
Hour Of The Day
DIURNAL VARIATION OF KEHR PROCESS PHOSPHATE
251
-------
;:
30
25
Q
:
pq
B
-
—
L5
LO
Fresh Water Feed
Unfiltered
Sample
Sewage Feed
p.—
Filtered
Sample
18 32 48 64 80 96
Time In Hours From Initiation Of The Study
Figure 12. EFFLUENT BOD AS A FUNCTION OF TIME FOR THE SHOCK LOADING STUDY
252
-------
:
-
:
*j
-
-
:
L6
80
96
32 48 64
Time In Hours From Initiation Of Study
Figure 13. EFFLUENT SUSPENDED SOLIDS AS A FUNCTION OF TIME FOR THE SHOCK LOADING STUDY
253
-------
60
40
u
30
r
—
-
20
10
,Unfiltered
Sample
A
•Fresh Water Feed
_l L_
Sewage Feed
16 32 48 64 80
Time In Hours From Initiation Of The Study
96
Figure 14. EFFLUENT TOC AS A FUNCTION OF TIME FOR THE SHOCK LOADING STUDY
254
-------
The concentration of BOD^ in the filtered effluent dropped to
nearly zero indicating that little biodegradable substrate was
present in solution. Because no food for the bio-mass was
present in solution, the production of new cellular mass was
reduced causing cell death and lysing to predominate over synthesis.
The concentrations of TOG, BOD5, and suspended solids in the
effluent after they leveled off probably corresponded to loss into
the effluent due to cell lysing.
After the sewage was turned back on, a rapid and immediate increase
was observed in the effluent concentrations of both the BOD5 and
TOG. This rapid increase was followed by a leveling off of the TOG
and filtered BOD5. The filtered BODs and TOG reached a level nearly
the same as that before the initiation of the water feed while the
level of TOG was less than before the water feed. The concentration
of unfiltered 6005 decreased after the initial rapid increase. The
sewage feed produced little change in the concentration of suspended
solids.
It is interesting to note that the KASP was able to go without any
organic loading for 48 hours with no apparent loss of treatment
efficiency when the organic load was returned. The KASP would seem
to be able to adequately handle wastes of periodic discharge.
The shock loading also had an effect, shown graphically in Figure 15 ,
upon the concentration of nitrogen and phosphate in the effluent.
During the fresh water feed, the concentrations of both decreased
rapidly to a fairly steady state value probably corresponding to the
rate of cell lysing. Upon the initiation of the sewage feed, the
concentrations of both increased rapidly but the concentration of
nitrogen soon leveled off well below the concentration in the sewage
while the phosphate increased in concentration to that in the sewage.
DISCUSSION
The operation of the KASP biological system using gravity settling
was very effective in 6005 and solids removal and performed with no
major difficulties. Once the air and liquid flows were set, the
system required no attention except for sampling and mechanical
maintenance. The biological system performed well throughout its
operation except for the period of July 9th through July llth, 1968
when the pH of the sewage ranged from 9.2 to 11.2. This high pH
waste raised the pH of the mixed liquor to 9.7, which adversely
affected most of the mixed liquor bio-mass because little or no active
microbiological life was observed by microscopic examination.
255
-------
-
-
.-
-
.-.
:
-
-
--
:~
Total -
Nitrogen
16 32 48 64 80
Time In Hours From Initiation Of The Study
Figure 15. EFFLUENT NUTRIENT CONCENTRATION AS A FUNCTION OF TIME FOR THE SHOCK LOADING STUDY
256
-------
The operation of the KASP showed that aerobic digestion could be
obtained in the aeration tank of a completely mixed activated sludge
plant at aeration times from 2 to 8 hours. This was done by having
no intentional wasting of sludge and operating such that the con-
centration of mixed liquor solids stabilized at about 10,000 ing/liter.
The KASP produced about a 90 percent reduction in BOD5 but the KASP
does have features that can limit its application.
One limitation is the low removal of TOG. Organic carbon is present
in the raw waste in two forms, non-biodegradable carbon and
biodegradable carbon. Non-biodegradable carbon is unaffected by
microbiological activity and therefore would not be removed by the
KASP or any other activated sludge process. This portion of the
organic waste will pass through unchanged and therefore limit the
amount of carbon that can be removed from the influent waste. In the
KASP, most of the biodegradable carbon, measured as BOD5, was readily
converted to a settleable activated sludge mass and carbon dioxide
allowing high removals of EOD^. However, a portion of the settleable
sludge mass cannot be biodegraded. Therefore, it cannot be oxidized
and removed as carbon dioxide. In conventional activated sludge
treatment, some of this non-biodegradable sludge is removed as waste
activated sludge. The KASP differed from conventional activated
sludge treatment in that no solids were intentionally wasted from the
system. Therefore the non-biodegradable sludge mass formed by the
KASP must either remain in the aeration tank or be washed out as
organic carbon in the effluent. McKinney (2) indicates that for
complete oxidation systems, approximately 10 to 13 percent of the
biodegradable organic matter being treated will appear in the effluent
as inert solids. For the operation of the KASP, the removal of organic
carbon was usually about 80 to 85 percent. The effluent from the KASP
was high in turbidity and color and did not have the sparkling
appearance that is observed in the effluent from a well run activated
sludge plant that produces waste activated sludge. The increase in
color and turbidity over a conventional activated sludge plant is
probably due to the added carbon carried into the effluent.
Because no waste activated sludge stream is removed from the KASP, no
inorganic nutrients are removed with the waste sludge. However, in
a conventional activated sludge plant inorganic nutrients are removed
with waste activated sludge. Therefore, BOD removal is the only area
where the KASP can equal the performance of an activated sludge plant
that produces waste sludge.
One logical extension of the KASP is increasing the volumetric BOD5
loading and thereby increasing the substrate level and correspondingly
the concentration of mixed liquor solids. The increased loading results
in a smaller aeration tank and reduced costs. However, the BOD5 loading
can only be increased if there is sufficient aeration capacity to supply
the increased oxygen demand that the increased loading would require.
Therefore, the BOD loading and the steady state concentration of solids
in the mixed liquor is limited by the aeration capacity. For this
study, 10,000 mg/liter was thought to be a good compromise.
257
-------
The KASP provides a lower removal of TOG, nitrogen and phosphorus
than can be achieved from a conventional activated sludge plant that
provides a waste sludge, but the KASP offers other advantages. The
KASP was found to be able to produce a high degree of treatment of
sewage following 48 hours with no organic loading.
One possible use of the KASP that will profit from its advantages
and not be hindered by its disadvantages is the treatment of high
strength industrial wastes prior to their discharge into a municipal
sewer. The municipal treatment plant would benefit from the
reduction in BOD5- Some of the advantages of the KASP that the
industry could benefit from are give below.
The KASP provides simple operation requiring no waste sludge
handling facilities. The KASP has the ability to withstand long
periods of no organic loading for industries that do not work over
weekends or do not have 24-hour a day production.
The KASP appears to be a suitable form of waste treatment where
about a 90 percent removal of BOD5 is desired and where little
removal of either phosphorus or nitrogen, as well as only an 80
percent removal of organic carbon, is acceptable.
An activated sludge plant such as the KASP using no intentional
sludge wasting and operating on the principal of steady state
concentration of active mass appears to have an application for
treating industrial wastes. As illustrated by the 10 fold
difference in the BOD removal rate constants found in this study,
each waste has its own characteristics. Therefore, prior to the
installation of an activated sludge plant such as the KASP, it is
desirable to operate a pilot plant to determine the necessary
design parameters for each individual waste.
258
-------
REFERENCES
1. Kehr, Dietrich, "Aerobic Sludge Stabilization in Sewage
Treatment Plants," Third International Conference on
Water Pollution Research, Section 11, Paper No. 8,
Conference Center, Messegelande Theresienhohe, Munich, Germany.
2. McKinney, Ross E.. "Microbiology for Sanitary Engineers,"
McGraw-Hill Book Company, Inc., New York, 1962,
3. Smith, H. S., and Paulson, Wayne L., "Homogeneous Activated
Sludge," Civil Engineering, 36_, 56_, May, 1966.
259
-------
CONCENTRATION 01 SUGARBEET WASTES FOR ECONOMIC TREATMENT
WITH BIOLOGICAL SYSTEMS
by
Ronald W. Brenton and James H. Fischer*
INTRODUCTION
The people of the United States consume about 10 million tons of
sugar annually, or slightly under 100 pounds per person. Of this
total, about 30 percent comes from domestic sugarbeets (6). Beet sugar
is produced at 58 factories located in 19 states from California to
Maine and from Arizona to North Dakota. The processing of sugar from
sugarbeets takes place at climatic locations and at times of the
year ranging from a high of 120°F. in the Imperial Valley of
California to a low of -AO°F. in the Red River Valley of Minnesota
and North Dakota. The sugarbeets being processed may have been
freshly harvested or may have been in open-pile storage for some
100 days.
With these vastly different conditions in mind it is easy to under-
stand that no single type of waste treatment system will operate
with equal success at all locations. The quantity and quality of
factory wastes do vary somewhat from factory to factory, however,
the basic nature of the system wastes remains quite similar.
McGinnis and Weckel (4) report that little or no attention was given
to waste disposal in the beet-sugar industry until recent years.
Complete waste treatment in conformance with modern standards would
require the equivalent of an average-sized municipal treatment plant
for each factory. Such treatment is additionally complicated by the
seasonal operations of the factories.
Beet-sugar factory wastes may be classified according to Eldredge
(2) as to source. They are: flume water; process water, lime cake
drainage; and Steffen or final waste when the Steffen process is used.
Respectively, Project Chemist, and Manager, Beet Sugar Development
Foundation, P. 0. Box 538, Fort Collins, Colorado 80521.
261
-------
Figure 1 is a typical ^diagram of the hydraulic flow in an average
modern sugarbeet factory. Those streams associated with waste and
byproduct disposal are shown. The sugarbeet, being a root crop,
brings with it to the factory high quantities of dirt and organic
particles. A high volume of water is required to flume and wash
the roots prior to the actual process of extracting and refining
the sugar. The fluming and washing waters become one of the most
difficult of the wastes to treat since it accounts for the largest
portion of the waste on a volume basis. Although water conserva-
tion practices are being exploited, 2200 gal. of water are required
per 1 ton of sugarbeets for fluming and washing operations (2).
Hungerford has (3) reported on work in Nebraska where large grass
fields were used to replace and follow the typical lagoons and ponds
to treat factory wastes. The encroachment of metropolitan areas
near to sugar factories has placed more emphasis on concentrating
treatment systems into smaller land areas. Ponds and lagoons do
not accomplish this unless they are accompanied by additional treat-
ment. Norman e_t^ al^ (5) determined that a laboratory activated sludge
unit and a pilot plant trickling filter unit could work to serve
such a purpose provided that large changes in pH could be avoided
as well as overloading.
In a report on water conservation in the beet sugar industry Browning
(1) states that recirculation of water used for fluming beets is a
common practice. The recirculation he mentioned is partial, without
supplementary treatment and not complete. Unpublished reports from
Europe have indicated success with complete recirculation accompanied
by continuous lime addition. This paper deals with total recircula-
tion of limed flume water and concentration of the dissolved and
suspended solids in the mud and recirculating water.
METHODS AND MATERIALS
It was determined that a full-scale closed-loop recirculating flume
water system was to be constructed at Longmont, Colorado. A sche-
matic diagram of the plan is shown on Figure 2. Construction began
during the spring of 1967 and was completed by September 1967 in
time for the system to become 100 percent operative as an integral
part of the Longmont Factory of The Great Western Sugar Company.
A fully-equipped mobile laboratory which contained the typical analy-
tical hardware was placed near the treatment system. The system
was designed to treat only those waters used in the fluming and
washing operations.
Relatively few changes were made in the fluming system per se;
the water treatment began after the delivery of sugarbeets past
the final spray wash.
262
-------
To River In Spring
or To City Sewage
Disposal System
[ Sugar |
Figure 1. WATER FLOW DIAGRAM FOR A TYPICAL SUGARBEET FACTORY
263
-------
Holding
Pond
Or
From Pluming Operations
To Normal —
Pluming Operations
Heated Water
To Soaking Hopper
Drain
1
DSM
Screens
Tare House
Steffen _
Evaporator j
Anaerobic
Pond
o
Emergency Bypass
First Settling
Ponds
Overflow
Figure 2. SCHEMATIC FLOW DIAGRAM, LONGMONT FLUME WATER TREATMENT SYSTEM.
-------
The flume water flow, which averaged 5000 to 6000 gpm, was first
screened thru five Dorr-Oliver parabolic screens, and then was
split. One-half of the flow returned directly to the flumes and
the other half was mixed with a hot lime slurry and sent to the
ponding system.
The ponding system was composed of two primary settling ponds
(first ponds) which received the limed flume water. The two-pond
arrangement allowed for cleaning of one pond while the other pond
was in operation.
The flow from the first pond entered the second settling pond which
was located in the middle of the two first ponds (Figure 2). The
effluent from the second pond went back to the fluming operations.
The anaerobic pond received any excess water from the second pond
and received the flume water at end of campaign. Arrangements were
available to use anaerobic pond water for make up to the system if
needed and available.
Flume water could also be sent to a holding pond as an emergency
in the event the system accumulated too much water. Flume water
could also bypass the screens enroute to the settling ponds in
situations which required more water than could be screened, again
an emergency provision.
Condenser water could be added to the transport water for fluming
beets during cold weather periods when heat was required to thaw
frozen beets from rail cars. Facilities were designed for the sec-
ond-year operation to heat the second pond effluent water for the
aforementioned purpose, thus reducing the build up of system waters
by direct addition of condenser water.
PROCEDURES
The experimental procedure and objectives included the following:
1. To operate the flume water system at various pH levels (which
were to be achieved and maintained by the addition of differential
amounts of lime). To determine: the effects of pH on the system;
the effects of the system on the factory's ability to control the
pH; the most economically feasible, yet operationally acceptable,
pH range; the amount of pollution reduction achieved by the ponding
system and the amount of residual pollution which would build up in
the system water; and the biological characteristics of the ponding
system.
2. To study the flume water after the beet processing season to
determine: the rate of degradation; the eventual fate of various
flume water components; the operational parameters (nutrient addi-
tion, aeration, etc.) which would achieve the most rapid and com-
plete biodegradation.
265
-------
During the beet processing season 24-hour composite samples were
taken from the first pond influent, first pond effluent, and second
pond effluent. These samples were taken and analyzed five days per
week.
After the beet processing season, the flume water was discharged to
the anaerobic pond. Samples were taken from various depths at two
points on this pond. The individual samples were composited and
analyzed from one to two times per week.
RESULTS
Recirculating Water, During Campaign
During the first six weeks of the processing campaign it was found
that pH levels above 10.0 could be maintained with as little as
from 200 to 300 Ibs of lime per hour added to a flow rate of about
3000 gpm, (Figure 3) but once the pH fell below 10.0, it would con-
tinue to fall to between 6 and 8; pH levels between 7-10 could not
be effectively controlled.
After the seventh week of campaign, cold weather necessitated the
use of heated flume water (as indicated by rising pond temperatures),
and it was found that adding heat to the system made high pH levels
impossible to maintain. In fact during the 10th week of operation
1000 Ibs of lime per hour could not reestablish a pH level above 8.0.
It was also found that when pH levels were greater than 10.0 dissolved
oxygen was maintained in the system which was not true when levels
were below 10.0. (See Figure 4). This was indicated by the fact
that dissolved oxygen was carried completely through the ponding
system only when the pH of the second pond effluent was greater
than 10.0.
Mud Settling, During Campaign
The primary function of the first settling pond was to remove the
majority of the suspended solids from the flume water. Almost
98,000 pounds of total solids (including suspended and dissolved
solids) were settled each day (See Table 1). The settled mud aver-
aged a moisture content of 83 percent. The fluid nature of the
mud made its removal from the settling ponds very difficult.
More than 3 percent of the mud was carbonaceous in nature, and
probably was susceptible to biodegradation. Very foul odors were
released from the mud, while it was being removed from the ponds,
indicating that it was very septic.
Although the settling efficiency of the system was good, the small
particle suspended solids seemed to remain in suspension and carry
through the system (See Figure 5) as the suspended solids concentra-
tion at the second pond effluent increased with continued recirculation.
266
-------
Periods
o
n
4J
c
o
u w
P*
0) W
00
a
<; P.
a) -d
0)
60 "
o) t3
M 0)
0) T3
> n3
•5
-------
a
a,
o
Q
oo
to
a)
ai
2=
2 -
1 .
12 .
11 -
33
a.
01
IU
OJ
10-
9-
Figure 4.
Weeks
COMPARISON BETWEEN PH AND DISSOLVED OXYGEN AT THREE SYSTEM
LOCATIONS, DURING THE PROCESSING SEASON.
268
-------
TABLE 1. - Weekly mud composites 1968-69 campaign average.
% on Total Solids
Average Pounds Per Day
Relative Density 1.366
wet , g/ml
Absolute Density
g/ml^ 2.59
pH3 8.6
Total CaO
Soluble CaO
Alkalinity as CaO
NJ
S Dissolved Solids
Suspended Solids
Total Solids
Total Carbon
6.36
1.25
0.16
12.89
87.11
100.0
3.25
6,220
1,222
156
12,606
85,190
97,796
3,178
"Relative Density based on maximum compacted volume.
;
"Calculated from Relative Density and percent moisture.
pH was taken on mud diluted to a particular volume, usually 500 ml.
-------
t
13,000 .
12,000 _
11,000 .
10,000 -
9,000 -
8,000 -
7,000 -
6,000 -
5,000 -
4,000 _
3,000 _
2,000 -
1,000
Dissolved Solids
p-o
Suspended Solids
2 4 6 8 10 12 14 16
Weeks
Figure 5. ACCUMULATION RATE OF DISSOLVED AND SUSPENDED SOLIDS,
DURING THE PROCESSING SEASON
270
-------
In the recirculating flume water much of the pollutant load was
removed by combined settling and biological activity. A proportion
carried through and built up increasing concentrations with contin-
ued reuse.
Build-up of Unsettled Solids, During Campaign
As much as 75 percent of the entering BOD was removed and almost
68 percent of the COD was removed. About 75 percent of the total
carbon was removed in the ponds and about 80 percent of the total
CaO added to the system was removed by the ponds (Table 2).
The build-up of residual solids in the system produced a peak BOD
concentration of more than 6000 ppm (See Figure 6). The COD reached
a peak concentration of more than 10,000 ppm and the total carbon
concentration rose to over 3,000 ppm. The dissolved solids con-
centration also reached a peak of greater than 10,000 ppm (See
Figure 5) .
The amount of total sugar in the system was dependent upon the
pH (See Figure 7). When the pH was maintained above 10.0 the
sugar concentration increased with the time. When the pH fell
under 10.0 the sugar concentration decreased and the organic acid
concentration increased. This was probably due to the influence
of pH upon the bacteria in the ponds. Apparently pH levels above
10.0 seriously inhibit bacterial activity and acid production, but
allow sugar to build up in the system.
Generally speaking, high pond temperatures and low pH seemed to
be the best environment for high bioactivity, but the data are
not clear in many instances. The fact that the total aerobic
count was almost always approximately equal to the total anaero-
bic count indicated that the bacteria were facultative in nature.
(Bio-analyses were performed by the Department of Microbiology at
Colorado State University).
Post Processing Season Treatment of Water
After the end of the beet processing season, the flume water was
discharged to a 1.8 surface acre x 15 ft deep anaerobic pond with
two floating 5-HP aerators. The water was held for a period of
31 weeks at which time the quality met discharge standards.
The changes in concentration of the constituents in the residue
water selected for analyses are shown in Figures 8 and 9. Note that
the pH of the water slowly rose from about 7 to slightly greater
than 8. The nitrate and nitrite concentrations consistently remained
below the 0.3 ppm level.
271
-------
Table 2. Analyzed substances entering and leaving the first and second ponds
in series, showing percentage removed by the ponds in pounds per ton
of beets sliced, 1968-69 campaign average.
Analyses
BOD5
COD
Soluble CaO
Total CaO
^ Dissolved Solids
VJ
10 Total Carbon
Total Sugar3
Total Acids^
(As CaO)
Organic Acids^
(As CaO)
Pounds, Per
Ton of Beets,
Entering
First Pond
2.76
2.77
0.61
1.40
3.13
2.03
2.24
—
—
Reduction (or
increase)
across both
Ponds
-2.082
-1.74
-0.34
-1.12
-1.68
-1.53
-1.93
+0.31
+0.59
Pounds, Per
Ton of Beets,
residual in
pounds
0.68
1.03
0.27
0.28
1.45
0.50
0.31
0.42
0.52
% Reduction
75.4
62.8
55.7
80.0
53.7
75.4
86.2
— —
_._
Calculated on static pond volume basis and as if no physical loss and subsequent dilution occurred.
Minus sign designates reduction, plus sign designates increase.
3
Calculated using periods characterized by high pond pH, where sugar concentration showed a tendency
to increase.
4
Increase across ponds approximately equal to residual in ponds, so no reduction occurred.
-------
12,000 -,
11,000
10,000
9,000
8,000
7,000
6,000
5,000
4,000
3,000 -
2,000 -
1,000 -
TOG
o—•—o BOD
O-——O COD
2 46 8 10 12 14 16
Weeks
Figure 6. ACCUMULATION RATE OF BOD, COD AND TOTAL ORGANIC CARBON,
DURING THE PROCESSING SEASON
273
-------
7,000_
6,000_
5,000_
4,000-
3,000-
2,000'
1,000-
Total Sugar
Organic Acid
O—.—O pH
/•
<}
A
n
!\
A
i \
l
V /V'
/ \! ;\\J
\; i
\
-12.0
-11.0
- 10.0
9.0
8.0
- 7.0
. 6.0
8 10 12 14 16
Weeks
Figure 7. COMPARISON OF PH WITH TOTAL SUGAR AND ORGANIC ACIDS,
DURING THE PROCESSING SEASON
274
-------
U3
J-l
•H
c
as
tx
E
a.
8
7
6 _
5
3 _
2 -
1 -
0—0
Q—a Total CaO (Right Scale)
Soluble CaO (Right Scale)
Sulfides
N03+N02-Nitrogen
Dissolved Oxygen
14 16 18 20
May I June
Weeks
Figure 8. CONCENTRATION CHANGES IN SELECTED CONSTITUENTS, POST-PROCESSING SEASON
Sept,
_ 1600
- 1400
~ 1200
- 1000
- 800
- 600
- 400
200
0
O.
a.
-------
5000 -
Dissolved Solids
COD
Organic Acids ppm Acetic
BOD
O—O Totw.j. Acids ppm CaO
Suspended Solids
Sulfate
4000 -
(X
3000 -
ho
••J
ON
2000 -
1000 -
Feb.
14 16 18 20
May I June
Weeks
Figure 9. CONCENTRATION CHANGES IN SELECTED CONSTITUENTS, POST-PROCESSING SEASON.
-------
The sulfide concentrations steadily increased to a peak during the
9th week. It was noted that the highest detection of odor was recorded
during this same period of time. The increase in sulfides was inversely
associated with sulfate until sulfate had essentially dissipated.
However, the slow decrease of sulfides after peaking (even with the
aerators in operation) would cause one to postulate that organically-
bound sulfide was being released.
Dissolved oxygen appeared during the 27th week. At this point the
BOD was less than 100 and green algae became evident. Within the
next 4 weeks, D.O. had increased to an average of 4 ppm and BOD
has been reduced to 31 ppm.
It is of interest to note that the post-campaign pond in which the
water was treated was expected to be anaerobic. Theoretically, two
5-HP floating surface aerators were to keep some entrained oxygen
in the top layer of the 15-ft deep pond,- thus an aerobic layer.
The deeper portions of the pond were expected to be anaerobic.
The biological data revealed that the pond existed continually
in a biologically falcultative state as stratified anaerobic counts
were the same as the aerobic counts. Biological examination also
revealed that the prevalent organism was of the genus leuconostoc;
leuconostoc is a facultative organism.
The nutrient analyses revealed that very little phosphate was avail-
able (See Figure 10). In an attempt to increase the rate of biode-
gradation phosphate was added to the pond six times during the re-
tention period. The first two times a dry fertilizer was used with
the results being only partially satisfactory, since the formulation
did not completely dissolve. The last four times, liquid ammonium
phosphate was used with the results being fully satisfactory as
shown by analysis. Each time phosphate was added the BOD degrada-
tion rate increased which indicated that additional phosphate was
required.
Note that after each phosphate addition the phosphate was quickly
utilized with the exception of after the last addition whereupon
the phosphate concentration remained consistent. Since phosphate
limits in discharge waters have been established for most states,
phosphate must be added carefully towards the end of the holding
period if the water is to be discharged. As expected temperature also
appeared to effect the rate of biodegradation. Temperatures above
15°C. seemed to produce a faster rate of activity than did temper-
atures below 15°G.
All of the data collected were statistically treated and computer
print out curves were produced. An attempt was made to develop
nomographs whereby COD data could be reliably used to predict BOD.
Converted COD values were established as adequate for internal con-
trol, but not accurate enough for the reporting to regulatory branches
of the State and Federal Governments.
277
-------
•z.
m
ca
§
hO
VJ
00
D.
e
OJ
H
-a-
O
CM
10 12 14 16 18 20 22 24
246
Feb. I March
April
June
May
Weeks
Figure 10. EFFECTS OF NUTRIENTS AND TEMPERATURE ON RATE OF BOD REMOVAL
Sept.
-------
It became evident from all the data, that if the impounded water
was of high enough quality for discharge, there was no valid reason
to release it. As a result the water from the first year of opera-
tion was used to fill the fluming system the second year and was
again used for the same purpose during the 1969-70 processing sea-
son. With the exception of a planned discharge from the deep pond
during the first year, none of the flume water has been discharged
to a nearby stream. From this standpoint the results have exceeded
the initial expectations.
SUMMARY AND CONCLUSIONS
Considerable progress has been made by the U. S. beet sugar industry
during the past several decades in reducing the discharge of suspended
and dissolved solids to receiving water bodies. A two year study on
the containment, treatment, recirculation and reuse of sugarbeet
fluming and wash water is reported.
Settleable solids were concentrated into a sludge (mud) by contin-
uous addition of a slaked lime slurry. Two alternately-used first
settling ponds in series with a second pond removed an average of
75 percent of the BOD and 54 percent of the dissolved solids. When
pH exceeded 10.0, greater than 90 percent of the settleable solids
were removed before water returned to the fluming and washing oper-
ation. BOD, COD and total dissolved solids continued to increase
during the operating campaign. With a water temperature rise above
20°C., pH dropped and could not be recovered with large quantities
of lime addition.
At the end of a 16-week processing campaign, the system and surplus
waters were discharged into a 1.8 surface acres by 15 ft. deep pond
with two 5-HP surface aerators. The initial 3000 ppm BOD, 6000 ppm
dissolved solids water was degraded to less than 50 BOD, 4 ppm D. 0.
The expected anaerobic processes proved to be facultative prolonging
the period of digestion to 31 weeks. The prevalent organism was
leuconostoc. Odor initially was very slight, rose to a peak in about
9 weeks and declined to a negligible degree 8 weeks later. "Odor was
associated with the analyses for sulfides. The rate of degradation
was enhanced by the periodic addition of liquid ammonium phosphate.
The ponded water was not discharged but was used to fill the fluming-
washing system for the next processing season.
Sugarbeet flume water can be recirculated without odor provided the
water temperature and pH can be controlled. Settleable solids can
be removed by ponding or by mechanical clarifiers. Non-odorous
treatment of the settled mud remains an unsolved problem. The
control of odors during peak activity during post campaign treat-
ment of residue water remains unsolved.
279
-------
REFERENCES
1. Browning, J. E. 1954. Water conservation and process water
utilization in the beet sugar industry. Am. Soc. Sugar Beet
Technol. Proc. 8(2): 254-257
2. Eldridge, E. F. 1942. Industrial waste treatment practice.
McGraw-Hill Book Company, New York, N. Y.
3. Hungerford, E. H. 1954. Factory waste stabilization by
aeration On large fields. Am. Soc. Sugar Beet Technol. Proc.
8(2): 258-259
4. McGinnis, R. A. 1951. Beet-sugar technology. Reinhold
Publishing Corp., New York, N. Y. 574 pp.
5. Norman, Lloyd W., James E. Laughlen, and L. 0. Mills. 1965.
Waste water treatment studies at Tracy, California. J. Am.
Soc. Sugar Beet Technol. 13(5): 415-424
6. Sugar Statistics and Related Data. 1969. Volume II (Revised)
U. S. Dept. of Agri. Statistical Bull. No. 244. Washington,
D. C. 78 pp.
280
-------
RECONDITIONING AND REUSE OF OLIVE PROCESSING BRINES
by
Walter A. Mercer, Harry J. Maagdenberg, and Jack W. Rails*
The preservation of olives in California is a small industry with
a wholesale product value of approximately $40,000,000 annually.
The preparation of olives for canning creates a strong liquid
waste which is high in both BOD and sodium chloride content. Since
the latter material is not biodegradable, discharges from olive
canning plants have the potential for saline pollution of receiving
waters.
The industry receives about 50,000 tons of olives for preservation
during mid-September to mid-November of each year. A portion of
the freshly harvested olives are treated with lye, aerated and
washed for black, ripe olive production (so called fresh-curing).
The bulk of the freshly harvested olives are put in storage using
3-4 percent sodium chloride solutions; these are increased in
concentration to 6-8 percent over a period of 4-10 weeks. Olives
from brine storage are taken for preparation of canned, black ripe
olives over a period of 6-11 months, depending on the production
schedule of the individual cannery.
The purpose of the project being described in this paper was the
reconditioning of spent olive storage brines on a commercial scale
and evaluation of their reuse potential. The technology under
investigation was based on exploratory work done at the Albany
(California) Laboratory of the USDA where olive brines were upgraded
using activated carbon treatment.
A pilot-plant-scale, brine, reconditioning unit was constructed
and used at four olive canneries in the Central Valley of California.
A schematic of the brine reconditioning unit is shown on Figure 1.
Figures 2 to 4 show photographs of the brine reconditioning unit
on site at an olive cannery.
* Western Research Laboratory, National Canners Association,
1950 6th Street, Berkeley, California 94710
281
-------
FRESH CARBON
SLURRY FEED-^'
TANK
>=COMPRESSED AIR
OPERATED VALVES
E
INF.
TANK
INFLUENT BRINE-
AIR
COMPRESSOR
I
FEED PUMP
A
X:
EFFLUENT BRINE—-
DESIGN CAPACITY;
3.5GALLONS/MIN. OF BRINE
CAR BON CAPACITY;
1040 POUNDS
COST OF UNIT;
$15,000 (WITH ALL ACCESSORIES)
UNIT CARBON COST;
$0.30 PER POUND
LU
CO
O
CO
oe.
o
o
•<
SPENT
I CARBON
D
U>
10 FEET
CARBON SLURRY
'PREPARATION TANK
CARBON
BLOWPOT
EFF.
TANK
Figure 1. SCHEMATIC OF BRINE RECONDITIONING UNIT
282
-------
Figure 2. BRINE RECONDITIONING UNIT
283
-------
Figure 3. DETAIL VIEW OF BRINE RECONDITIONING UNIT
284
-------
Figure 4. BLOWPOT AND CONTROL PANEL
285
-------
At each of the four locations, substantial volumes of spent
storage brine were put through the carbon column. The influent
feed rate was varied from 0.5 to 3.5 gallons per minute. This
flow rate corresponded to 0.36 to 2.5 gallons per minute per
square foot of carbon area. The extent of treatment of the brine
was followed by measuring COD, suspended solids, pH, total acidity
and color (transmittance at 425 millimicrons).
The results of a series of analyses made on brine samples are
illustrated on Figures 5 to 9. The data indicate the surges
which were always observed when the brine contacted fresh carbon.
In the figures showing changes in pH, total acidity and COD, there
were six additions of 60 pound portions of fresh carbon, and a
significant change was noted in the analytical measurements
determined immediately after each addition of carbon.
The measurement of light transmittance followed a consistent
pattern (Figure 7). The effluent was essentially transparent
to light of 425 millimicrons wavelength until approximately
6,000 gallons of brine had contacted a single charge of 1000
pounds of activated carbon; the transmittance then decreased with
increasing effluent gallonage to a value of about 75 percent at
10,000 gallons of collected effluent. The suspended solids content
followed no definite pattern in one series of samples. In general,
the suspended solids content decreased due to the carbon bed acting
as a filter, but the inconsistent generation of carbon fines from
loading of carbon slurry and abrasion gave a random pattern to the
results. The fluctuations in values on fresh carbon addition were
of short time duration and typical values for composited effluents
are tabulated in Table 1. These results show that carbon treatment
significantly reduced COD and color, but had little effect on pH,
total acidity, salt content, and suspended solids content.
A secondary disposal problem in olive canning plants is caused by
the large volume, low salt strength, processing waters. Two types
of processing waters were treated with activated carbon with the
results shown in Table 2. For the lye rinse water and transport
brine, there were substantial reductions in suspended solids
(apparently due to larger particle size compared to storage brines)
and COD. These treated rinse waters were reused in commercial
production with no detectable effect on the quality of the canned
olives as judged by production personnel. There is considerable
promise of using carbon treatment of processing waters to recondition
these liquid wastes for reuse at considerable saving in potable
water and reduction of saline pollution potential.
The project plan included storage experiments of freshly harvested
olives in carbon treated brine. The success of the storage was
judged by the quality of the final canned product as determined by
a taste panel.
286
-------
9,-
3 —
_ T FRESH CARBON
ADDED
111 III
L5 3 45 6 7.5 9
1 1 1
10. 5 12 12. 5
GALLONS TREATED (IN THOUSANDS)
Figure 5. pH OF RECONDITIONED LINDSAY STORAGE BRINE
287
-------
CD
O
o
§•»
o
i
2.2
o
I
FRESH CARBON
ADDED
L5
45 6 7.5 9
GALLONS TREATED (IN THOUSANDS)
10.5
12
Figure 6. TOTAL ACIDITY OF RECONDITIONED LINDSAY STORAGE BRINE
288
-------
18000 f-
1.5
4.5 6 7.5 9
GALLONS TREATED (IN THOUSANDS}
10.5
12
12.5
Figure 7. CHEMICAL OXYGEN DEMAND OF RECONDITIONED
LINDSAY STORAGE BRINE
289
-------
15
0
1
1
1
Z5 5.0 7.5
GALLONS TREATED (IN THOUSANDS)
10.0
Figure 8. COLOR INDEX OF RECONDITIONED VISALIA STORAGE BRINES
290
-------
90
I
I
I
2. 5 5.0 7. 5
GALLONS TREATED (IN THOUSANDS)
10.0
Figure 9. SUSPENDED SOLIDS CONTENT OF RECONDITIONED VISALIA
STORAGE BRINE
291
-------
Table 1. Brine Reconditioning
pH
Acidity
Color
NaCl
SS
COD
Influent
4.0-4.3
0.3
70.5
5.5
200-286
18,000
Effluent
4.2-8.8
0.02-0.39
75-100
5.2-5.5
164-220
5,000-16,500
Table 2. Removal of Suspended Solids and Chemical Oxygen Demand
From Brines
Liquid Treated
Lye Rinse Water
Transport Brine
Domestic Spanish Olive Brine
Imported Spanish Olive Brine
and Process Waters
Percent Removal ,
Suspended Solids
91
100
21
30
Average
COD
62
25
38
65
Table 3. Quality of Olives Stored In Reconditioned Brine
Extent of Treatment
Gallons Brine :
Pounds of Carbon
14 : 1
2 : 1
5 : 1
Similar Commercial Sample
Brine Storage
Time, Days
64
36
4
Canned Product
Flavor Ranking
1 = Best, 4 = Worst
2.13
2.85
2.62
2.39
292
-------
A 15 ton lot of olives were stored in 1530 gallons of
reconditioned brine for several months. A sample of olives
stored in this tank was prepared and canned at. the six month
storage time. A taste panel could not detect any difference
between the sample from reconditioned brine storage and a
similar sample stored in virgin brine. The olives in the tank
of reconditioned brine were used in commercial production and
sales with no adverse comment by quality assurance inspection
or consumer reaction.
A more extensive testing of reconditioned brine was made during
1969-1970. One hundred pound lots of olives were stored in
three types of reconditioned brines having different effluent
volume to carbon ratios and different lengths of storage after
reconditioning in the absence of olives. After 4.5 months of
storage, samples from each of the three types of brine were
prepared and canned. A similar sample from virgin brine storage
was used as a control. The canned samples were presented to a
laboratory taste panel who were asked to rank the samples, in
order of preference with 1 for the best and 4 for the worst. If
the samples were identical in flavor, then each sample would have
a ranking number of 2.5 after a number of individual judgements
were averaged. The results of this test are summarized in Table 3.
The sample from the brine having the highest gallonage to carbon
ratio and the longest storage period before olives were introduced
has a ranking comparable to the commercial control samples.
Actually, all four samples were of good quality and difficult to
distinguish from one another.
We conclude from this study that the reconditioning and reuse of
olive storage brines is a commercially feasible process. We see
two options for use of this carbon treating of brine in the
industry. In a first stage, two to four canneries with an urgent
need to reduce saline waste liquid discharge would each have a
brine reconditioning unit capable of producing in a 12-week period
enough brine to store 5,000 tons of freshly harvested olives. No
carbon regeneration would be done at this stage. In a second
stage, 10 canneries would each have a brine reconditioning unit
rated at 5,000 tons of olives storage capacity. A centrally located
carbon regeneration unit would be operated cooperatively to
regenerate the carbon from all ten plants.
The cost of brine reconditioning would be about $0.50 per ton of
olives in the first stage and $1.00 per ton of olives in the second
stage, using a 10-year amortization period. Freshly harvested
olives average about $300.00 per ton, so the increment in cost due
to brine reconditioning is a relatively small figure.
293
-------
TRICKLING FILTER TREATMENT OF FOOD CANNING WASTE WATERS
by
Walter W. Rose*
INTRODUCTION
The treatment of food canning waste waters, by the trickling filter
process, has had a long and varied history. Under optimum condi-
tions, the system has produced satisfactory results. Many investi-
gators including sewage treatment plant operators have, at times,
experienced little success with the operation of conventional rock
filled filters. I believe some of the reasons arock filters have
not been able to provide adequate treatment to food wastes are as
follows:
1. Canning operations may, by necessity create a sudden change in
the volume of the effluent discharged or a sudden change in the
character of the waste. The most common changes are increases in
acidity or alkalinity or in the organic load.
2. Related to these changes is the possible stop and go nature of
plant operations which is caused by breakdowns or fluctuations in
the quantity of incoming raw product.
3. Wastes from fruit canning are deficient in certain nutrients
needed in a biological treatment system. Most food wastes are low
in nitrogen and possibly phosphorous.
4. Food canning wastes are generally high in carbohydrates and other
simple organic compounds. These compounds are readily metabolized
by bacteria and as such impose a high immediate oxygen demand on the
treatment system.
Points number 1 and 4 create conditions which seriously affect the
performance of rock filters.
In recent years significant changes have been made in certain aspects
of trickling filter treatment systems. One of the most important
changes has been in development of uniform filter media as a substi-
tute for rock. In particular, the development of plastic to replace
rock has greatly expanded the potential of using the trickling filter
*National Canners Association, 1950 6th street, Berkeley, California 94710
295
-------
process in treating high strength industrial wastes. Much credit for
exploring the use of plastic filled trickling filters must go to Dr.
Chipperfield in England and Germain in this country.
The National Canners Association began an evaluation of these new
filters in 1962. In this paper I will attempt to highlight some of
our findings that have been obtained from three different units.
Figure 1 illustrates an important principle which is applicable to
any treatment system. Any product whether it be artichoke or zucchini'
will have certain operations that produce a high strength waste stream.
In our surveys we have found that within a plant there may be 2 to 3
waste streams that account for most of the BOD discharged and that
the volume of these waste streams will be a low percentage of the
total flow. For economical treatment, the strong waste should be
segregated from the weaker wastes streams.
Outlined below are four reasons or justifications for conducting
research on the use of plastic filled trickling filters.
1. High Concentration of Soluble Solids - Studies have shown that
80 - 90 percent of the BOD is in the soluble form. Compounds in the
dissolved state are not readily removed from solution by either phy-
sical or chemical means.
2. Necessity for Microbial Removal of Soluble Solids - If physical
and chemical methods are not appropriate for the removal of soluble
solids, then one must rely on biological methods.
3. Need for High Rate-Space Saving Treatment Method - This is par-
ticularly true in urban areas where land costs are high or nearly
non-existent as a result of plant expansion over the years.
4. High Organic Loadings and Removals - past research has indicated
that plastic filters are especially adaptable to both of these cri-
teria that are not easily met by other treatment systems.
The following is a list of parameters which we have investigated.
The first five items are basic to any biological process. The sixth
item is the only one unique to the trickling filter process. It has
been necessary for us to evaluate each of the parameters as there
was a scarcity of available information related to treatment of food
processing wastes using trickling filters.
1. Toxicity - This was not a problem since we were investigating
waste from food processing.
2. Nutrients - Bacteria require a balanced diet if they are to be
efficient in the removal of BOD.
296
-------
PO
vO
Pre-Peeling^
Operations
BOD= m
Flow 3%
ALL OTHER WASTE FLOWS
A \
BOD-39%
Flow=79%
COMBINED WASTE FLOW
Caustic Peeling
Solution
BOD= 5%
Flow* 0.1%
After Peeling
BOD= 40%
Flow* 18%
BOD= 61%
Flow- 21%
Figure 1. SOURCE AND STRENGTH OF WASTES WATERS FROM PEACH CANNING
-------
3. pH - Must not be too acid or alkaline otherwise performance of
the treatment system is adversely affected.
4. Organic Loading - To determine the ability of the filter in
accepting high organic loadings.
5. Hydraulic Loading - To determine the reaction of the system to
both high and low flow rates.
6. Depth of Packing - To determine the distribution of BOD removals
with filter depth for optimum use of the total height.
Figure 2 is a schematic drawing of a pilot scale trickling filter that
was used in the initial phases of our investigation. The filter is
34 feet in height and 3 feet in diameter. It is packed to a depth of
21.5 feet with plastic media. The unit contains pumps, rotary distri-
butor, nutrient feed and sample collection equipment.
Separate pumps are used to introduce raw waste and a recycle stream
to the top of the filter. A distributor spreads the combined flows
over the top of the packing media. Experience has shown that within
3 to 4 days after the start up, the unit is operational.
Figure 3 is a photograph of one bundle or module of plastic media
that was used in all three trickling filters. This particular media
is manufactured by the Dow Chemical Company and is distributed under
the trade name of Surfpac. There are other media of varying design
and compositon but we have not had any personal experience with per-
formance from such units.
The plastic media, using a honey comb design, has a high void (94
percent) and a high surface to volume ratic (27 sq. ft. per cubic
ft.). The material is polyvinylchloride and is self-supporting
to a height of 21.5 feet. The module has nominal measurements of
19-1/2" wide, 39" long and 21.5" in depth.
Listed below are results obtained from the pilot filter. The results
are for different products and illustrate that the treatability of
each should be considered in designing a treatment system. In obtain-
ing these results, the hydraulic loading remained constant. However,
the organic loading was not uniform -and must also be considered in
designing a treatment system.
298
-------
RAW a RECYCLE
ORIFICE BOXES
RECYCLE
OVERFLOW
MOTORIZED
DISTRIBUTOR
SHELL —
36" Sch.~40
STEEL PIPE
SUPPORT
GRATI NG
AIR
PORTS
NUTRIENT
FEED-
TANK
NUTRIENT
PUMP
6 LAYERS
of
PACKING
10.8'
RAW OVERFLOW
TREATED WASTE
RAW WASTE
Figure 2. SCHEMATIC DIAGRAM OF THE FILTER UNIT SHOWING THF
ARRANGEMENT OF THE FEED SYSTEMS AND FILTER LAYERS
299
-------
Figure 3. SINGLE BUNDLE OF PACKING MEDIUM.
300
-------
Table 1. Filter Performance
Product
Peas
Peaches
Apricots
Pears and Carrots
Beets
BOD ppm
642
722
320
801
153
Percent Removed
24.7
33.0
33.9
41.4
86.6
Table 2 illustrates the effect of various hydraulic rates and the
importance of adding nutrients on BOD removals. At a high recycle
to raw waste ratio (14:1) over 85 percent BOD removal can be obtained.
As the ratio is decreased, the percent removal also decreases. The
importance of nutrients, in this case diammonium phosphate, is illus-
strated by comparing the two columns on the right. The loadings were
the same in each case but better removals were obtained when diammon-
ium phosphate was added to the waste water. This was expected because
the waste is low in nitrogen which is a requirement for good bacterial
growth.
Table 2
Performance of Filter Unit based on
Removal of Biochemical Oxygen Demand (BOD)
Flow Rates (GPM)
Raw Waste
Recycle Effluent
Total Flow
Filter Performance
(PPM BOD Removed)
Influent
Effluent
Reduction (%)
*Diammonium phosphate added
1
14
15
4033
580
85.5
3
14
17
3200
1395
56.5
5-
14
19
7
7
14
7
7
14*
2700
1800
33.3
3210
2040
36.3
2750
1515
45.0
Figure 4 is a curve relating the effect of pH on the performance of the
filter. It can be seen that the percent BOD reduction decreases with
either a low pH or a high pH. The poor performance is a result of not
301
-------
0
H-
O
Z3
O
LU
oi
t—
UJ
O
Q_
50
40
30
20'
.
/ \
/ \
/ \
/ \
/ \
/ \
/ \
/ »
/ •
• NW
S ^sSi^
/ ^S*x*»
•^^^»
• ^
i i » i t i i i t \ *
40 .5.0 6.0 7.0 8.0 9.0
pH RANGES
Figure 4. EFFECT OF pH ON BOD REMOVAL
302
-------
providing optimum growth conditions for the microflora on the packing
media. An acceptable pH range is between 6.5 and 8.0.
Figure 5 plots the effect of organic loading vs. percent reduction.
As shown, the percent removal decreases as the applied load increases.
Beyond a loading of 1500 Ibs. BOD/day/1000 cu.ft. the removal drops
off rapidly. At this and higher loadings, the filter has become sat-
urated with organic matter and the microflora cannot remove additional
quantities from the waste water.
Figure 6 is a schematic drawing of the second trickling filter system
to be discussed. The screened waste passes through two 55 gallon drums
then into one side of the wet well. A pump withdraws the waste from
the sump and transports the liquid to the top of the filter. There, a
motor driven-open trough distributor spreads the waste over the top sur-
face of packing media. The waste flow, after passing down the filter,
enters the other side of the wet well. A portion of the effluent over-
flows out of the sump and the remainder passes under the baffle to be-
come recycled
This past season (1969) two intermediate sample points were added in
addition to the normal influent - effluent sample points. The new
sample points made it possible to obtain data on removals at different
filter depths. The depths of the sample points are as follows:
A = 7.2 ft. from the top of the packing media
B = 14.4 ft. from the top
C = 21.5 ft. or effluent from the bottom of the filter
Figure 7 is a summary of the data collected last year and shows the
percent removals at the various sample points. The first one third
of the filter removed 28% of the total load removed by the filter. At
14.4 ft., 33 percent of the total load was removed and at 21.5 ft., 40
percent removed. Breaking down the removals by section we see that
the top third of the filter removed 28 percent, the middle section
removed an additional 5 percent and the bottom third removed the re-
maining 7 percent.
Figure 8 is another way of plotting the distribution of the pounds
removed by each section of the filter. If we take the 40 percent that
was removed by the total filter and determine on a percentage basis, the
contribution of each section towards that removal, results are obtained
as follows:
75 percent was taken out by the first 7.2 ft.
14 percent was taken out by the next 7.2 ft.
11 percent was taken out by the remaining 7.3 ft.
We believe there are at least two factors which contributed to the
disproportionate removals at the different filter depths. First a
303
-------
80
o
ID
Q
UJ
O
C£
LU
Q_
60
40
20
• = PEARS & CARROTS
*= PREVIOUS STUDY
500 1000 1500
ORG. LOADING, IBS/ 1000 FT'/ DAY
Figure 5. EFFECT OF ORGANIC LOADING ON BOD REMOVAL
304
-------
ROTARY
DISTRIBUTOR
FRESH WASTE-
RECYCLED WASTE
T METER
6" VARIABLE SPEED
PUMP
METER
AIR
rPORT
TREATED
WASTE
TREATMENT COLUMN
FRESH-SCREEN ED
WASTE
METER
TREATED WASTE
OVERFLOW
BAFFLE —
|
1
i
_ ~^
\ r
ANHYDROUS
AMMONIA
(NO SCALE)
WET WELL SUMP
Figure 6. HIGH RATE TRICKLING FILTER SYSTEM
305
-------
LU
O
40
30
20
10
14.4 21.5
FILTER DEPTH (FT.)
Figure 7. COD REMOVAL VS. FILTER DEPTH (NATURAL AERATION)
306
-------
UJ
o
o
o
100
75
50
25
11
7.2 14 4 21.5
PACK ING DEPTH (FT.)
Figure 8. DISTRIBUTION OF TOTAL COD REMOVED
307
-------
low hydraulic rate was applied to the filter. The raw and recycle
rates were each 0.44 gpm/1 sq. ft. of cross sectional area. Second,
fruit waste from peach and fruit cocktail processing was being fed to
the filter. As indicated before, fruit waste is composed primarily
of simple sugars and acids which are readily metabolized by bacteria.
Therefore, a combination of low-flow and easily oxidizable compounds
probably account for the high removals shown by the results for samples
taken from the first one third filter depth- The remaining filter depth
took out more complex compounds which are not as readily removed from
the waste as the sugars and acids. Next season we will increase the
hydraulic rate by increasing the flow of raw and/or recycle to deter-
mine if better use can be made of the total filter depth.
Figure 9 is a schematic of the third trickling filter system to be
discussed. As indicated by the title, the unit has provisions for
forced aeration and elevation of the influent temperature. The res-
ervoir tank on the right is used to raise the temperature of the
influent. A plate coil was used in 1968 which did not work very well.
In 1969 this was replaced by direct steam injection which performed
better than the plate coil.
The treatment column on the left is 3-3/4 ft. in diameter and contains
7.5 ft. of plastic packing media. Distribution of the waste is by fixed
nozzles. Air is introduced at the bottom of the treatment column and
measured by a manometer-orifice plate system.
Table 3 is a summary of 1969 results and compares the effect of three
air flow rates on percent COD removals while operating at ambient tem-
perature.
Low air flow (100 SCFM) =15.9 percent COD removal
Medium air flow (200 SCFM) =22.9 percent COD removal
High air flow (300 SCFM) - 28.8 percent COD removal
As indicated above, there was an improvement in removals as the air
flow increased. Some dissolved oxygen determinations on the effluent
indicated inadequate air flow at the lower flow rates. We did not have
time to increase the air flow beyond 300 SCFM to determine if additional
removals would be achieved. This will be considered in a test program
for the 1970 season.
Some studies were made with this unit when operated in the thermophilic
temperature range of 120°F. The results were not as good as expected
but did indicate better removal rates than at ambient temperature.
While evaluating this phase of the program, we did experience diffi-
culties with the nutrient feed system and this may have contributed to
the poor performance of the filter. The use of elevated temperature
is another area that will be investigated in 1970 in an effort to
optimize the operation of the high rate trickling filter system.
308
-------
Air Out
Spray Nozzles
Packing Medium
Treatment Column
UJ
O
Air
Heater
Blower
Effluent
Reqycle Reservoir
Raw Waste From Two
55 Gal. Drums
V-^K
Heated
Influent
Vari-speed
Pump
Overflow
_Steam
Ejector
Grinder
V
\
\ \ Support Grating
Immersion Heater
Figure 9. SCHEMATIC DRAWING OF FORCED AERATION, CONTROLLED
TEMPERATURE TRICKLING FILTER. 1969
-------
Table 3. Effect of Air Flow Rates on COD Removal
at Ambient Temperature
Influent Air Flow
Low Medium High
3580 3000 (15.9)*
3400 2620 (22.9)*
4180 2980 (28.8)*
*Percent COD removed
310
-------
FOOD PROCESSING WASTE TREATMENT
BY SURFACE FILTRATION
by
Louis C. Glide*
A food processing plant can be in harmony with nature as witness the
outstanding results of the Campbell Soup Company, Paris,.Texas, plant
whereby 3.6 mgd per day of wastewater are treated by a unique system
accomplishing the equivalent of tertiary treatment. This is a new over-
land flow technique by which water sprayed on impervious or semi-pervious
soils becomes purified by flowing through vegetative litter on the soil's
surface and collects in terraces at the bottom of slopes. For over five
years the system has averaged 99 percent BOD reduction with major reduc-
tions of nutrients. The before and after picture of Figure 1 shows how
the badly eroded soil due to mismanagement of cotton farming was recon-
structed and rehabilitated to uniform slopes and terraces for treatment
purposes, thus becoming an outstanding example of soil conservation.
Detailed data were obtained on the system by conducting a one-year
comprehensive survey to study the interacting relationship of climate,
biological, and soil conditions. During this study, the Robert S. Kerr
Water Research Center obtained data on the water balance and treatment
efficiencies; Campbell Soup Company sponsored work by C. W. Thornthwaite
Associates who investigated the microclimate and agricultural aspects,
while North Texas State University evaluated the microbiological relation-
ships. Campbell Soup Company also provided project coordination and
on-site supervision of daily operations.
To accomplish this research, a pilot installation was set up where the
conditions of waste treatment were carefully controlled. On approximately
15 acres of a functioning system consisting of 500 acres, detailed bio-
logical studies identified and enumerated the various species of micro-
organisms. Microclimatic data were recorded to provide insight into the
microfauna flux as a function of temperature. Concurrently, the soil
chemical changes within the pilot plant were related to microbiology and
temperature. The effectiveness of this treatment was then evaluated
according to the rate of application and the soil and temperature condi-
tions. Evaluation of the agricultural economics of this disposal system
was also included.
The spray irrigation disposal system of Campbell Soup Company in Paris,
Texas, is one of fourQ' overland flow systems in the country which have
been installed on an area of impervious soil. Wastewater is applied
*Director-Environmental Engineering, Overall Project Coordinator, Campbell
Soup Company, 375 Memorial Avenue, Camden, New Jersey 08110
311
-------
Figure 1. Before and after photographs represent an excellent example
of soil conservation practice by adapting badly eroded non-
usable farmland to carefully rehabilitated uniform slopes
and terraces for treatment. The erosion gullies in some
places were deep enough and wide enough to hide complete-
ly a D-7 Bulldozer.
312
-------
along the top of a carefully prepared slope and becomes purified by flow-
ing through the vegetative litter on the soil surface and is collected
in a collection terrace at the bottom of the slope. Since the only
significant water loss is through evapotranspiration, it follows that this
technique is exceptionally well adapted for the collection and reuse of
treated water. It is also evident that the overland flow method of
treatment can be used in areas where an impervious soil or a high water
table would have ruled out the possibility of conventional spray irriga-
tion that normally relies upon infiltration.
The phases of the investigation included: (1) climatology, (2) biology,
(3) chemistry, and (4) agriculture. The climatological investigations
included an official weather station which was established to record
rainfall, maximum and minimum temperatures, and provide a continuous
record of relative humidity. The record was made of the soil heat flux
at selected levels below the surface. Net radiometers were used to record
the difference in energy balance between sprayed and unsprayed areas.
The biological program involved identification and enumeration of micro-
organisms (bacteria, fungi, algae, protozoa, etc.) which exist in the
spray fields and compared these populations with those in control fields.
Physiological activities of certain selected microorganisms were studied
to isolate and identify specific segments of the microbial population that
were responsible for the degradation of the organic substances in the
wastewater. Evaluations were also made of the effect upon the microbial
population produced by the use of certain insecticides.
Chemical analyses as conducted by the Robert S. Kerr mobile laboratory
facilities followed classical analytical tests normally associated with
pollution control programs.
Finally, the agricultural evaluation took into consideration the growth
index and hay quality along with the harvest rate for various disposal
potentials. This study involved the logistic problems of large-scale
harvest while continuing to operate the system for treatment purposes.
In order to evaluate the agricultural economics of the disposal system,
samples of fodder were collected during the growing season and analyzed
for animal nutrient value and palatability. Growth indices and evapo-
transpiration loss were evaluated according to microclimatic data.
Evaluations were made of the rate of harvest related to the gain or loss
in disposal potential.
DESCRIPTION OF PARIS SYSTEM(2)
The effluent arrives at the waste treatment building from two sources
within the factory. The first, containing grease from the cooking area,
passes through a gravity grease separator. It then joins the stream from
the vegetable washing area and the combined flow passes through 10-mesh
screens to remove large pieces of solids. From here the water is pumped
to the disposal field at high pressure.
313
-------
The treatment system consists of some 30,000 feet of underground force
main, 50,000 feet of portable aluminum irrigation pipe above ground, and
180,000 feet of control tubing in the ground operating 77 pneumatically
controlled valves. There are over 700 sprinklers in the system varying
in size from 14 gpm to 30 gpm, depending upon the particular treatment
capabilities of the individual slopes on which they are located.
The entire system is controlled automatically. Three clock timers similar
to golf course watering systems are connected electrically to each of the
five high pressure pumps. As the water level in the surge tank rises,
each pump is started in turn and energizes its own clock timer, which
automatically opens the valves in the field that are programmed to that
pump. When a sprinkling cycle for one group of lines has been completed,
the timer automatically advances to the next group and so on until all
sprinkler lines have received their predetermined allotment of water.
A plug board provides for convenient regrouping of lines as field condi-
tions or harvesting schedule requires.
Prior to initial operation, the heavy clay soil on the 500-acre tract was
remolded to form watersheds of two to ten acres each, depending upon the
lay of the land and the natural watercourses. Each watershed is from 200
to 300 feet wide and pitched toward a collecting terrace at the foot of
each slope. The land was then land-planed to smoothness and planted to
water-tolerant grasses. Near the top of the slope, irrigation sprinkler
lines run parallel to the collection terrace at the bottom. Wastewater
from the factory is discharged from the sprinklers and flows slowly
downhill in a thin sheet,becoming purified enroute through microbiological
activity. The purified water is then collected at the terrace and flows
to the receiving stream via a prepared waterway. The basic concept of
the system is diagrammed in Figure 2.
The grass which has been planted upon the slopes serves a multi-purpose
function. First, it protects the soil surface from erosion and retards
the flow of water across the slope. Next, it provides-^a protected habitat
for microorganisms and a vast surface area for the adsorption of impurities
contained in the water. Finally, when cut for hay it is a valuable cash
crop and provides an effective means for reclaiming plant nutrients which
are released in soluble form when the organic waste material decomposes.
DISCUSSION OF RESULTS OF ONE-YEAR RESEARCH PROGRAM(3)
The work in climatology centered around the establishment of a temperature-
moisture base which could be related to weather bureau records throughout
the world. To accomplish this, temperature, rainfall, and relative
humidity data were tabulated for several locations within and without the
disposal tract and at several levels above the ground. In addition, the
net radiation and soil heat flow were also recorded. The data showed a
remarkable similarity in all locations with the mean temperatures, differ-
ing only a degree or two. Furthermore, there was no consistent pattern
to indicate that one observation site was usually higher or lower than
any of the others, except that the U. S. Weather Bureau Station in the
City of Paris usually recorded higher temperatures than the duplicate
314
-------
SPRAY IRRIGATION
SURFACE FILTRATION
TREATMENT
Figure II
315
-------
equipment located in the disposal area. Unquestionably, this was due to
buildings near the city observation site. The same lack of pattern was
noted in the observations of relative humidity. Soil heat flux and net
radiation did show differences between wet and dry areas, but no more so
than would be expected. Thus, it can be said that the growth of grass
upon the disposal field responds to temperature and moisture in the same
manner as any hay field in the humid east and that any U. S. Weather
Bureau records can be used to predict growth rates anywhere in the east.
From the temperature and rainfall data, potential evapotranspiration was
calculated by means of the Thornthwaite formula. This, in turn, was
used in the hydrology studies. It was also used to plot growth rate
curves, predict hay harvest dates, and relate hay quality to growth rates.
During the season three cuttings of hay were made and reserved for later
palatability tests. In addition, weekly samples were collected for nutri-
tional and chemical analysis. The analysis results showed the Reed
Canary grass which had been grown on the disposal tract was of exception-
ally high nutritional value ranging up to 23 percent crude protein. The
mineral content was also high, nearly double the quantity found in other
good quality hay. Thus, harvest of hay provides an efficient method of
reclaiming part of the plant nutrients which are released when the vege-
table matter contained in the effluent decomposes. When the feeding tests
were conducted, cattle exhibited a definite preference for hay grown on
the disposal sites.
An attempt to relate potential evapotranspiration to soil tractionability(4-)
was disappointing because of the exceptionally high rainfall during 1968.
The rainfall also delayed the general harvest of the entire tract, result-
ing in a loss of quality.
The chemical investigation was primarily directed toward determining the
fate of phosphorus and nitrogen, but included much other work as well,
including BOD evaluation, a record of dissolved oxygen, pH and conductiv-
ity and an analysis of mineral salts contained in the soil. To accomplish
this, four watersheds were selected and isolated so that only wastewater
and rainwater would flow across the area. Several sampling stations were
located within the sprinkler pattern to collect raw waste and proportional
samplers were located at the outfall of the watersheds. The four water-
sheds represented two soil types, one each in areas which had operated
four years and one each in areas operated for two years. These areas
received similar quantities of water as all other watersheds in the
disposal system.
Early observations revealed that while BOD and nitrogen removal were high--
99 and 90 percent respectively—the removal of phosphorus was dis-
appointingly low, about 45 percent. A later change in the operating
procedure to provide a longer rest period between applications, with no
change in the total volume increased the phosphorus removal to nearly
90 percent without affecting the BOD or nitrogen removal.
316
-------
Table 1. Treatment Efficiency of the Overland Flow System
Mean Concentration, mg/1 Percent Removal
Parameter
Total suspended solids
Total organic carbon
Biochemical oxygen demand
Total phosphorus
Total nitrogen
Wastewater
263
264
616
7.6
17.4
Section
Effluent
16
23
9
4.3
2.8
Concentra tion
Basis
93.5
90.8
98.5
42.5
83.9
Mass
Basis
98.2
-
99.1
61.5
91.5
It is interesting to note that the disposal field acts as a buffering
agent and maintains the field effluent close to neutrality while the
factory effluent ranges widely over the pH scale. This is clearly shown
in Figure 3, which follows, showing the diurnal variation of pH for the
wastewater applied to the slopes and the field effluent.
The same condition is obtained for the diurnal variation of electrical
conductivity shown in Figure 4.
Analysis of groundwater samples showed that while mineral salts had
increased markedly in the old area, the total accumulation was not crit-
ical and there was some indication that the rate of increase might be
lessening.
H-type flumes were installed at the outfall of each watershed and equipped
with instruments to record the water depth. A complete water balance was
obtained from a known quantity of water being applied and measurement of
runoff, spray evaporation, evapotranspiration, and deep soil percolation.
The water data is believed accurate to plus or minus ten percent. Water
balance measurements accounted for 93 percent of the total liquid applied,
runoff measurements accounted for about 61 percent, and the remaining
21 percent percolated through the soil. Table 2 summarizes the hydro-
logical data for the area studied.
317
-------
12
11
10
9
8
7
6
5
4
3
2
1
£ Influent
O Effluent
A A
O O O
A
6 6 0060006006 o o 66
A A A A AA A A
A A
A A
O O O O O
8 9 10 11 12 1 2 3
AM N
456789
PM
10 11 12 123
N
4567
AM
Figure 3. DIURNAL PH VARIATION
318
-------
OIUCHAL MaiAT/OM Of
OF iMsre
° - SUMMER
X, - FALL
v -
DHJ8NAL MR/AT/OfS OF
o -
x -
• - SPtHt/6
)00-
9 H 3
TIME OF DAY
M
900
too
-IT A-
T/Mf
FIGURJS IV
-------
Table 2. Hydrology of the Overland Flow Treatment System
Water Applied TT «. A ^ j .c n
p Water Accounted for, Percent
Percent
Season
Spring
Summer
Fall
Winter
Spray
76
86
73
63
Rainfall
24
14
27
37
Evaporative
losses
23
28
13
10
Runoff
56
55
64
68
Deep-soil
percolation
21
17
23
22
Annual 74 26 18 61 21
Perhaps the most interesting single fact emerging from the study in biology
is the reason behind the continued high efficiency in winter(->). it has
long been recognized that a system would continue to purify water when
temperatures were near freezing, but since it is also known that the
respiration of microorganisms slows down as temperature decreases, it was
believed that the impurities were being adsorbed upon the surface of the
vegetation and held there until the weather warmed again. Biological
studies have shown that as temperature decreases the number of organisms
increases, thus maintaining a constant level of mass activity. This fact
places the importance of adsorption in a position of lesser significance.
The increase in organisms in winter is clearly shown in Figure 5.
Other work in biology sought to identify the groups of organisms which
were actually doing the work and revealed an extremely complex and chang-
ing ecology. Composite samples were taken at the top and bottom, of two
watersheds and also on the control lot. The data revealed that the sum
of the three most populous organisms represented less than 50 percent of
the total, indicating a very heterogeneous population. As might be
expected, the population density was greater at the top of the watershed
than at the bottom and both were greater than on the control, thus
reflecting the effect of a diminished food supply. This condition is
shown in Figure 6.
Since the type of organisms present in the control and those in the test
plot were essentially the same, it is inferred that the microflora
naturally present in the soil are those responsible for the biodegrada-
tion of the waste. However, other experiments showed that the oxidative
activity of samples inoculated with soil from the disposal field was far
greater than that of samples inoculated with soil from the control plot.
This would indicate either an evolutionary or a seeding process whereby
microorganisms specific for the effluent developed on the disposal site.
320
-------
TOTAL M/CGO&/AL
OM COUT8OL AAJD
LOT
resr tors
600
\
321
-------
TOTAL M/caOB/AL £>Of>ULAT/OM OM TOP
AMD Borrow OF" SLOf>£~ 0/V T^ST LOTS
A
o
T0f>
BOTTOM
(&io)
JOG T
APR MAY Jt/A/f- JULY AUG SfPT OC7 A/OV OfC J4V f£3 MA A APR MAY
322
-------
One might speculate that the time required for the evolutionary process
is one reason for the apparent greater capacity of an older disposal
system. A similar speculation suggests that the maturing of the system
might be hastened by seeding it with specific organisms.
It was also shown that some soils will support microorganisms much better
than others in the laboratory. The significance of this on a complete
field system is not known since no individual area or slope could be
identified as being less efficient due to soil character or structure.
The overland flow system has the outstanding capability of handling shock
loads on the one hand and periods of long shutdown and immediate start-up
on the other, producing excellent results in either case. In addition,
wide variations in effluent character, such as occur during night cleanup,
produce no adverse effects. Also, if a single terraced slope is accident-
ally overloaded or suffers a mechanical failure, the effluent treatment
continues in the long terraces and waterways before the effluent reaches
the receiving stream. In other words, a great deal of "fail-safe"
capacity is naturally built into the system.
CONCLUSIONS RELATED TO DESIGN PARAMETERS
The Paris system was designed with the length of downhill slope from
200 feet to 300 feet. Studies in 1968 showed that 175 feet would give
effective purification,and there is strong suspicion that 150 feet is
sufficient at this site, provided the slope is not too great. Basically,
the downslope area requirement is 50 feet beyond the perimeter of the
sprinklers.
The pitch of slopes in Paris ranges from less than one percent to more
than twelve percent. The investigation led to the conclusion that a flat
slope encouraged puddling and subsequently anaerobic conditions, while
the retention time on a steep slope was insufficient for complete degrada-
tion at normal application rates. This established a design criterion of
no more than six percent, but not less than two percent for any slope.
In Paris the design was based upon an application rate of 0.25 inch per
day in winter and 0.50 inch per day in summer based upon an estimated
wetted area coverage of approximately 75 percent. As it turned out, the
actual wetted area was considerably less than the estimate, but the
disposal capacity per wetted acre was much greater. The greater hydrau-
lic load capacity combined with more efficient land utilization results
in a more compact system requiring less acreage. The application to
wetted watershed in the test areas was normally 0.6 inch per day or 3.0
inches for a five-day week. During the summer of 1968, the duration of
the sprinkling cycle was increased from 6 to 8 hours per day, while hay
harvest was in progress.
Reed Canary grass will yield a large quantity of exceptionally high
quality hay which, according to nutritional analysis, approximates the
value of first quality alfalfa. The number of harvestswhich may be under-
taken during the growing season has not been completely established but
323
-------
certainly one and probably two harvests are feasible. The study further
established that late harvest of a third crop resulted in a poor quality
crop because of over-maturity. The value of the hay, however, was about
8 percent of the operating cost of the system.
The spray irrigation technique of wastewater disposal is now being used
widely throughout the food and wood products industry. Most of these
installations achieve a high degree of water purification, but the use
of this technique is usually limited to soil areas with high infiltra-
tion capacity, and thus the purified water is usually lost to the original
user. The system at Paris, Texas, is a notable departure from the usual
spray irrigation concept, since it employs the overland flow technique
whereby water is applied to impervious soil and the purified effluent is
collected at the foot of a vegetative slope. Thus, consideration can be
given to the collection and reuse of water from such a system.
This technique has been developed to the point that an exceptionally high
degree of waste purification has been demonstrated. The system at Paris,
Texas, has averaged in excess of 99 percent reduction in BOD for over
five years. This technique now extends the highly efficient performance
to areas of impermeable soils where its use was formerly considered
impossible.
324
-------
ACKNOWLEDGMENTS
The major portion of this paper has been abstracted from a com-
pendium report entitled, "An Evaluation of Cannery Waste Disposal
by Overland Flow Spray Irrigation" which is published in Publica-
tions in Climatology, Volume XXII, No. 2, and is available from
the Laboratory of Climatology, R. D. 1, Elmer, New Jersey.
Individuals who have made a major contribution to the research
effort are:
Campbell Soup Company
L. C. Gilde, Director-Environmental Engineering,
Overall Project Coordinator
C. H. Neeley, Manager-Plant Services (Paris Plant)
Field Project Coordinator
C. W. Thornthwaite Associates
Principal Contractor for Research
D. M. Parmelee, Vice President, CWTA
Coordinator of Contracted Research
J. R. Mather, Ph.D., President, CWTA
Consultant in Climatology
J. P. Ford, Agricultural Engineer, CWTA
Field Scientist
North Texas State University
A. S. Kester, Ph.D., Assistant Professor
Chief Investigator, Microbiological
Population Studies
G. R. Vela, Ph.D., Assistant Professor
Chief Investigator, Microflora Activity Studies
Robert S. Kerr Water Research Center, U. S. Department of Interior
J. P. Law, Jr., Ph.D., Research Soil Scientist
Project Leader
R. E. Thomas, Research Soil Scientist
In charge of Field Operations
L. H. Myers, Research Chemist
In charge of Analytical Operations
Special appreciation is extended to Mr. D. M. Parmelee for his assis-
tance in the preparation of this paper.
325
-------
REFERENCES
1. Hill, R. D., Bendixen, T. W., DuByne, F. T., and Robeck, G. C.,
"Cannery Waste Treatment by Spray Irrigation - Runoff." Presented
at the Water Pollution Control Federation convention, September 1966.
2. Gilde, L. C., and Parmelee, D. M., "Natural Land Filtration Treatment
System, Campbell Soup Company, Paris, Texas." Presented at the 49th
Texas Water and Sewage Works Associations Short School, Texas A&M
University, College Station, Texas.
3. Gilde, L. C., Kester, A. S., Parmelee, D. M., and Thomas, R. E.,
"Research Evaluation for Overland Flow-Spray Irrigation Treatment
of Cannery Wastes." Presented at ISA-AID Symposium, May 5-7, 1969,
New Orleans, La.
4. Thornthwaite, C. W., Mather, J. R., and Krimgold, D. B., "Estimating
Soil Tractionability from Climatic Data." Publications in Climatology,
Laboratory of Climatology, vol. VII, No. 3. (1954).
5. Vela, G. R., 1969 Spring Meeting, Texas Section, American Society
for Microbiology in Austin, Texas.
326
-------
BIOLOGICAL TREATMENT OF FOOD PROCESSING WASTES
by
Perry L. McCarty*
The principles of biological waste treatment have been developed to a
high level of understanding over the past twenty years. The papers
presented during the first two days of the National Symposium on Food
Processing Wastes have shown that these principles now are being used
successfully for the solution of practical waste treatment problems.
The success which is being achieved clearly demonstrates that the
relatively few years and few dollars spent by the Federal government for
basic research on biological processes had paid-off decidedly.
No longer is it necessary to consider each different waste as being
highly unique and requiring years of empirical trial and error research
to reach a practical solution to a waste treatment problem. No longer
is it necessary to continue polluting streams with organic waste because
of the absence of suitable methods for removing organic materials from
waste effluents. Full scale treatment systems can be designed, can be
built, can be operated, and can be depended upon to achieve a high
efficiency of treatment. This can usually be done now with a relatively
straight-forward waste characterization and with only limited laboratory
studies.
This paper is a summary of the basic principles which have been applied
and discussed during the first two days of this Symposium. During this
time, twelve papers primarily concerned with the use of biological
processes for treatment of food processing wastes were presented. Eleven
of the papers dealt primarily with the use of aerobic systems, with major
emphasis on the activated sludge process. The anaerobic treatment process,
however, was discussed in only three of the papers. The later process
has some important advantages for treatment of food processing wastes, but
has not been widely used to date. This is partly due to a former lack of
understanding of the process, but this deficiency has largely been erased
during the past decade. Additional emphasis' on the anaerobic process will
be given in this paper.
*Professor of Environmental Engineering, Civil Engineering Department,
Stanford University, Stanford, California.
327
-------
PRINCIPLES OF BIOLOGICAL TREATMENT
The engineer can consider the design of a waste treatment system
similar to the way in which he considers the design of a structure. He
determines first the stresses which can be placed upon his materials
before failure is reached. He does not design for this failure point,
but applies a safety factor. The magnitude of the safety factor is based
upon judgements which consider possible overloads on the system, sudden
and localized stresses, temperature effects, permissible strains on the
system, and allowance for unknown or non-calculable risks.
The ultimate stress which can be placed upon a biological treatment
system before failure is related to the maximum reproduction rate of
the microorganisms involved. If the system is stressed to the point at
which the microorganisms are being removed from the system faster than
they can reproduce themselves, then failure will result. This is closely
related to another phenomena, the maximum rate at which microorganisms
can consume food. If efforts are made to force them to consume more food
than their physiological ability allows, then the process will fail. The
failure point for biological systems can be evaluated from a consideration
of the kinetics of bacterial growth and waste utilization.
Process Kinetics
The growth of microorganisms as a function of time after they are mixed
with organic wastes is generally (2) estimated from the following
equation:
dX/dt = Y(dF/dt) - bX (1)
where *
dX/dt = Net growth rate of microorganisms in mass
per unit time,
dF/dt = Rate of waste utilization in mass per unit time,
X = Mass of microorganisms present,
Y = Growth yield coefficient in mass or organisms
produced per unit mass of waste consumed,
b = Organism decay coefficient, time~l.
Equation 1 states that the rate of growth of microorganisms is proportional
to the rate of waste utilization (dF/dt), minus the loss in microbial mass
due to decay (bX). In biological processes the mass of active micro-
organisms is frequently assumed to be equal to the volatile suspended
solids in the system. However, the actual mass of active organisms may
represent only a small fraction of this, the exact proportion depending
upon the quantity of other organic suspended solids present.
The rate at which organic matter is utilized by microorganisms is a
function of the concentration of a limiting nutrient as illustrated in
Figure 1. In waste treatment, the organic waste itself is generally the
limiting nutrient. However, with food processing wastes required inorganic
nutrients such as nitrogen and phosphorus are often limiting and should be
added for optimum rates of treatment and for good removal efficiency.
328
-------
dF/dt
u>
to
X JL._
SUBSTRATE CONCENTRATION, S
Figure 1. EFFECT OF WASTE CONCENTRATION, S, ON
SUBSTRATE REMOVAL VELOCITY (dF/dt/X)
-------
Monod (1) presented a relationship similar to the following to describe
the effect of waste concentration on rate of waste utilization as
illustrated in Figure 1:
dF m kSX (2)
dt K + S
s
where: fc _ M,^^^ rate of waste utilization with high waste
concentration ,
K - Waste concentration at which the rate of waste
Q
utilization is 1/2 the maximum rate,
S » Concentration of waste surrounding the microorganisms .
Equation 2 can be rearranged to give the following:
_ kS
~ K + S
s
(3)
The ratio (dF/dt)/X is commonly called the substrate removal velocity or
the food to microorganism ratio. It represents the rate of waste
utilization by a unit mass of microorganisms. The maximum removal velocity
as illustrated in Figure 1 is equal to k and is achieved at high waste
concentrations. This represents the upper limit for any biological process
above which failure results. Much lower removal velocities than this
maximum must be maintained for dependability and efficiency. A low removal
velocity is obtained by adding a small amount of waste to a large reactor
containing a high concentration of microorganisms.
If Equations 1 and 2 are combined, the rate of growth of microorganisms
as a function of waste concentration can be obtained:
YkS
X K + S
s
where (dX/dt)/X is the net specific growth rate of the microorganisms in
units of time" . The maximum growth rate is obtained at high waste
concentration and equals (Yk - b) .
A schematic diagram for a typical biological process is illustrated in
Figure 2. A waste with flow rate Q and waste concentration S0 is treated
in a biological reactor with volume V. In the reactor, a mass of micro-
organisms, X, metabolize the waste and reduce its concentration to S,
after which the microorganisms and remaining waste pass to a settling
chamber where the microorganisms are removed from the waste stream.
Considering that the settling tank is small in volume compared to the
biological reactor and that no biological action takes place there, the
concentration of waste in the treated effluent, S would be the same as
in the reactor. The settled microorganisms are recycled and mixed with
new incoming waste for treatment. If waste were continuously added to
this system, the microorganisms would grow in mass until the system could
330
-------
U)
Q,S<
RECYCLE
' Q.S
dX/dt
Figure 2. SCHEMATIC DIAGRAM OF A TYPICAL COMPLETELY-MIXED
BIOLOGICAL PROCESS WITH RECYCLE
-------
no longer hold them. At this point, the excess microorganisms produced
each day (dX/dt) would have to be removed from the system or else they
would fill the settling tank and would pass out with the treated effluent,
deteriorating its quality.
Continuous operation with deliberate wasting of a fixed quantity of micro-
organisms would soon lead to steady-state conditions of waste stabilization
and microorganism growth. The rate of waste treatment, dF/dt, is equal to
(SQ-S)Q, and the percent efficiency of waste treatment is 100(SO-S)/S0. For
high efficiencies of treatment, the concentration of waste in the effluent,
S, must be small. This concentration based on Equation 3 is a function of
the substrate removal velocity, or based on Equation 4 is a function of net
specific growth rate. The reciprocal of growth rate is more convenient for
use and gives the solids retention time (8C) , or sludge age, which represents
the average retention time of microorganisms in the system:
6C = X/(dX/dt) (5)
The value for 6C is obtained by a suspended solids balance on the overall
treatment system:
_ suspended solids in system _
c suspended, solids removed from system per day (.6)
The suspended solids removed per day include both the solids taken from
the system deliberately as well as those which pass out unintentionally
with the effluent. Figure 3 illustrates the relationship between effluent
waste concentration and 6C. As 6C is decreased, the concentration of
waste in the effluent increases. There is a minimum 0C below which the
treatment process will fail and is reached when the organisms are removed
from the system faster than they can reproduce themselves and "wash-out"
or loss of microorganisms responsible for treatment results. This
minimum solids retention time, 6m, can be approximated by the following
formula, derived from Equations 4 and 5, by considering b to be negligible
and S to be much larger than Kg:
Table 1 summarizes the usual ranges of the kinetic coefficients for both
aerobic and anaerobic processes (2) (3). The maximum rate of waste utili-
zation, k, varies with temperature and within the range of biologically
acceptable temperatures, about doubles for each 10° C rise. In aerobic
processes, a large portion of the organic waste is converted to biological
cells so that Y is high. Here, 6*? is only a fraction of a day. In
anaerobic processes, on the other hand, only a fraction of energy contained
in organic matter is available for microbiological growth so that Y is
quite small. For this case 6° is several days.
332
-------
to
u>
CONCENTRATION
SOLIDS RETENTION TIME, 0C
'Figure 3. RELATION BETWEEN SOLIDS RETENTION TIME, EFFLUENT
WASTE CONCENTRATION AND PROCESS EFFICIENCY
-------
Table 1. Usual Range of Kinetic Coefficients For Aerobic and
Anaerobic Treatment Processes
Coefficient
Y
b
K
s
k
em
c
Units
Ib vol suspended solids/lb BOD
Li
day
mg/1
Ib BOD /day/lb vol suspended solids
LI
days
Aerobic
0.3-0.4
0.02-0.06
1-50
4-24
0.1-0.3
Anaerobic
0.03-0.15*
0.01-0.04
20-300
4-20
2-6
*For anaerobic systems Y is 0.03-0.05 for protein and fatty acid wastes
and 0.14-0.16 for carbohydrate wastes.
Both substrate removal velocity, (dF/dt)/X and solids retention time, 6C.
can be used as basic parameters in the design and control of all biological
treatment systems (2). The former has received most attention for the
aerobic treatment of organic wastes, while the latter is more widely used
for anaerobic processes and for the control of nitrification in aerobic
processes. The solids retention time has certain distinct advantages for
biological processes in general. With this parameter, the concentration
of active microorganisms present does not have to be known. For example
if it were desired to maintain a 9C of 5 days, this could be done by
removing 20 percent of the total suspended solids from the system every
day. By doing this 20 percent of the active microorganisms are also
removed.
To use the concept of substrate removal velocity, however, the quantity
of active microorganisms in the system must be known or estimated. A fair
approximation to the active microorganism concentration is given by the
mixed liquor suspended solids concentration in activated sludge systems so
that the parameter can be reasonably well applied. However, in anaerobic
processes or with nitrification, the concentration of active organisms is
very difficult to determine so that the loading velocity concept is not
easily applied.
Excess Sludge Production
A necessary by-product of all biological unit processes is excess sludge.
Such sludge must be removed periodically from the treatment process and
disposed by some means. This is frequently one of the most difficult and
expensive operations associated with biological treatment. The quantity
of sludge will vary with the type of biological treatment, whether aerobic
334
-------
or anaerobic, and with the method of operation, whether at long 6 or
short 0C. Excess sludge is composed of two major fractions, (1) the
excess microorganisms produced during waste treatment, and (2) the
relatively stable organic suspended solids contained in the raw waste
which are not altered by the treatment process, but which settle in tanks
and are removed along with the excess biological cells. Excess sludge
may be estimated as follows:
dX /dt = dX/dt + dX./dt (8)
s i
where :
dX /dt = Excess pounds of volatile sludge solids
produced per day,
dX/dt = Excess microorganisms produced per day,
dX./dt = Stable organic suspended solids contributed by raw
waste per day.
The excess microorganisms (dX/dt) can be estimated from Equation 1. The
relation between excess microorganisms and 6m or substrate velocity can
be found by combinations of Equations 1 and 5:
dX dF dX _ a(dF/dt)
dt~Y IF ~ b ec dt ~ i + bec
or,
= (Y - b)X (10)
The value for dX^/dt varies with the type of waste. In raw domestic
wastewater, about 30 to 60 percent of the volatile suspended solids are
relatively stable. Thus, this fraction would represent the solids to
consider in evaluating the stable volatile suspended solids for this
waste. Food processing wastes on the other hand, differ considerably and
dX-^/dt may vary from zero for wastes containing no suspended solids, to
a very large number for wastes with large quantities of suspended solids.
Nutrient Requirements
The proper operation of biological waste treatment processes depends upon
the presence of sufficient inorganic nutrients for biological growth.
With food processing wastes, the major nutrients which may be lacking are
nitrogen and phosphorus. The quantity of each required can be estimated
from the weight of biological cells produced per day and the proportion
of nitrogen and phosphorus which they contain. The nitrogen and phosphorus
content of bacterial cells is relatively constant for all biological
processes and equals about 12 percent for nitrogen and 2 percent for
phosphorus. Thus, the requirements for nitrogen and phosphorus can be
evaluated as follows:
N = 0.12 (dX/dt) (11)
P = 0.02 (dX/dt) (12)
335
-------
where: N _ pounds of nitrogen/day required
P = pounds of phosphorus/day required
dX/dt = excess microorganisms produced per day in pounds
of volatile solids.
Air Requirements and Methane Production
The basic equations presented can also be used to evaluate the quantity
of air required for an aerobic biological process or to evaluate the
quantity of methane produced in an anaerobic process. When waste is
consumed biologically in an aerobic process, a portion is oxidized,
thus consuming oxygen. Oxygen is also consumed during decay of micro-
organisms under aerobic conditions. From these considerations and from
a materials balance using Equation 1, the oxygen required in waste
oxidation can be determined from the following:
C = dF/dt - n(dX/dt)
or,
C = (1 - nY)(dF/dt) + nbX (13)
where C is the pounds of oxygen used per day and n is a conversion factor
equal to the oxygen required to oxidize a unit mass of cells to inorganic
end products. The value for n is generally close to 1.4 when Y is
expressed in pounds of volatile suspended solids per pound of waste COD,
X in pounds of volatile suspended solids, and dF/dt in pounds of COD
consumed per day.
The oxygen requirement can also be expressed in terms of the solids
retention time as follows, using units similar to those above:
c = n _ nY \ dF_ ,14v
C C1 1 + b8c} dt U^;
During anaerobic treatment, the quantity of consumed waste not converted
to biological solids is converted to methane gas. Equations similar to
those for oxygen consumption in aerobic processes can be used for pre-
diction of methane production in anaerobic processes, as follows:
M = m[(l - nY) (dF/dt) + nbX)] (15)
Here, M is cubic feet of methane produced per day and m is a conversion
factor which equals the methane equivalent of organic matter. For
standard conditions of temperature and pressure m equals 5.6 cubic feet
of methane per pound of COD (17). The other units in Equation 15 are
the same as for Equations 13 and 14. Methane production as a function
of solids retention time is as follows:
"-•a-rvatf f (16>
The units in Equation 16 are the same as for Equation 15.
336
-------
Safety Factor for Design
The failure point for aerobic biological treatment occurs at a solids
retention time, 6™, of about 0.2 days and for anaerobic systems, 6|?
is about 3 days. In order to prevent process failure, longer solids
retention times must be used in practice (2) . Processes which are
considered conventional and have stable operation and high efficiency
of treatment normally are operated with solids retention times 10 to
70 times greater than 0m. Processes which are high rate, on the other
hand operate closer to the failure point, are in general less efficient,
and require closer control. They generally employ safety factors in the
range of 3 to 20. Processes designed for operation with little supervision
or maintenance or designed for highly stable operation with low excess
biological solids production generally have very high safety factors of
50 to 100 or more. These safety factors apply whether the process is
aerobic, anaerobic, or designed for autotrophic reactions such as
nitrification.
Processes designed and operated on the basis of substrate removal
velocity can also be considered in terms of a safety factor. The usual
maximum removal velocities for both aerobic and anaerobic systems based
upon active organisms is 4 to 24 pounds of COD per day per pound of
organisms. By applying the safety factors discussed above to reduce the
applied removal velocities, typical values for high rate, conventional,
and low rate systems can be obtained.
AEROBIC TREATMENT OF FOOD PROCESSING WASTES
Most of the papers presented at this National Symposium on Food Processing
Wastes have been concerned with use of aerobic processes for treatment of
food processing wastes. A summary of the various results reported will
be made and will be compared with parameters outlined under the previous
sections on principles of biological treatment.
Biological Growth Yield
The biological growth yield coefficient, Y, gives the portion of waste
converted to biological solids. This term is important as it indicates
both the quantity of biological solids which require disposal and the
quantity of inorganic nutrients such as nitrogen and phosphorus required
for treatment. As indicated in Table 1, the value for Y for aerobic
systems in general is about 0.3 to 0.4 pounds per pound of BODL or COD
consumed, which would equal about 0.4 to 0.6 pounds in terms of BOD5
consumption values.
Esvelt (8) found values for Y of 0.46 - 0.57 for peach, pear, and apple
wastes. Church (6) found a similar value even when fungi, rather than
bacteria were used for treatment, suggesting the general validity of this
range for all types of aerobic processes.
337
-------
Guttormsen and Carlson (4) reported a high value for Y of about 1 pound
of suspended solids per pound of BOD for potato processing wastes, but
this was a combined value including not only biological solids, but
digestible suspended solids in the influent.
The organism decay rate, b, is normally low, about 0.02 to 0.06 per day.
This means that if left with no food, the microbial mass would decrease
at a rate of 2 to 6 percent per day. Esvelt (8) reported an unusually
high value of about 11 percent per day of 20° C for fruit packing wastes,
but this dropped to the more normal lower values at reduced temperatures.
At low loadings, the cell decay plays a significant part in reducing the
total quantity of biological solids for disposal.
Substrate Removal Velocities
Maximum substrate removal velocities are about 4 to 24 pounds of BOI^ per
day per pound of microbial mass, depending upon temperature. Streebin (9)
was the only individual reporting operating values near this maximum with
a loading of 4 to 6 pounds of COD per pound of mixed liquor volatile
solids in the first stage of a two stage process for cannery wastes. How-
ever, without a second stage operating at a much lower loading and without
the recycle scheme that was used, the process no doubt would have failed.
Most of the food processing waste treatment systems described were operated
with conventional loadings of about 0.5 pounds of BOD^ per pound of mixed
liquor volatile solids or less. Richter and Pailthorp (5) reported values
of 0.1 to 0.4 for potato processing waste treatment, Graham and Filbert
(7) indicated values of 0.03 to 0.42 for canning wastes combined with
domestic wastes, Esvelt (8) stated 0.4 was used for fruit processing
wastes, Jones (10) found values ranging from 0.014 to 0.38 with an average
of 0.14 were satisfactory for citrus processing effluents, and Palmer (12)
used comparable values ranging from 0.03 to 0.47 for cannery waste treat-
ment. In almost all cases efficiencies of BOD removal of 90 percent or
better were obtained.
Several of the above values are much below the normal value of 0.5
pounds of BOD per day per pound of solids and are typical of the low-loading
values used for extended aeration systems. The lower loadings were
generally used to reduce the biological solids production. This was
indicated by Richter and Pailthorp (5), Graham and Filbert (7), Esvelt (8),
Streebin (9), and Palmer (12).
Microbial Concentration
A significant conclusion which can be drawn from the concept of substrate
removal velocity is that more waste can be treated per day in a given
reactor by increasing the concentration of microorganisms or mixed liquor
volatile solids. For a given removal velocity, the greater the mass of
organisms the greater will be the quantity of waste which can be treated.
This was well illustrated by Richter and Pailthorp (5) when they increased
338
-------
the mixed liquor suspended solids concentration to 4000 mg/1 and by
Palmer (12) who used mixed liquor suspended solids concentrations as
high as 10,000 mg/1. In both cases volumetric loadings of 100 to 200
pounds of BOD per thousand cubic feet of aeration volume per day were
obtained and with a high efficiency of treatment. Difficulties with
maintaining such high concentrations, however, were emphasized. This is
frequently true when attempts are made to maintain mixed liquor solids
concentrations greater than about 2500 mg/1.
Aerated Lagoons
The general biological principles which apply to activated sludge treat-
ment also apply to treatment in aerated lagoons. In a flow through
system of this type without recirculation of active biological solids, the
solids retention time is about equal to the hydraulic detention time. Thus,
for conventional loadings using safety factors of 10 to 70, and with 6m for
aerobic systems of 0.2 days, hydraulic detention times of 2 to 14 days
are indicated.
Since aerated lagoons are not usually designed for removal of the micro-
organisms developed during treatment, the effluent generally contains a
high concentration of microbial solids which consume oxygen in the
stream or BOD bottle during organism decay. For this reason, efficiency
of treatment in aerated lagoons is seldom greater than 70 to 80 percent.
Esvelt (8), for example found an efficiency of about 70 percent BOD
removal for treatment of fruit processing wastes in aerated lagoons with
a detention time of 5 to 10 days. Richter and Pailthorp (5) and Dostal
and Burm (15) indicated similar experience with other fruit processing
wastes treated in aerated lagoons.
ANAEROBIC TREATMENT OF FOOD PROCESSING WASTES
Food processing wastes generally have characteristics which should make
them susceptible to anaerobic treatment. They generally have BOD
concentrations within the desirable range of 1000 to 4000 mg/1. Many of
the food processing wastes (exceptions are seafood (11) and meat processing
wastes) have nitrogen and phosphorus concentrations which are low for
aerobic treatment, but sufficient for the anaerobic process. A significant
advantage of anaerobic treatment is that microbial production is low. Only
about 10 to 20 percent of the BODr is converted to cells with carbohydrates,
and only about 2 to 6 percent with protein or fatty acid containing wastes.
Another characteristic of food processing wastes as pointed out by Palmer
(12), Brenton and Fischer (13), Soderquist and Williamson (11), and others
is the seasonal nature of the wastes, requiring a treatment process which
can be brought to full operation within a very short period of time.
339
-------
The anaerobic process has not been widely used for food processing waste
treatment. Most anaerobic systems have been deep lagoons with results
as described by Brenton and Fischer (13), Dostal and Burm (15), and
Guttormsen and Carlson (4). In order to prevent washout of the sensitive
methane bacteria, anaerobic lagoons should have detention times of 10
days or longer. Even then, contact with air would tend to limit the
effectiveness of the anaerobic microflora. Also, odorous end products
such as hydrogen sulfide may limit the application of an open lagoon
system in urban areas. Dostal and Burm (15) mentioned the use of a
floating styrofoam cover on an anaerobic pond for treatment of potato
processing wastes which may serve to overcome some of the problems with
anaerobic lagoons.
Anaerobic treatment of food processing wastes using high concentrations
of microorganisms as in the activated sludge or trickling filter systems
has been limited. The main usage has been for wastes from the meat
packing and alcoholic beverage industries (17). Such systems, however,
offer much promise for treating wastes with concentrations and temperatures
typical of food processing wastes. The anaerobic process has been studied
extensively over the past 15 years from a fundamental standpoint and is
ready for wider usage. The process is not suitable for all wastes, but
its limitations are well known. For those wastes which are susceptible
to the process, the advantages are significant. The potential savings
to the food processing industry should be sufficient to overcome the
reluctance to assume the risk involved in the trial of a new method.
A relatively new anaerobic process called the anaerobic filter has been
developed and has had several years of laboratory experimentation. At
this time it appears to offer the most promise for treatment of wastes
with BOD ranging from 500 to 6000 mg/1 and at temperatures of 15° to
35° C.
The Anaerobic Filter
In Figure 4 is shown a schematic diagram of an anaerobic filter. Its
operation has been described in detail previously (18). The filter
consists of a bed of submerged media through which waste flows in an
upward direction. The anaerobic microflora grow within the bed and either
cling to the media or grow in the voids between. The filter has a large
capacity for retaining microorganisms so that concentrations up to
20,000 mg/1 are typical after a period of operation. Only a few types of
media have been explored and one-inch rounded stone appears to be one of
the best for use.
The effect of detention time on efficiency of treatment is illustrated in
Figure 5 for operation at 25° C. With waste concentrations about 750 mg/1
the efficiency of waste removal is a function of detention time. Experi-
ments at other temperatures have indicated that effective treatment can
be obtained at temperatures as low as 15° C, although the rates are much
lower than at 25° C.
340
-------
GAS
CONCENTRATED
FEED -
TAP
WATER
^
o
PUMP
PUMP
]
DISPERSION
RING
FILTER
STONE
FILTER
Figure 4. SCHEMATIC DIAGRAM OF LABORATORY ANAEROBIC FILTER
-------
THEORETICAL HYDRAULIC DETENTION TIME-HOURS
72 36 18 9 4.5
i -
t J
1 \J\J
80
55
1
-J 60
§
O
z
111
"40
O
0
CD
20
0
C
Figur
^
T
d£
r^
1
•
«»^_
r^-\
WASTE COD -mg/l
V.A. P-C
7!
ISC
30(
60(
I20C
JO D
)0 — o
)0 A
)0 — v
)0 — 0
•
A
"\
i
^L
cS
) O.I 0.2 0
(THEORETICAL HYDRAULIC DETENTION TIME -HOURS)"
e 5. RELATION BETWEEN DETENTION TIME AND EFFICIENCY OF WASTE REMOVAL FOR
3
DIFFERENT WASTE CONCENTRATIONS USING THE ANAEROBIC FILTER. V. A. repre-
sents an acetate-propionate containing waste and P-C represents a protein-
carbohydrate containing waste (Young, J.C. and McCarty, P.L., "The Anaerobic
Filter for Waste Treatment," Tech. Report No. 87, Civil Engineering Dept
Stanford University, March 1968).
-------
The anaerobic filter is of simple design and requires no sludge or effluent
recycle to maintain a high treatment efficiency. Unlike aerobic treatment
processes, the anaerobic filter results in exceptionally low solids
production. With low solids-producing wastes such as fatty acids and
proteins, the filter can be operated for well over a year with no need for
solids wasting. High carbohydrate containing wastes, however, produce
more biological sludge and sludge wasting a few times a year may be
necessary.
The filter responds rapidly to changing loads. It has an added significant
advantage for food processing wastes in that it operates well on a periodic
basis. Laboratory filters have been started immediately after receiving no
waste for periods up to a year. The efficiency of treatment may be a
little below normal for the first few days after a long period with no
feed, but the normal efficiency is regained rapidly.
Their are certain limitations for the anaerobic filter which must be
recognized. The efficiency of treatment may be less than shown in
Figure 5 for wastes with concentrations less than about 500 mg/1. The
filter operated best on soluble wastes which will not tend to clog the
void spaces. Ways to eliminate this problem are currently under study.
Control of pH is also necessary as the methane bacteria are sensitive
to large variations. Wastes must have an alkalinity of at least 500 mg/1
and preferably 2000 mg/1 to avoid a significant pH drop during treatment.
A high sulfate concentration presents some special problems. With sulfate
concentration about 200 mg/1, the hydrogen sulfide produced under
anaerobic conditions from sulfate reduction may be toxic to the methane
forming bacteria. In addition, the hydrogen sulfide may produce an
objectionable odor. Finally, the initial start-up of an anaerobic filter
must be done carefully in order to establish the proper microflora.
With wastes which are susceptible to treatment by the anaerobic filter,
significant advantages can be realized. Power requirements, maintenance,
and excess biological solids production are very low. The filter operates
exceptionally well under seasonal conditions of operation. Methane gas is
a useful by-product. Many of the food processing wastes appear ideal for
treatment by the anaerobic filter. Results of pilot plant studies for
treatment of potato processing wastes by the anaerobic filter were reported
by Guttormsen and Carlson (4). More studies on this scale or larger are
needed to determine the overall feasibility of the process.
OTHER ASPECTS OF FOOD PROCESSING WASTE TREATMENT
The need for waste treatment represents somewhat of a failure, a failure
to successfully use all components of the processed food. A desirable
goal is complete utilization of all by-products and extensive reuse of
process waters so that waste disposal is not'necessary.
343
-------
In accordance with the goal of by-product recovery, Church (6) reported
on the use of fungi for waste treatment hoping to recover the mycelium
for use as an animal food supplement. More basic research with fungi is
desirable as they have long been recognized to be capable of consuming
wastes that are not readily digestible by the bacteria normally used in
waste treatment.
The use of waste activated sludge from food processing for animal feed
has been suggested and would help considerably in the solids disposal
problem.
Jones (10) indicated efforts to treat the effluent from citrus processing
wastes to such a degree that the water could be reused. Although he felt
the reprocessed water presently was not as inexpensive as raw water, the
economics may change when effluent requirements become more rigid.
Some wastes from the food processing industry are not susceptible to
biological treatment and chemical and/or physical methods must be used.
Mercer, Maagdenberg, and Rails (14) discussed the treatment of olive
brine which was perhaps too saline for biological treatment. They used
activated carbon to remove dissolved organics so that the brine could be
reused rather than discharged to waterways. Technology for reclaiming
spent carbon has advanced rapidly so that chemical and physical methods
are reaching a competitive point with biological methods, especially
when both organic and inorganic nutrient removal are being emphasized.
Gilde(16) suggested another approach in which the ground surface was
used as a biological system. Waste was allowed to flow over the surface
and was consumed by attached organisms. Nutrients were also removed
and converted to hay, a usable by-product.
Another desirable direction for the food processing industry is to take
the processing step directly to the field, or as Soderquist, Williamson,
and Blanton (11) suggested, to the sea, and combine it with the harvesting
step. Thus, the residuals from processing could be returned directly to
the soil or the ocean to add to their enrichment. As the cost of waste
disposal increases, initially unlikely solutions such as these may
suddenly become attractive.
SUMMARY
The fundamentals of aerobic biological treatment are well understood and
application in full scale has been sufficiently demonstrated and reported
at this meeting and elsewhere. The fundamentals of anaerobic treatment
are also well understood, but adequate field demonstration is still a
required step. Biological treatment, however, is not the final answer.
The public desire for cleaner water and the new and promised government
regulations resulting from this desire no doubt will give the proper
incentives so that in the future the more desirable reduction of waste
quantities and improvement in the efficiency of product recovery will be
achieved to a high degree by the food growing and processing industry.
344
-------
REFERENCES
1. Monod, J., Recherches sur la croissance des cultures bacteriennes>
Herman and Cie, Paris (1942).
2. Lawrence, A. W. and McCarty, P. L., "A Unified Basis for Biological
Waste Treatment Design," Proc. Sanitary Engineering Div., American
Society of Civil Engineers, in Press.
3. McCarty, P. L., "Energetics and Bacterial Growth," Proceedings of
the fifth Rudolf Research Conference, Rutgers-The State University,
July, 1969.
4. Guttormsen, K. and Carlson, D. A., "Status and Research Needs of
Potato Processing Wastes," Proceedings National Symposium on Food
Processing Wastes, April, 1970.
5. Richter, G. A. and Pailthorp, R. E., "Aerobic Secondary Treatment
of Potato Processing Wastes," Proceedings National Symposium on
Food Processing Wastes, April, 1970.
6. Church, B. D., "Use of Fungi Imperfecti in Waste Control," Proceedings
National Symposium on Food Processing Wastes, April, 1970.
7. Graham, J. L. and J. W. Filbert, "Combined Treatment of Domestic and
Industrial Wastes by Activated Sludge," Proceedings National Symposium
on Food Processing Wastes, April, 1970.
8. Esvelt, L. A., "Aerobic Treatment of Liquid Food Processing Wastes,"
Proceedings National Symposium on Food Processing Wastes, April, 1970.
9. Streebin, L. E., "Cannery Waste Treatment by a Two-Stage Aeration
Process," Proceedings National Symposium on Food Processing Wastes,
April, 1970.
10. Jones, R. H., "Lime Treatment and Inplant Reuse of an Activated Sludge
Plant Effluent in the Citrus Processing Industry," Proceedings National
Symposium on Food Processing Wastes, April, 1970.
11. Soderquist, M. R., Williamson, K. J. and Blanton, G. I., "Seafoods
Processing: Pollution Problems and Guidelines for Improvement,"
Proceedings National Symposium on Food Processing Wastes, April, 1970.
12. Palmer, W. J., "Cannery Waste Treatment by a High Solids Activated
Sludge Process," Proceedings National Symposium on Food Processing
Wastes, April, 1970.
345
-------
13. Brenton, R. W. and Fischer, J. H., "Concentration of Sugarbeet Wastes
for Economic Treatment with Biological Systems," Proceedings National
Symposium on Food Processing Wastes, April, 1970.
14. Mercer, W. A., Maagdenberg, H. J. and Rails, J. W., "Reconditioning
and Reuse of Olive Processing Brines," Proceedings National Symposium
on Food Processing Wastes;, April, 1970.
15. Dostal, K. A. and Burm, R. J., "Status of R and D Development Efforts
on Food Processing Wastes," Proceedings National Symposium on Food
Processing Wastes, April, 1970.
16. Gilde, L. C., "Food Processing Waste Treatment by Surface Filtration,"
Proceedings National Symposium on Food Processing Wastes, April, 1970.
17. McCarty, P. L., "Anaerobic Waste Treatment Fundamentals," Public
Works, September - December, 1964.
18. Young, J. C. and McCarty, P. L., "The Anaerobic Filter for Waste
Treatment," J. Water Pollution Control Federation, 41, R160-R173
(1969).
346
-------
WURDD'S TASK FORCE ON AGRICULTURAL POLLUTION
by
J. Peter Clark1
One of the first acts of Dr. Arthur I. Morgan, Jr., upon becoming
the new director of the Western Utilization Research and Development
Division of the U.S. Department of Agriculture, was to create a special
Task Force on Agricultural Pollution. I was made chairman of this
group, which came to consist of about fifteen scientists from various
disciplines and commodity specialties. After numerous meetings, we
evolved both a plan of action and a number of specific research
suggestions. This paper is a brief discussion of these two areas of
accomplishment. Also, it is a general introduction to some of the
other pollution-related research conducted at our Division.
Process Modifications vs Waste Treatment
We have come to believe that modifying food processes is a most
promising approach to reducing pollution from agriculture. The classic
examples, discussed in greater detail in other reports, are Dry Caustic
Peeling of Potatoes (Graham, Smith*) and In-Field Processing of Tomatoes
(Wagner). In both cases, not only is the waste quantity reduced but
product yields are also increased. It should be dbvious that any
increase in yield must reduce the potential quantity of waste. Yet many
food processes remain unchanged, out of tradition or for lack of
attention, or, perhaps, because the benefits of change have not been
made sufficiently impressive. It is only relatively recently that waste
disposal and pollution control have demanded attention so aggressively
from the processor. In response to such a demand, it is normally
convenient to consider forms of biological treatment. But, as various
studies by the FWQA have shown, biological treatment of food wastes in
the volume frequently encountered demands enormous capital investment
and unpleasantly high operating expenses. Thus may be provided the
required incentive for studying food processes carefully and for adopting
new processes that reduce wastes and improve yields.
One of the ways yields can be improved is to produce new products
such as animal feeds, chemicals, or building materials along with the
original primary product. In this area, our Division is doing research
on the conversion of both cellulosic and leafy plant wastes to animal
feeds (Kohler). Also, we have sponsored an investigation of various
uses for rice hulls (Staackmann).
Chemical Engineer, Engineering and Development Laboratory, Western
Utilization Research and Development Division, Albany, California.
References are to other papers presented at the National Symposium on
Food Processing Wastes.
347
-------
Frequently the necessary modifications are not very sophisticated,
but yet quite effective; the experience of Artichoke Industries in
Castroville is a good illustration (Perkins).
The examples I have cited represent on-going, rather large projects
for the most part, with which the Task Force has had little direct
connection. On the other hand the Task Force is looking carefully for
new opportunities for the same sort of discoveries. For example,
blanching is recognized as a major source of water pollution in freezing
operations. It may be that new forms of heat transfer using gases,
recycled liquids, or perhaps microwave energy can be justified by a
reduction in waste water. Or, more exciting, new approaches to freezing
using protective chemicals may eliminate the need for blanching while
improving product quality.
Another process we feel could stand examination is the conversion
of feed into meat and manure in the factory we call a cow. With
concentrated animal feeding threatened by water quality control
restrictions, the volume and properties of wastes produced by a given
feed mix may be as critical as the animal weight gain in optimizing feed
lot operations.
In other divisions of the USDA, concepts have appeared which we
seek to apply to new situations. For example, our Weslaco, Texas,
laboratory has made pureed drink concentrates from whole citrus fruit,
greatly reducing the waste produced. Perhaps other commodities can be
used with less peeling or trimming than at present.
Many agricultural commodities rely on strong salt brines for
processing. We have learned how to reuse olive brine by treating it
with charcoal (Rails) and we have also investigated processing leather
tanning brines. The pure culture fermentation process for pickles
developed by our Southern Laboratory permits packing the brine instead
of throwing it away. Perhaps other commodities can also be packed in
their associated process liquids, for example, fruits in their own
sugars, as pineapple often is.
The opportunities for process modification are clearly very broad.
The general principles seem to be: minimize water use, recycle water
or other liquids where possible, put as much of the raw material into
the final product as possible, and make salable material from whatever
cannot go in the primary product.
Other Research Problems
One of the major difficulties of evaluating a waste stream for
either treatment or value recovery is the lack of a rational means of
expressing the composition of the stream and the relationship of the
stream to its environment. A good illustration of the inadequacy of
current measurements, such as BOD, COD, TDS, SS, etc., is the waste
stream from olive processing. This strong salt brine is a serious
pollutant in the Central Valley of California, but it probably would
not be so along the Pacific Ocean. There is clearly a need for some
practical representation of the interaction of a given stream with its
environment—that is, a kind of quantitative ecology.
348
-------
Still another difficulty with traditional measurements is their
presumption that a given stream is a waste and should be reduced to
innocuous CC>2 and l^O. This is why the biological or chemical oxygen
demand is a frequent measurement. But if a waste stream is to yield
biological or chemical value, then some nutritional property or chemical
analysis is a more meaningful measurement. There is a certain
psychological benefit in referring to a dilute sugar stream in terms of
its caloric content, for example, instead of its tendency to putrify.
One would like to recover calories, but would dispose of putrification.
On a more practical level than analysis or "psychology," the Task
Force has examined a number of diverse potential research problems in
the pollution area. One example is uses for the Uni-flow Filter (Popper)
which we see as an important solid-liquid separation device. Many
conventional waste treatment processes have serious sludge handling
problems which may be amenable to this new approach.
Sometimes we look at a specific waste problem and then expand our
consideration. At one time the waste from some castor oil production
was thought to be toxic to all animals. We found that fish could live
well on castor pomace that was toxic to warm blooded creatures. Now,
the castor meal produced in this country usually is not toxic because of
the treatment it normally receives, but the tolerance of the fish to the
poison does provide a margin of safety which thus creates a possible
new use for this material. Castor is not the only potential source of
nutritious fish feed so we are now doing research in the general area of
fish culture. We are learning about both catfish and carp, their
nutritional requirements, habits, diseases, and other cultural needs.
Our Laboratory has a wide background in enzyme technology and so
we are looking at the possibility of growing useful enzymes on waste
streams and also at the use of enzymes to convert or treat waste streams.
As a side issue, we have become concerned about the potential health
hazards of widespread use of enzymes and are trying to learn whether the
enzyme itself or some impurity may be responsible for allergic reactions.
We started looking for a way to clean up the waste stream from wool
scouring and are now trying to find a better way to clean raw wool while
recovering lanolin for sale.
We have discussed uses for whey, the largest food processing waste
in this country, and have tried to find alternatives to the solid waste
disposal problems created by commodity diversion under various marketing
orders. We have unfortunately not had any bright ideas in either area
so far, but they illustrate the breadth of topics considered by the
Task Force.
Basically, the Task Force on Agricultural Pollution is a new form of
organization intended to develop research ideas and to sponsor new projects
until they are absorbed by our commodity research groups. By mixing
scientific disciplines in an informal, non-critical atmosphere we have
promoted imaginative and perceptive ideas which we hope will lead,
someday, to successful ventures comparable to dry caustic peeling and
others recently reported.
349
-------
IN-FIELD PROCESSING OF TOMATOES
by
Joseph R. Wagner
Tomatoes are an important processing vegetable in the United States.
From 5 to nearly 7 million tons of tomatoes have been delivered to our
canneries per year. The State of California produces and processes
approximately two-thirds of the national total. The industry in
California is unique with regard to its dependence on machine harvesting;
approximately 95 percent of the processing tomatoes delivered to can-
neries are now machine harvested.
The adoption of machine harvesting in California was accelerated by
an acute shortage of field labor. However, it is now the cheapest and
the only practical available method for harvesting the crop. It seems
safe to assume that large scale production and machine harvesting of
tomatoes will continue in California.
The growth of machine harvesting has had many effects on the
industry, among which is the large amount of waste material. As now
practiced, the crop is harvested in a single pass. The vines are cut at
the base and lifted by the harvester for separation of fruit and vines.
The vines and much of the cull fruit and dirt are dropped behind the
harvester, and if the harvester is operated without a sorting crew the
tomatoes may include 20 percent or more of dirt, trash and culls (1).
Even under good operating conditions there is apt to be more damaged
fruit than commonly encountered in hand picked tomatoes. A crew of
sorters usually rides the machine to remove and return to the field culls,
trash, and dirt. Sorting on a moving harvester is a hard, unpopular job.
Even with a large sorting crew it is impossible to achieve complete
separation on the harvester. Much of the fruit damage may go undetected,
and later contributes materially to spoilage during long hauls to can-
neries and other delays prior to processing (1)• Collectively the dirt,
trash, and cull tomatoes, regardless of cause, create a serious waste
disposal and sanitation problem at the cannery.
Industry concern over the difficulty of assuring the quality of raw
fruit delivered to the cannery has lead to the investigation of in-field
washing and sorting stations. The location of sorting, cleaning and
sanitizing facilities in the immediate harvest area offers important
potential advantages. Rejected raw product, trash, field soil, and
product wash water would remain in the field. Waste disposal there can
be by simpler methods and at less cost than at a central plant in an
urban community. Unfortunately, this system also has flaws.
Chief, Vegetable Laboratory, Western Utilization Research and
Development Division, ARS, USDA, Albany, California.
350
-------
Studies conducted in recent years by the tomato processors, the
National Canners Association (NCA) and the University of California
indicate that elimination of dirt, trash and culls at the in-field
stations is quite feasible (l-j4). However, the handling and washing
causes additional damage to the ripe fruit. The sorted, washed fruit is
very vulnerable to mold and other types of in-transit and holding losses.
Therefore a considerable portion (10 to 20 percent) of the clean, ripe
fruit collected from a washing station may spoil in the period required
to deliver it to a distant processing line. To the loss represented by
the value of the spoiled fruit one must add the costs of transportation
and additional waste disposal.
The industry now recognizes the potential advantages of processing
these clean, washed and highly perishable tomatoes before large in-transit
and holding losses can occur. Processing tomatoes in the rural harvesting
areas should reduce the total amount of waste and increase the yield of
processed products per ton of tomatoes harvested. Thus, virtually all of
the solid and liquid wastes derived from the preparation of tomato juice
and concentrates would be retained in the harvest area. Only liquid
product from such preprocessing operations would be transported to urban
areas where plants exist. Such an in-field preprocessing station would
include the facilities of a sorting and washing station plus those for
crushing the clean tomatoes, heating of macerates and pulping to separate
the juice from the cores, skins and seeds. Evaporative cooling of the
juice or partial concentration or acidification during preprocessing are
all possible methods for stabilizing the liquid product for transporting
to a central plant for final processing and packing. At least two large
tomato processing firms have investigated field processing. However,
their data are not available and we cannot effectively comment on their
experience.
During 1969 a novel field research and demonstration tomato
processing unit was built at Albany and operated at the Anzac Sorters
station near Dixon, California by engineers and chemists of the Western
Regional Research Laboratory.
Our unit combined the concept of field processing with the applica-
tion of a new acidified tomato juice extraction process developed at our
Laboratory for the production of tomato products of very high consistency
(5^-7)' The adaptation of the acid extraction process to our field
studies is now referred to as the Acidified-Field-Break process.
The field unit consists of three mobile components: The process
line mounted on a large flat-bed trailer, a portable steam boiler, and an
air-conditioned laboratory. The processing unit performs the following
major steps:
(1) Preparation of tomatoes by washing and inspection.
(2) A hot-break for rapid enzyme destruction by heat within
seconds after crushing the tomatoes.
(3) The optional acid treatment; the addition of acid to maximize
extraction and product consistency and subsequent neutralization of added
acid (either before or after pulping).
351
-------
(4) Pulping to separate pulp from seeds, skins and fiber.
(5) Cooling and deaeration.
Additional detail on the process line is shown schematically in
Figure 1 (8).
The unit was operated at Dixon from early August through the first
week of October. Washed and sorted tomatoes were obtained from the
nearby sorting and washing station. In a number of holding experiments
comparisons were made between tomatoes processed within 2 to 4 hours after
delivery and tomatoes processed 24 hours later. It was found that holding
firm ripe tomatoes in bins at ambient temperature even without hauling
resulted in a 6 percent increase in culls unfit for processing. Under
similar conditions, 12 percent of soft, fully ripe tomatoes were rejected
after holding for 24 hours. Therefore, prompt processing resulted in 6
to 12 percent increases in product yield in these experiments. These
yield increases should be largely attributed to the advantage of a field
location which favors prompt processing and minimization of handling
damage.
When macerated tomatoes were processed with heat and the acid
extraction treatment, 4 percent more solids were recovered from raw
tomatoes than when heating only was used. This effect should result in
an equivalent increase in product case yield per ton of tomatoes processed
for manufacturers of concentrated tomato products. The increase in yield
is confirmed by the reduction in weight of tomato pomace. The more
complete transfer of soft tissue from the pomace fraction to the juice
fraction results in a drier pomace, which should be easier to dehydrate
for conversion to animal feed. In our studies the pomace, culls, leaves,
stems, and other solid wastes were disposed of on nearby farm lands.
Wash water was also returned to the land.
Juice would be the only material transported to a central plant.
For stability in the transportation period, the product should be cooled.
To obtain sufficient stability, products at natural pH (approx. 4.4)
should be chilled to 40°F. If neutralization is postponed until
delivery to the final processing point, product prepared at pH 2.75 can
be held up to 24 hours at 70°F. without significant deterioration from
microbiological action or loss of physical attributes such as color and
consistency.
In closing, it seems appropriate to identify the people who
contributed extensively to the project: The acid extraction process
was originally developed by a group of chemists from our Laboratory
headed by chemist J. C. Miers. In 1969 the process was scaled up to
field operations under the leadership of engineer R. P. Graham. The unit
was designed by engineer W. G. Schultz, who subsequently supervised the
field operations in collaboration with Mr. Miers. Others who contributed
to the success of the project were engineers J. L. Bomben and W, C.
Rockwell and chemists M.-D. Nutting, D. W. Sanshuck, W. C. Dietrich,
R. Becker, and H. J. Neumann.
352
-------
PROCESS LINE
PRE-WASHED
AND SORTED
TOMATOES -
6IN DUMPER
WATER
R/NSE TANK
T£MP. coMrftoueR-
Reco*o£K PUMP
/A/S0ECT/OA/
6£LT
BACK PKSSSUfte VAL\(£_
P*
HCI
ADOJT/ON
CHAMBER*
pH CONTROLLe*-
\
CONTROLLED
T/ME
TREATING
CHAMBER
NaOH
ZUrfiA
CHAMBER**
f/N/SHER
VACUUM
FLASH
COOLER
*
WASTE
PUMP
Acid treating and /Vet/fra//z/no steps opt/ona/.
Optfono/. Juice may be transported in ac/'d cond/t/'on.
Figure 1
KEITZ
0/SINTEGRATOR
STEAM
/NJECT/ON
TEE
WATER CffffLfR
I
TUBULAR
COOLER
PRODUCT
353
-------
REFERENCES
1. Reports of the NCA Industry Committee on Tomato Harvesting and
Handling. 1966. National Canners Association Research
Foundation, 1950 Sixth Street, Berkeley, California 94710.
2. Tomato In-Field Washing Station Study. December 1967. National
Canners Association Research Foundation, 1950 Sixth Street,
Berkeley, California 94710.
3. Final Report. Tomato In-Field Washing Station Study. January
1969. National Canners Association Research Foundation, 1950
Sixth Street, Berkeley, California 94710.
4. O'Brien, M., Lorenzen, C., Chesson, J., Reed, G., and Angle, T.,
Engineering Aspects of Central Sorting and Washing of Tomatoes,
Research Progress Report. Dumping, Washing, Soil Disposal,
Sorting and Filling. 1968. Department of Agricultural
Engineering, University of California, Davis, California 95616.
5. Wagner, J. R., Miers, J. C., Sanshuck, D. W., and Becker, R.,
Consistency of Tomato Products. 5. Differentiation of
Extractive and Enzyme Inhibitory Aspects of the Acidified Hot
Break Process, Food Technology 23(2): 113, 1969.
6. Miers, J. C., Sanshuck, D. W., Nutting, M.-D., and Wagner, J. R.,
Consistency of Tomato Products. 6. Effects of Holding
Temperature and pH, in press.
7. Wagner, J. R., Miers, J. C., and Burr, H. K., U.S. Patent
No. 3,366,488, Production of High Consistency Tomato Juice, 1968.
8. Figure 1 was part of press release #302-9-69, September 25, 1969,
from this Laboratory.
A 16-mm color film was shown during the presentation to illustrate
a typical field harvesting situation, the operation of the field
processing unit, the reductions in wastes, the normal post-harvest
cultivation of tomato fields, which offers a convenient route for
field disposal of culls and debris, and the proposed utilization of
cooling and other waste waters. The film is available on a loan
basis from the Engineering and Development Laboratory, Western
Utilization Research and Development Division, Agricultural Research
Service, USDA, 800 Buchanan Street, Albany, California 94710. The
film runs 13 minutes.
354
-------
"DRY" CAUSTIC PEELING OF VEGETABLES AND FRUITS
by
Robert P. Graham
The peeling of vegetables and fruits has long been a source of
pollution in the processing industry. Hydraulic procedures to remove
softened peel after lye or steam treatment have been customary. Present-
day rulings regarding pollution, however, have led to a search for
alternate methods of peel removal.
At the Western Regional Research Laboratory of the United States
Department of Agriculture we have studied various methods of peeling
vegetables and fruits to reduce the amount of peel waste which enters
plant effluents. Such methods must also be economically feasible. It
was recognized that some method of removing peel without the use of
water would be desirable. Also, the procedure should work acceptably
on both fresh and stored produce.
White Potatoes
Our initial efforts were directed toward white potatoes because of
the large amount of peel loss involved. The procedure developed was
called "dry" caustic peeling. We tried various methods of abrading
the peel material from the potatoes after treating them with hot lye.
Early tests on single potatoes impaled on a spit showed that the
potatoes could be peeled with a dry brushing action using nylon brushes
or flexible rubber-tipped brushes. This process was made more effec-
tive by heating the lye-treated potatoes with infrared heaters prior to
brushing. Heating with steam and hot air did not produce the same
results. It was also noted that if the lye-dipped potato was held for
a period of time after dipping the effect of the infrared heating was
enhanced and less caustic was required.
To study the technique further, experimental equipment with a capacity
of about 500 pounds per hour was built. The first peeling device con-
sisted of a single rubber-tipped roller. While this single-roll unit
further encouraged our use of rubber-tipped rolls as a peeling medium,
it had too many limitations. Two other peeling devices were built with
different configurations to get a unit that would continuously handle
potatoes in a small pilot plant. Both of these devices worked, provided
the tip speed of the rolls was sufficient to make them self-cleaning.
Chief, Engineering and Development Laboratory, Western Utilization
Research and Development Division, Agricultural Research Service,
United States Department of Agriculture, Albany, California.
355
-------
A forty-inch diameter barrel was used to turn the potatoes while
they were being irradiated. The barrel, after several modifications,
provided continuous operation. A 500-pound per hour pilot plant was
then built to permit experimentation on potatoes stored for varying
lengths of time and to provide enough peel material for examination.
Because the potatoes were partially peeled in the barrel, two different
waste products were produced, one essentially dry and the other with
about 25% solids content.
This plant demonstrated the general feasibility of the "dry"
caustic peeling procedure, especially as the amount of peel removed
could be controlled by caustic dip time. It did, however, have
several rather serious limitations. First, scaling up the barrel pre-
sented some serious difficulties, and second, residence time in the
barrel was not uniform, since it depended on the shape of the potato.
To overcome these difficulties a live-roller conveyor was tried. This
is the type of equipment used for tomato inspection. When fitted with
infrared burners it provided continuous turning of the potatoes and
positive control of residence time under the burners. With this system
almost all of the peel material was removed at one time, as only a
small amount of the peel stuck to the rollers. Also, this unit used
only 1/4 to 1/2 as much gas as the barrel. The amount of peel removed
could be controlled by the caustic dip time and could be compensated
when desired by control of the irradiation time. Potatoes leaving the
peeler had a small amount of sticky residue not removed by the rubber-
tipped rollers. This was removed by brushes with a small amount of
water.
Peeling potatoes under proper conditions does not produce any
significant heat ring. The potatoes do tend to darken sooner than
potatoes peeled in the conventional system, but this can be prevented
by a sulfite dip after peeling.
Further work done on this system was carried out with commercial-
size equipment manufactured by Magnuson Engineers.
Sweet Potatoes
Some tests have been made on sweet potatoes, both on individual
potatoes and on potatoes in modified commercial-size machines. The
peeling loss for sweet potatoes is usually larger than for white
potatoes. Most of the peeled sweet potatoes are cut into about 1-inch
thick pieces and are canned. The potatoes for this purpose should have
a smooth surface so that when they are autoclaved none of the potato
will sluff off into the syrup. Sweet potatoes after washing are
usually preheated in hot water and then immersed in about 20% hot
caustic for 2 to 3 minutes. Peeling conditions are similar to those
for white potatoes except that the brushing state following the stud
rubber peeler is somewhat more critical.
356
-------
Fruit
At the Western Laboratory we are now studying the "dry" caustic peeling
of fruits for canning. A caustic dip similar to that used for conven-
tional peeling with water sprays is used. The infrared heaters are not
used because the heat from the burners tends to bake the peel on the
fruit rather than let it soften:
Cling peaches are cut in half, pitted, and placed cup down on a
mesh belt. Boiling 3% caustic is cascaded over the fruit for 5 to 10
seconds followed by a 10 to 15 second drain in a warm humid atmosphere.
The fruit is then run through a rotating rubber disk peeler to wipe off
the softened peel. The 1/32-inch rubber disks are stiffened by metal
side disks which allow the peach halves to turn over but not drop
through the rotors. The peeler as shown is mounted at a 45 degree angle
and the fruit flows down against the rotation of the disks. We have also
operated the unit in a nearly horizontal position so that the fruit is
peeled as it is being conveyed along the peeler. Peeling losses are 4
to 5%, which is comparable to the loss when water sprays are used for
peeling. Whole apricots, freestone peaches, and Bartlett pears have
been peeled on the disk peeler. When peeling ripe pears the edges of
the disks tended to mark the soft pear flesh, especially on the neck of
the pear. This marking was reduced by cementing 1/8-inch diameter
rubber tubing on the edge of the disks.
A 12-inch wide by 10-foot long model of the rubber disk peeler
will be tested during the 1970 canning season in a local cannery.
In conclusion, at the Western Regional Research Laboratory of the
United States Department of Agriculture we have attempted to attack
pollution caused by vegetable and fruit peeling through process modifi-
cation. We have attempted where possible to keep the peel materials
from ever entering the plant effluent. In the case of the peeling of
white potatoes we feel that this procedure will also result in
economies in excess of the savings through control of water pollution.
357
-------
REFERENCES (Showing Pictures of the Process)
1. Graham, R. P., Huxsoll, C. C., Hart, M. R., Weaver, M. L. and
Morgan, A. I., Jr., "'Dry' Caustic Peeling of Potatoes," Food
Technology, 23:2, 61 ff, (1969).
2. Graham, R. P., Huxsoll, C. C., Hart, M. R., Weaver, M. L. and
Morgan, A. I., Jr., "Prevents Potato Peel Pollution," Food
Engineering, 41(6), 91-93, (1969).
358
-------
PILOT PLANT .EXPERIENCE OF USDA-MAGNUSON
DRY CAUSTIC PEELING PROCESS
Traver J. Smith1
In 1968, the Western Regional Research Laboratory had successfully
experimented with peeling white potatoes utilizing infrared radiation
to accelerate caustic destruction of the skin, followed by wiping off
the softened peel with stud rubber rolls. We could see a potential
application of our MAGNUPEELER in this process and about the same time
we acquired a basic patent on an infrared-caustic peeling method which
had been issued to Mr. Miles Willard, a well known food processing
consultant. Consequently, we entered into a cooperative agreement with
the Potato Processors of Idaho (a trade association) and the Western
Regional Research Laboratory to construct a pilot line for commercial
production tests on the application of Dry Caustic Peeling to white
potatoes. The Idaho Processors furnished the standard portions of the
line, Magnuson provided three new machines especially for the Dry
Caustic Process, and the Laboratory provided technical supervision.
The pilot line was set up in parallel with a conventional caustic
peeling line at the Aberdeen, Idaho, plant of Western Farmers Association.
Potatoes were diverted just ahead of the commercial process, run through
the Dry Caustic Process and then mixed back with the commercially peeled
potatoes. This permitted us to make precise comparisons between the two
processes on identical potatoes.
The first unit in the pilot line was a small ferris wheel type
conventional caustic dip unit. The potatoes were dipped in a 12% caustic
solution for one minute, while in the conventional process they were
dipped in 20$ caustic for four minutes. Both caustic solutions were at
a temperature of 170°F. The caustic consumption on the dry peel line
was only about 20$ to 30$ of that used on the conventional line because
of the more dilute solution and the shorter dip time.
Following the caustic dip, the potatoes were stacked 10 to 12
inches deep on a holding belt for 3 to 10 minutes (depending on their
condition) to allow the caustic to penetrate a thin layer at the surface.
1
Vice President, Magnuson Engineers, Inc., San Jose, Calif.
359
-------
After holding, the potatoes entered the Infrared Unit, the first of
the new Magnuson machines. A SHUFFLO was used to deliver one row of
potatoes at a time onto a roller conveyor which rotated them as they
passed closely beneath gas-fired infrared burners. The infrared source
was a porous ceramic material heated to a red hot l650°F. All surfaces
of the potatoes received remarkably uniform exposure to the infrared
radiation. The effect of the infrared was to quickly activate the
caustic to totally soften the thin surface layer without forming the
heat ring which usually results from caustic peeling. The infrared
exposure time was only about one minute.
For the next few seconds, the potatoes were passed through the MAGNU-
SCRUBBER. This is a rotating cylindrical cage of stud rubber rolls. The
potatoes are tumbled through the cage, with their flow being regulated
by a central screw conveyor. The stud rubber rolls are rapidly rotating
in a planetary action to wipe off the softened peel and fling it away
by centrifugal force. This material is collected on- the inner surface
of a rotating drum and wiped out at the bottom. The peel waste is a dark
brown thick paste of 23 to 25% solids at pH 11. It is more concentrated
than the potatoes, because of the surface evaporation which occurs during
the infrared treatment. It can be conditioned for animal feeding.
Up to this point no water has been used in the process. The potatoes
leave the MAGNUSCRUBBER with a thin coat of this tacky paste over their
surface. This is removed by the MAGNUBRUSHER, the next machine in the
process. It is a cylindrical brush washer and uses a small amount of
water which can run into the existing plant waste system or can be blended
with the peel waste. This will reduce the solids content of the peel
waste to possibly 15$, but no waste water at all will be discharged from
the process. After brushing, the potatoes are briefly dipped in a 0.5%
solution of sodium bisulphite to fix the color. This is also frequently
done in conventional caustic peeling, but it is essential in the Dry
Caustic Peeling process because these potatoes have no heat ring.
Incidentally, many people are confused by the term "Dry Caustic
Peeling," thinking that it somehow involves undissolved dry flake caustic.
Consequently, we and the Laboratory have adopted the name "USDA-Magnuson
Anti-Pollution Peeling System," and because that's so long, we usually
call it "Infrared Caustic Peeling." The process is referred to by all
three names in published literature.
The Infrared Caustic Peeling process definitely accomplishes the
objective of keeping the peel waste out of the plant waste water. For
a French fry plant this reduces the solids in the plant waste system by
approximately 15%- In a flake or granule plant, where less peeling is
required, the reduction is about 50$. This obviously is very beneficial
to waste treatment requirements.
The water used in the Infrared process is only about 5% of that used
in the conventional caustic peeling process and only about 8% to 10$
of that used in a typical steam peeling plant. This is another tremendous
benefit to waste treatment requirements.
360
-------
The amount of peel waste produced is reduced by about one-third.
This, of course, also reduces the peel waste handling problem, but more
importantly, it substantially reduces the cost of peeling potatoes by
increasing the product recovery.
Because of the absence of the heat ring and the reduced caustic
consumption, the potatoes peeled by the infrared process are much whiter
than those produced by the conventional caustic process and they do not
have the customary slippery surface which is associated with caustic
peeling. This is an obvious quality benefit. The infrared treatment
also selectively softens dark areas of defect and they are more completely
removed in the subsequent wiping processes. This reduces the amount of
hand trimming required to finish the potatoes for processing, another
quality benefit.
This process shows great promise in other product applications.
For example, the Western Utilization Research Laboratory has done much
experimental work with sweet potatoes, where processors face waste
treatment problems similar to those of white potato processors. Here,
too, the process successfully removes the peel without using water and
keeps the peel waste out of the plant waste disposal system. A very
acceptable, high quality, smoothly peeled sweet potato has been produced.
We have had some most successful short tests with Magnuson commercial
equipment on sweet potatoes and hope to have a commercial line operating
soon.
Other products, including beets, carrots, onions, and some fruits,
have been run experimentally in laboratories. At this point we are
confident of commercial success with beets and carrots, while the other
products are most promising, but require more development.
We would like to express our appreciation to the Western Utilization
Research Laboratory and particularly to Mr. R. P. Graham, Dr. C. C.
Huxsoll, Mr. M. R. Hart, Dr. M. L. Weaver, and Dr. A. I. Morgan, Jr.,
Director of the Laboratory. All of these men have been personally
involved in the development of this process and we at Magnuson Engineers,
Inc., recognize them for their help and cooperation.
361
-------
POSSIBLE USES OF UNI-FLOW FILTER
Karel Popper^-
Most effluents contain both suspended and dissolved solids and in
very many such effluents the BOD load of the suspended material vastly
exceeds that contributed by the solute.
The primary treatment task to be carried out vith such material then
is the removal of the solids from the water phase. By extension, one may
be able in some effluents to bring about precipitation of some of the
noxious materials so that their removal would also become possible. In
some rather common instances the removal of the solids will make the
liquid either valuable or reusable. A tool that will do the job at a very
low investment cost and a low operation cost had therefore to be invented.
Figure 1 may help to explain the function of the unit. Sludge enters
the top of the hose and flows in it downwards. The solids move to the
bottom and the solution oozes out through the cloth. As we go on, the
sludge in the bottom of the hose accumulates to an ever higher amount and
compaction and when it reaches a satisfactory stage, is ejected.
While the function of the unit is to filter, that is to separate
solids from liquids, it would be wrong to think of it in engineering terms
of a filter. The principle, as I see it, is LOW PRESSURE. Low pressure,
no filter cake. Ho filter cake, no flow interruption. It really is as
simple as that. Just let the sludge seek its own merry way to the bottom
of the tube both by gravity and with the assistance of flow.
Uni-Flow filter inlet is shown in Figure 2. The slurry enters the
distributor and flows through the nipples into the individual hoses.
Figure 3 is the Uni-Flow filter bottom. The hoses fit into the sludge
take-off nipples which are fitted into a divisor plate. The filtrate
stays above the divisor plate whence it is piped away and the sludge
moves through the nipples to the sludge collector cone.
An example of an effluent that never needs to be discharged is water
softener regenerant.
1
Fruit Laboratory, Western Utilization Research and Development Division,
ARS, USDA, Albany, California.
362
-------
SLURRY
TUBE
o
•-
VALVE
FILTRATE
Figure 1
-------
o o
, V" X - -
>•• . •--
» '
- ^ - ,
.
Figure 2.
364
-------
Figure 3
Figure 3.
365
-------
Zeolite water softening is exchange of hardness cations, calcium
and magnesium, from the water for sodium on the ion exchange resin.
2RNa + Ca (HCO-j)?*'' R2 Ca + Na^CC^ + C02
+H20
2RNa + Mg Cl2 ' R2Mg + 2NaCl
I, for one, see no substantial advantage in such an operation except
for boiler feed, and for a number of reasons would oppose its application
on a municipal scale. Nevertheless, the general public has been sold on
the idea of soft water, pretty much as it has been sold on other harmful
products, and the result is a massive discharge of high concentrations
of sodium chloride during the regeneration of the exchanger.
R2Ca + R2Mg + UNaCl - ItRNa + CaCl2 + MgCl2
Because the usually employed sulfonic cation exchange resin has a much
higher affinity for the divalent than for the monovalent ions, a
considerable excess of salt is usually, if quite unnecessarily, used to
convert the exchanger to the usable sodium form. The regenerant effluent
thus is a fairly concentrated solution of sodium chloride contaminated
with calcium and magnesium, of no use to anybody and a troublemaker.
A quick think-back to high school chemistry teaches us that addition
of sodium carbonate to the spent regenerant will precipitate calcium
carbonate.
CaCl2 + Na2CC>3 — > CaC03 + 2NaCl
Calcium carbonate settles very easily and if one wanted to filter, would
filter with no trouble. However, when the effluent is made alkaline,
we also get magnesium hydroxide, a substance known to everybody from
his prep, chemistry course as that doggone goo that won't filter. It
also does not want to settle, an illustration being the extremely large
settling tanks involved in production of magnesium hydroxide from
sea water.
Now, everybody knows that the way to handle magnesium hydroxide
is to settle it out, just as everybody knows that a primary settler is
a fairly big and expensive, and sometimes evil smelling, tank.
The problem is that a small operator, say a relatively small canner
or freezer who softens his boiler feed, and, possibly, final wash water,
or a small town soft drink bottler, does not want to tie up his money
and valuable sp'ace in a settling tank just because some government
official tells him that he must not ruin everybody else's drinking water.
Yet he may well want to reuse his spent regenerant, if only he did not
have to make that investment.
So back to the settling tank. Settling tank size and thus cost is
a function of solids settling rate and concentration. That is, it takes
a given time to settle out a given amount of a suspended solid of a
certain settling rate without it making much of a difference, up to a
certain compaction point, how concentrated the suspension is.
366
-------
Thus, it can be reasoned that were one to build a tank with canvas
walls, the solution would go out through the fabric and the required
holding capacity would diminish. This in turn leads to the concept of a
very narrow and tall tank and to our way of thinking the logical outcome
is the unit shown in Figure k.
The regenerant effluent to which we added sodium carbonate is pumped
to the top distributor of the unit. From there the slurry flows down
through the hoses and the clear filtrate permeates through the fabric.
The fabric is cotton tubing manufactured by the Tape-Craft Company in
Anniston, Ala., and sold as "Soaker Hose." It is normally used for slow
irrigation in gardening. The permeate collects in the filtrate collector
and flows out through the outlet. The compacted sludge collects in the
discharge cone from which it is intermittently removed by opening a
valve. Instead of a valve, one really should have some continuous removal
device, such as a screw, a Moyno pump, or some other rate-programed gadget.
Figure 5 is a close-up of the Uni-Flow filter bottom. The white
material in the cone is a mixture of calcium carbonate and magnesium
hydroxide and the clear stream is a solution of sodium chloride. This
sodium chloride, rather than going to the sewer, can go back to regeneration
and save money.
That savings may be able to persuade the processor to stop discharging
his pollutant.
The principal reject material of the wine industry is wine lees.
This is the material left in the tank after the wine is drawn off. Lees
have a common feature with magnesium hydroxide. They are a gooey mess
that does not want to filter. Yet they are found in the bottom of the
tank; ergo, they settled out.
We pumped them up and let them run through the Uni-Flow filter and
collected the filtrate at a rate of 7 gallons per minute. The filter had
T hoses Ik feet long. The rate was thus, as near as makes no difference,
0.3 gpm/sq.. ft. of hose. I built a similar filter that handles 50 gpm
of lees for about $800. If you wish to operate the unit 3 months out of
the year, you find the following fixed charges:
Depreciation 10 years $80.00
Capital cost 1% p.a. 56.00
Maintenance 3% p.a. 2U.OO
Hose replacement in 16 months
at U cents per ft. of hose 3.25
At 22 days per month
= 1,572,000 gallons $163.25
Per 1000 gallons 1<£
The recovery of 1000 gallons of wine worth at least $500 will thus require
an expenditure of 1 cent in facilities. I did not include a pump since a
winery has plenty of idle pumps during that period in which it wishes to
process lees.
367
-------
Figure 4.
368
-------
Figure 5.
369
-------
Figures 6, 7, 8 and 9 show the results. Note the exponential type
of settling.
While on the subject of wine, I would like to say that the Uni-Flow
filter is an intrinsic part of a still slops disposal process that we
developed. Still slops are an extremely noxious pollutant and the
Uni-Flow filter does what some would have a centrifuge do by way of
solids separation. To differ from a centrifuge, it does it at a lower
order of magnitude cost.
Figure 10 shows the filtrate with a subsequent cold throw of
saleable potassium bitartrate. The tartrate value pays for processing
and gives a profit. Here again we have a situation I like to see.
It is too early and always will be too early to enumerate all that the
Uni-Flow filter can do to alleviate pollution. We have looked so far at
potato starch, where we had very encouraging results on cold cut starch
waters - the elimination of suspended solids was around 98$, with no
problems. With blanching waters, however, we had some clogging. Preliminary
experimental work with nylon, rather than cotton, was encouraging enough
to make us think in terms of non-wetting fiber tubes for some of the more
difficult to handle materials, such as cane juice or tomato juice.
We tried to eliminate clay from flume waters and were successful,
and we are toying with the idea of algae skim-off and compaction. We
also installed a unit at a local sewer plant and found a reduction of
92% of the suspended solids of raw sewage after the comminutor.
Laboratory experiments teach us that better than 99% can be obtained
after very low liming.
No doubt we will find things we cannot handle; so far we have found
only very few and none of them related to the topic of this meeting.
One thus would be likely to want to think in terms of re-use of
handling waters, of elimination of fiber in pineapple waste, concentration
of suspensions from stick water, and such like.
Ending, as I should, on an optimistic note, I say: Eureka, we have
a poor man's centrifuge!
370
-------
INPUT
FILTRATE ^" SLUDGE
WINE LEES
Figure 6.
371
-------
INPUT
FILTRATE
WINE LEES
SLUDGE
Figure 7.
372
-------
INPUT
FILTRATE
WINE LEES
SLUDGE
Figure 8.
373
-------
INPUT
FILTRATE
WINE LEES
SLUDGE
Figure 9.
374
-------
FILTRATE
AFTER COOLING
SLUDGE
Figure 10.
375
-------
FILTRATE
AFTER COOLING
SLUDGE
Figure 11.
376
-------
A CASE HISTORY IN FOOD PLANT WASTE WATER
CONSERVATION AND PRETREATMENT EXPERIENCE
Granville Perkins1
Artichoke Industries, Inc. is a grower-shipper owned specialty food
processor of frozen and canned artichoke products. The plant is located
in Castroville, California, an unincorporated town of 3800 population
in the hear of the artichoke growing area on Monterey Bay.
The town and 10,000 acres of artichokes occupy the mouth of the
Salinas Valley in Monterey County. The acreage is served by four fresh
shippers and two processors specializing in this commodity. The two
processors perform a market support function for the fresh shippers by
absorbing small sized marketable artichokes to be converted into artichoke
hearts and canned in glass with vegetable oil and spice as marinated
artichoke hearts, or packed as frozen hearts in-retail packages.
Large unmarketable artichokes are prepared to recover the bottoms
of the artichokes that are processed in glass with brine, oil or tomato
cocktail sauce or fresh frozen in institutional packages for restaurant
use.
In the fall of 1955 Artichoke Industries, Inc. started operations
in a ^0 x 200 foot metal building. That first year the plant processed
1,350,000 pounds of raw product over a seven month period. The preparation
system developed for processing artichokes, a difficult, highly oxidizable
product, depended upon water driven cutting equipment called hydroughts.
Cone shaped knives rotating at high speed shaved the stem end of the
artichoke until all of the stem and 3A of the leaves were cut from the
heart. The water used for driving the cutters was exhausted in a flume
and carried away the waste portions of the artichoke. The resulting
slurry of artichoke parts and shavings was transported by flumes and
gutters to a dewatering belt which separated most of the waste from the
water. The waste was hauled away and fed on the ground as cattle feed.
The water was discharged into a small sump and from there to the sewer.
1
Artichoke Industries, Inc., Castroville, California.
377
-------
The community of Castroville is served by the Castroville County
Sanitation District. This district defines the city limits of the town
and receives all domestic and industrial waste from this community and
treats it in a common two stage municipal treatment plant, designed
principally for domestic waste treatment. The plant consists of a
primary clarifier, bio filter, aeration tanks, secondary clarifier, and
sludge digester. The effluent from the plant discharges into a slough
known as the Tembladero, and from there, with the help of irrigation
tail water and rain water runoff, to Moss Landing Harbor and Monterey Bay.
The average design capacity of this system is 300,000 gallons per
day, hydraulic capacity for peak flows, 900,000 gallons per day, and a
design population equivalent of 3000. In 1955 the population of the
town Was 1900.
Visits to our plant by Sanitation District representatives started
in 1955 when the flow from Artichoke Industries was probably ldO,000
gallons of water per day. The flow increased to 200,000 gallons in 1956
and so did the frequency of visits from the Sanitation District. In
1957 the flow had increased to about 300,000 gallons per day. The BOD
was not too high but the hydraulic load on the Sanitation District by
all contributors was about 3 times what they could handle in a satis-
factory manner.
In 1957 mechanical trimming machines not driven by water were
replacing the water-driven hydroughts. The business was growing and
the logical use of water for pumping and handling of the highly oxidizable
artichoke hearts was growing at nearly the same rate. The mechanical
trimmers helped to reduce the load but fluming with all fresh water was
still in general practice.
In 1958 "we reached the peak of our water usage of over U00,000 gallons
per operating day. The artichoke tonnage had increased 7 times and the
waste water volume had increased k times over a period of U years. You
can imagine the concern of the operators of our District Treatment Plant
when over a period of four years the hydraulic volume went from 250,000
gallons per day to over 1,000,000 gallons per day, all because of two
small processing plants. They were constantly in trouble. They had
talked to us many times over this period about reducing our hydraulic
load. We listened, but we had more important problems.
Then one day they caught our attention. We were told that if we
did not take substantial measures to reduce our hydraulic load to the
sewage treatment plant, they would fill our sewer line full of concrete.
We decided not to test this threat and if you can't whip them the
next best bet is to join them. We visited the treatment plant, not once,
but often. We learned how it was operated and learned about their problems,
This started the long, hard road back to sensible utilization of our
important resources—water and the tax dollar.
The first effort was water re-use. A lot of time and effort went
into pumping water from one use to another. Over a period of one year,
we reduced our daily usage from over ^00,000 gallons to about 300,000
gallons.
378
-------
Then we received a shock. Our BOD had gone from 1500 ppm to
ppm. Originally, we had short time contact between waste and water.
With the re-circulation technique fully developed, we used water as
many as nine times for gutter flushing as a final use "before discharging.
We had concentrated soluble solids by re-circulating, even though the
water was continually screened before each use. The period of 1959 to
I960 can be called the "re-circulating era".
Then came the 'screw conveyor and pneumatic era," 196l to 1963.
We jack-hammered the bottom out of our floor gutters and placed steel
shells and screw conveyors in our gutters, back pumping the shells with
concrete.
We used a variety of slingers and blowers, winding up with a
Buffalo Forge 10" cast iron shell with our own design armor plate
impellor. We used 10" well casing for blow pipe. Blowers cut down on
water, but they do not accommodate metallic objects. We were constantly
patching up holes caused by misplaced wrenches and screwdrivers. In
1963 operating officials of our Sanitation District developed an ordinance
which when passed by the County Board of Supervisors created the
Castroville Public Works Advisory Committee to the Board of Supervisors.
I was invited by our district supervisor to be a member of this Advisory
Committee. I jumped at the chance. The Committee held public meetings
and advised the Public Service Division of the County Road Department
on matters affecting the various service districts they had charge of
for the community of Castroville. The Sanitation District was one of
these.
In 196^ the Advisory Committee started a study on an Industrial
Sewage Ordinance. The Ordinance limited the gallons per industry per
day and established the pounds of BOD per day. Based on a formula
including assessed valuation and payroll, additional amounts of BOD
could be purchased from the District if treatment plant capacity was
available. There is a free base allowance of treatment capacity with
additional treatment capacity for a fee, if available. It is not
considered a fine for exceeding your allotment.
The Ordinance controls the types of things that are not discharged,
limits ether soluble materials, BOD, and suspended solids, and it
requires daily meter records and reports from an independent laboratory
at regular intervals concerning limiting factors. Our plant base
limit is 60,000 gallons per day at 500 ppm BOD, which converts to
pounds of BOD per day at no fee.
In 196U, while the Advisory Board was studying an ordinance,
Artichoke Industries employed a consulting engineering firm to study
pre-treatment. After several months of study and design, a chemical
precipitation system preceeded by a 60,000 gallon receiving and mixing
tank was recommended by the consultant. A centrifuge was to be used
to concentrate the sludge. We started building our treatment plant
in 1965 following the consultant's design recommendation. We had
decided that we had to reduce capacity of a pre-treatment system to
60,000 gallons in order to fit the system into the space available and
to reduce the pre-treatment cost to something that would not put us out
379
-------
of business. ¥e had decided by this time that we had better not get
the product or its waste wet, if we could possibly help it. Once you
decide this, lots of ideas occur that help reduce the use of water.
Again, an inventory of water uses was taken. As a result, spray
nozzle sizes were reduced for wetting product with a fog in order to
reduce the oxidation, rather than literally flooding it with fan
sprays, and pneumatic separators replaced rod reel washers. Trimming
machines were re-designed to incorporate pneumatic separation at the
point of cutting. At every step, where possible, waste was separated
from product and handled dry. Product cooling water for the freezer
line is re-circulated, screened, refrigerated, and used over and over.
A low pH citric acid solution minimizes bacterial buildup at the 50°F
temperature. Blancher make-up is held to an absolute minimum. Blancher
water is the highest BOD water in our plant and highly respected as
such.
Jar washing serves a dual purpose. The temperature is 165°F and
the water contains non-foaming detergent. This hot water is used as
a pre-sterilizing tempering wash and is constantly recirculated to serve
as the first stage in cooling. The balance of the jar cooling is
accomplished by a combination of fog nozzles and air to make use of
the latent heat of evaporation.
In 1965 the Advisory Committee recommended that an improvement
tax be levied on property to pay for an addition to the District Sewage
Treatment Plant. This tax was figured in the budget for 1966. When
the property owners received the increase in their tax bills, they
were ready to run the Advisory Committee out of town.
By 1966, the District Treatment Plant addition was under construction.
Artichoke Industries had brought waste water usage down to 150,000
gallons of water per day with an average BOD of 8^0 ppm. The raw
material handled over an eleven-month period had increased better than
10 times since 1956, and we were using 50,000 gallons less water than
in 1956.
In 1967 our own pre-treatment plant began operation. It was designed
for 100 gallons per minute, or our quota of 60,000 gallons to be run
in 10 hours of daytime operation. It did not work out that way. At
^5 gallons per minute, we had 5-day BOD of ^60 ppm, suspended solids
20 ppm, BOD reduction of 75 per cent at a 50 per cent of design flow.
After a year of practicing the art of precipitation, we were able
to obtain BOD of 700 ppm with suspended solids of about 25 to 35 ppm,
at 85 gallons per minute. The cost of operation was fantastic in labor
and chemicals. We could not operate the above-noted levels at normal
plant waste water flow and ran the pre-treatment only during the day
so we could watch it.
After one year of pre-treatment, we had our daily volume down to
80,000 gallons by constant attention to plant water usage reductions.
380
-------
BOD was still bouncing from 600 to 1000 ppm, depending on how we loaded
our plant. Our percentage of BOD reduction was, at best, 60 per cent.
Because of the precipitation chemicals, our effluent went from 5-5 pH,
because of acid blanching that our product requires, to a pH of 10-11.
It is now 1968. The District Sewage Treatment addition, consisting
of a large mixing basin, additional secondary clarifier and additional
sludge digester, giving the municipal system 800,000 gallons per day
added capacity, is completed. We are gradually cutting down on our
processing plant water requirements. We have, by late 19&8, reduced
our hydraulic load to 40,000 gallons per day.
As we reduced our daily gallonage from 80,000 to U0,000, something
else happened. We began concentrating our soluble solids. By pre-treatment,
we were removing good percentages of ether soluble materials from our
oil packaging operations and doing a good job on suspended solids. With
no oxidation or secondary treatment, we had reached the end of the line
with soluble solids.
The area we have for pre-treatment measures 50 feet by 80 feet.
This certainly limits how much you can expect to do. It is surprising
how much money you can spend in 4000 square feet.
By the end of May, 19&9, we thought we had turned ourselves
inside out. We were not happy with our expensive precipitator, but it
was reducing BOD by 50 to 60 percent. We were then told that our high
pH discharge was causing problems at the District Treatment Plant.
Something was causing their clarifiers not to work. Their suspended
solids were remaining suspended to a high degree. After several days of
experimentation, we shut down our precipitator and just allowed the
waste water to flow through with no chemicals added. Within a few days,
the District suspended solids problem cleared up at the sewage treatment
plant. Our BOD jumped from an average 800 ppm to 1200 ppm. We
accomplished nearly a one-third reduction in BOD by just running our
plant discharge through the precipitator tank. The processing plant
discharge was about 2,000 ppm BOD.
Disappointed in cost and results with the first consultant's
design, we hired another. The chemical engineer we employed analyzed
data based on a 2U-hour sampling of our plant discharge. He came to
the conclusion that, due to the specific gravity of our suspended
solids, we could more efficiently float them than precipitate them.
We agreed with his analysis.
We are now in the process of converting a 6,000 gallon precipitator
into an 8,000 gallon clarifier by reversing the flow and making some
modifications. We intend to make additional modifications to create
a flotation unit.
We are continually analyzing our processes to reduce water usage
and reduce water contamination.
At present, we are separating our water by quality and plan to
handle these separately. Why run 50 ppm BOD water through pre-treatment
381
-------
when it can be blended with post pre-treatment effluent ahead of the
meter? In our particular case we take the hydraulic load off the
treatment plant and dilute the BOD of the pre-treated effluent before
discharge.
We know there are many things left to do in the processing plant
and in the pre-treatment area. This project will be with us as long
as we are there. We have accepted waste water handling as an extension
of our manufacturing process.
With all the trials and tribulations, we believe this story is
worth telling. It shows what can be done when it has to be done. If
there is something to be learned from our mistakes, they are submitted
to you along with our accomplishments.
382
-------
ANIMAL FEEDS FROM VEGETABLE WASTES
George 0. Kohler-'-
My objective this afternoon is to present to you a progress report
on two agricultural waste problems which are causing serious environmental
pollution problems in agricultural areas. These wastes are lignified
cellulosic materials such as straws, oilseed hulls, corncobs, etc., and
the non-edible green wastes from truck farms.
More than 150 million tons of cellulosic agricultural wastes are
produced annually in the United States. This figure does not include
sawdust, which amounts to ^0 million tons, or animal wastes, which contain
another 150 million tons of solids. Commercial production of vegetables
contributes an additional more than 300 million tons of fresh waste
annually.
Cereal straws constitute the largest single source of wastes which
up until now represent a relatively neglected agricultural resource.
To orient the problem more regionally, about 1 to 2 million tons of rice
straw grown in the Central Valley of California go to waste and a similar
quantity of perennial grass straw grown in the Pacific Northwest is
also wasted. Traditionally it has been cheaper to burn these straws
after harvesting the seed than to utilize them in any other way. However,
recent anti-air-pollution legislation makes it imperative to develop
alternative disposal procedures and we have been working along these lines
for some time. Simple ploughing back into the soil is impossible with
the perennial grass crops and is considered unsound and uneconomical by
most rice farmers.
Most of the discussions we have had during the past several days of
this conference have been directed toward getting rid of pollutants
after they have been generated by developing new or improved process
steps. For some types of pollution these are undoubtedly the most
logical approaches. My interest is in converting waste materials into
feeds. As a neo-Malthusian I believe we will need greater food supplies
before the population density is stabilized. Even today some agricultural
wastes have the potential of providing cheaper feeds than crops grown
primarily for that purpose. For the past eight years we have been
carrying out research on lignified cellulosic wastes or byproducts,
1 Chief, Field Crops Laboratory, Western Utilization Research and
Development Division, ARS, USDA, Albany, California.
383
-------
including alfalfa stems, wheat straw and corn cobs, in cooperation with
the Nebraska Experiment Station; on grass straw from seed production in
cooperation with Oregon State University; and on rice straw in cooperation
with the University of California at Davis. Ruminants and other
herbivorous animals are uniquely adapted to digest the cell wall con-
stituents in these fibrous wastes. But lignin, an indigestible substance,
is associated with the cell wall cellulose and hemicellulose and tends to
decrease their digestibility. The partial crystallinity of cellulose
itself may prevent complete digestion of high fiber feeds.
Research has shown that processing straws with or without chemicals
can modify both lignin and cellulose and increase overall digestibility.
This is important because we know from experience that animals cannot
physically consume enough low digestibility feed to maintain themselves
adequately. For example, unprocessed rice straw is a poor substitute
for dehydrated alfalfa because it contains too little energy as well as
too little protein. However^ based on our present laboratory evidence,
we feel that lignified cellulosic wastes can be upgraded to the point
that they can provide adequate energy for ruminant maintenance, and
even for growth and/or production. Of course protein and other appropriate
supplementation will be necessary to provide proper nutritional balance.
Our laboratory processing experiments have been carried out in
several types of equipment. On the bench scale, we have used bombs of
two different sizes made from copper or stainless tubing with standard
fittings for closures. For larger scale work, we have used the pilot
plant equipment at Dierks Forest, Hot Springs, Arkansas. This facility
can batch process 1-1/2 cubic feet of material at steam pressures of
up to 600 psig.
To test digestibility of processed materials without time consuming
animal tests, we have developed an in_ vitro technique which uses two
commercially available enzyme preparations ('Anozuka Cellulase SS" and
"Pronase," a protease) to digest and solubilize the cellulosic, hemi-
cellulosic and proteinaceous components of potential feedstuffs. This
test correlates very well with standard artificial rumen studies using
rumen fluid from fistulated animals, and with actual in vivo animal
performance tests. A biochemical or biological evaluation of digestibility
is essential because standard chemical analyses do not detect differences
in digestibility of treated and untreated materials. Laboratory tests
show that alkali treatment of grass straw can double the digestibility,
while acid detergent fiber, crude fiber, pentosan, and lignin change
very little.
All straws tested can be improved by treatment. For example,
laboratory treatment of wet rice straw with 3$ by weight of NaOH at
231° for 3 minutes almost doubles the apparent digestibility; corresponding
treatment of barley straw increases apparent digestibility by 120$. We
are presently carrying out intensive laboratory investigations of the
effects that process variables, including alkali concentration, moisture
level, temperature, and reaction time, have on the digestibility of rice
straw. Promising treatments are already being expanded to the pilot
plant scale and enough materials will be prepared for large animal
feeding trials at Davis.
384
-------
A variety of lignified cellulosic wastes like sugar cane bagasse
and oil seed hulls accumulate at processing plants. Since no additional
transportation costs are required to make these materials available for
treatment, they represent a prime source of potential feedstuffs.
Laboratory data show that excellent apparent digestibility increases can
be achieved by processing them with alkali. Even manure digestibility
is greatly improved by alkali treatment. Feed potential of these
materials looks considerably brighter than that of sawdust, probably
because the vegetable products have a. relatively lower lignin content
initially.
Turning now to the second waste problem, waste materials from green
vegetables, in the relatively small Salinas valley of California, about
60,000 tons of cauliflower waste are produced annually. Production is
spread out over 10 months so that some material is available throughout
most of the year. Presently, following removal of the blossoms for the
fresh vegetable market, the fresh cauliflower leaves and stems are
trucked back to the fields from the processing sheds and dumped out on
the ground. The fresh leaves decompose rapidly and so become a breeding
ground for flies and create a serious odor problem. As a result of our
experiences in dehydration of alfalfa, we set up a cooperative project
with a commercial firm to try to solve this particular agricultural
waste problem in a way that might profit both the farmer and the
processor.
Whole cauliflower plants were obtained from two fields near Salinas
and after removal of the blossoms, the entire plants were freeze-dried.
A leaf -stem separation was then made and the respective fractions were
analyzed for xanthophyll, carotene, protein, fat, fiber and ash. The
analyses indicated that the very high xanthophyll and protein content of
the leaf, along with a low fiber content, makes the leaf fraction look
promising indeed as a poultry feed. Xanthophyll is the yellow pigment
in feeds which causes skin pigmentation of broilers and the orange color
of egg yolks. Likewise, the adequate protein and fiber content of the
stem fraction indicates it has good potential as a cattle feed.
The limiting economic factor in the dehydration of cauliflower
waste is its high moisture content. Fresh material varies from 7-5 - 10.0%
solids. This, of course, means almost 100% more water has to be evaporated
to obtain one pound of dehydrated cauliflower than has to be evaporated
from alfalfa to obtain one pound of dehydrated material, because alfalfa
has a higher initial solids content (l.6-2.k%) , To circumvent this
difficulty, the freshly harvested and chopped cauliflower waste was
pressed by means of a sugar cane type of roll-press which we had been
using in our work on wet-fractionation of alfalfa. The finely chopped
cauliflower proved more difficult to press than the more fibrous alfalfa;
however, by this means the solids content of the pressed cauliflower
was increased to 22.0%. Thus about 30$ of the water was removed with a
loss of only lUjf of the soluble solids.
Our next step was to make a trial run with our pilot Heil dryer.
Dehydration conditions employed were similar to those used to dehydrate
alfalfa and grasses. It was found that the leaf and the stem came through
385
-------
the drum with very different moisture contents. This differential was
of benefit in carrying out the next step which was to separate the
dehydrated product into leaf and stem fractions. An excellent leaf and
stem separation was achieved by this means as indicated by the
xanthophyll analysis for the two products. The leaf contained 191 mg/lb
xanthophyll while the stem contained only 51 mg/lb.
Proximate analysis of the meals indicated that the high quality
leaf meal containing up to 28$ protein with a low fiber content (12.5$)
should be excellent for poultry rations. In addition, the stem meal
fraction for ruminant feeds was only slightly lower in protein and
slightly higher in fiber. It compared very favorably with a good quality
20% protein dehydrated alfalfa meal with a 20% fiber content.
We found that pressing the chopped cauliflower prior to dehydration
actually reduced the through-put time in the dehydrator by about one
third. This gave a good retention of both carotene and xanthophyll in
the pressed leaf fraction along with a suitable protein and fiber content
for a high grade poultry feed. The corresponding stem meal from this
preparation had the necessary protein and fiber content to make it an
excellent cattle feed. An enzyme digestibility test run on the stem
fraction indicated 86.3$ digestibility, which when compared to alfalfa
stem, at about 50-60$ digestibility, looks promising indeed as a cattle
feed.
Presently cattle feeding trials are oeing carried out on the stem
fraction. Broiler pigmentation studies are also being conducted on a
preparation of the leaf fraction to ascertain availability and potency
of the xanthophylls present. These studies, combined with our work on
the technique and economics of dehydrating fresh cauliflower waste, will
determine the feasibility of solving a perplexing fresh-vegetable-
processing pollution problem.
Economic evaluations of treated straw products and cauliflower
waste products are being carried out by computer using parametric
linear programming techniques.
386
-------
RICE HULL UTILIZATION
*y
Milton Staackmann^-
INTRODUCTION
The disposal of rice hulls presents a major problem to the rice
industry, since industrial applications for the hulls currently consume
only a small fraction of the total. Nationwide, nearly a million tons
of hulls are produced annually and about 20% of these originate in the
Sacramento Valley area of California. Present methods of disposal
include open-air burning and dumping. Such methods are becoming ever
more unattractive due to air pollution in the case of burning and
because of increasing hauling and landfill charges in the case of
dumping.
Research activities aimed at finding potential uses for rice hulls
have been going on for the past sixty years. Most of these efforts
have attempted to take advantage of a single property of the hulls
themselves (such as the fuel value or abrasive character) or of the ash
formed by burning the hulls (as an absorbent or a ceramic raw material).
In addition, much of the past work considered technical feasibility
only, paying insufficient attention to economics. URS Research Company
has completed a study for the US Department of Agriculture examining
the technical and economic feasibility of various utilization concepts
involving more than one of the properties of the hulls. The general
areas examined included the following:
Concepts making use of both the fuel value of the organic
(cellulosic) portion of the hulls and the silica content;
Concepts utilizing both the structural characteristics
and the unique texture of the rice hulls;
Concepts combining both of the above.
The most promising approaches to solution of the problem of rice
hull disposal fall into two basic categories. The first category
includes methods for using the heating value and the silica content of
the rice hulls as elements in the manufacture of industrial products
1
URS Research Company, San Mateo, California.
387
-------
such as Portland cement, water glass, and porous silicate structural
materials. The other category includes the chemical or physical bonding
of rice hulls into board for architectural use. One of the most
attractive approaches appears to be a manufacturing complex which
converts part of the hulls into water glass which is used in the
bonding of the remainder of the hulls into panel board.
CONCEPTS INVESTIGATED
Portland Cement
This concept makes use of both the heating value and silica
content of the rice hulls. Hopefully, a standard Portland cement
manufacturing facility could be modified to permit substituting the
rice hulls for a portion of the silica supplied as a raw material and
for most of the hydrocarbon fuel used in manufacture of the cement.
It appears that "by merely charging the rice hulls into the rotary
kiln of the plant, some difficulties may arise. During burning of the
rice hulls to produce the heat necessary for chemical conversion of the
various raw materials into cement, it is possible that considerable
quantities of incompletely burned rice hulls could be incapsulated by
the fusing materials and complete combustion of the hulls would thereby
be prevented. In this manner, the total heating value of the hulls
would be unavailable, and the resultant cement would be contaminated
with cellulose and char. Due to this difficulty, an operable cement
plant utilizing rice hulls may have to be of unconventional design, or
additional equipment may be required to completely burn the hulls
before entry into the kiln. Such alternative procedures are still
considered to be technically feasible, since furnaces for complete
combustion of rice hulls have been designed, and unconventional processing
techniques such as fluidized bed reactors are used in other manufacturing
processes.
The economic feasibility of using rice hulls in cement manufacture
is greatly dependent upon the size of the plant being considered. It
appears that new plants must have annual capacities ranging from five
to ten million barrels in order to compete economically. Such a plant
has very stringent requirements for the total amount of rice hulls
required each year and the fraction of silica in the product cement
which may be supplied by the hulls. It appears that a plant based on
using the typical amounts of rice hulls available in any one geographic
area would be limited in annual production of cement to two to three
million barrels. The economics of production of ce'ment utilizing rice
hulls are, therefore, not favirable.
Water Glass
The production of water glass using rice hulls as a source of fuel
and silica in a conventional manner could experience the same difficulties
noted in the manufacture of cement. That is, significant amounts of
rice hulls could be prevented from undergoing complete combustion in
the process. The requirement for additional equipment for use in the
conventional manufacture of water glass or for an unconventional manu-
facturing process is evident.
388
-------
The most promising unconventional approach to manufacturing water
glass would utilize the Zimmerman process (wet air oxidation). The
Zimmerman process is already being successfully used in over 30 sewage
disposal plants across the country and for wet oxidation of paper
mill wastes, antibiotic residues, manure, etc. It appears that water
glass manufacturing plants using rice hulls would be competitive in
size with existing plants.
Porous Silicate Materials
This concept employs bonding the rice hull ash to form materials
with a wide range of specifications. The basic idea was initiated some
years ago to produce porous silica insulating refractory bricks. This
product is a specialized material and requires close control of many
elements of its manufacture. Whereas it appears to be a profitable
process, complete penetration of the total market for this product
(rather unlikely) would still provide an outlet for, at most, only a
few percent of all rice hulls.
The manufacture of products with less stringent specifications
appears more favorable. Such products could include the following:
Building blocks-,
Architectural insulating slabs;
Pipe lagging;
Aggregate for plaster, lightweight concrete, or concrete
building blocks.
Several processes for manufacture of such products have been investigated.
They fall into two basic types. The first type consists of combining
the hulls with a suitable fluxing agent and burning the hulls to produce
the heat required to achieve bonding of the silica residue of the rice
hulls. This process has the same shortcoming as the cement and water
glass concepts (some hulls could be incompletely burned), but in this
case the product specifications may permit it. The other process type
consists of burning the hulls to produce steam and silica with sub-
sequent use of the steam in an autoclave to bond the silica blended
with fluxing agent.
The latter process, in modified form, is currently used extensively
in Europe to produce calcium silicate bonded structural materials. Rice
hulls have been burned to produce steam as the means of hull disposal with
no attempt to recover the ash. Such boilers have become uncompetitive
with conventional steam production in recent years.
In the process outlined above, the steam would be used on location
and subsequent use would be made of the hulls. This process appears,
therefore, to be economically feasible.
389
-------
Board for Architectural Use
This concept makes use of the unconverted rice hulls (not the ash)
bonded together to produce an architectural board with attractive
textural qualities.
Processes for physically bonding the rice hulls include producing
vulcanized cellulose fiber and bonding the hulls together in a matrix
of sodium silicate. Difficulties in achieving high bond strength and
adequate water resistance are present for these processes, but it is
felt that such difficulties can be overcome by means of limited technical
development programs. A number of other processes have been considered
to achieve chemical bonding of the rice hulls. Such processes include
bonding by means of ester or ether linkages between cellulose molecules.
The primary difficulty encountered in examining these methods of
bonding is excessive degradation of the cellulose.
For a successful process of this type, preliminary treatment of
the hulls may be necessary to prevent such degradation. An overall
system requiring pretreatment of the hulls appears to be
economically unfavorable.
Combined Concept Plant
After review of the above concepts, it appears that the opportunity
for combining concepts into one manufacturing complex deserves further
examination. The combined plant would have two main processing areas.
One area would consist of a sodium silicate manufacturing facility
producing the sodium silicate using rice hulls as one of the raw
materials. The output of this facility would be used in the other
manufacturing area where the remainder of the rice hulls would be
bonded into architectural board using sodium silicate as the binder.
Operational economy would accrue from reduction of raw material
input to a few basic materials, chiefly rice hulls and sodium carbonate,
as well as from the savings in packaging, shipping, and sales cost of
sodium silicate that would result from its manufacture in a captive
facility. In addition, this plant would be very flexible since two
independently salable products would be made.
CONCLUSIONS
Investigations by UBS Research Company to date have been primarily
limited to paper studies of these concepts with very little laboratory
work having been performed. Results which have been obtained indicate
that several of the concepts are deserving of further examination.
Recommended future work would resolve some of the unanswered technical
problems by laboratory studies followed by pilot plant evaluations of
the most deserving concepts. There is great hope that this major
agricultural waste problem will be turned into a profit for the
industry and will illustrate the rewards which can be achieved by
application of imaginative systems engineering to pollution problems.
390
-------
REGISTRATION LIST
GEORGE ABEL
Northwest Region, FWQA
501 Pittock Block
Portland, Oregon 97205
THOMAS S. ALLEN
Ethyl Corporation
Ethyl Tower
Baton Rouge, La. 70821
STEVE M. ANDERSON
U.S.P. Corporation
Box 230, 570 Race St.
San Jose, Ca. 95103
JOHN L. ARNQUIST
State of Washington Water
Pollution Control Commission
3700 Rainier Avenue South
Seattle, Washington 98144
N. R. ARTHUR
ITT Electrophysics Labs.
3355 52nd Avenue
Hyattsville, Md. 20781
DR. JORG AUGUSTIN
University of Idaho
Branch Experiment Station
Aberdeen, Idaho 83210
GERALD BABCOCK
Stayton Canning Co.
Stayton, Oregon 97383
JAMES F. BARBOUR
Oregon State University
Ag.-Chem. Department
Corvallis, Or. 97331
TIBOR BATHONY
Early California Foods
1833 Blake Avenue
Los Angeles, Ca. 90039
A. C. BATTALION
United Flav-R-Pac Growers, Inc.
P. 0. Box 3288
Salem, Oregon 97310
C. L. BEARDSLEY
Diamond Fruit Growers, Inc.
P. 0. Box 180
Hood River, Or. 97031
ERNEST W. BECK, JR.
Spreckels Sugars
#2 Pine St.
San Francisco, Ca. 94111
BUD BENNETT
FMC Corporation
1185 Coleman St.
Santa Clara, Ca. 95052
DON H. BERRYHILL
Del Monte Corporation
P. 0. Box 150
Vancouver, Wa. 98660
CLARENCE L. BOLT
Prosser Packers, Inc.
1001 Bennet Avenue
Prosser, Wa. 99350
JAMES R. BOYDSTON
Pacific Northwest Water Laboratory
FWQA
200 S.W. 35th St.
Corvallis, Or. 97330
RONALD W. BRENTON
Beet Sugar Development Foundation
P. 0. Box 119
Longmont, Colorado 80501
DR, K. M. BRINK
Colorado State University
Dept. of Horticulture
Fort Collins, Colorado 80521
DANIEL E. BROOKS
National Canners Association
1600 S. Jackson St.
Seattle, Wa. 98144
WILLARD R. BROSZ
Green Giant
232 Regency Rd.
LeSueur, Minn. 56058
391
-------
STEPHEN C. BROWN
Envirotech Systems, Inc.
100 Valley Drive
Brisbane, Ca. 94005
C. J. BRUDE
Cudahy Company
100 W. Clarendon
Phoenix, Arizona 85013
ALLYN BURIC
FMC Corporation
Prudential Plaza
Chicago, 111. 60067
HOWARD L. BURKHARDT
Idaho Department of Health
Yellowstone Plaza, Suite D
Pocatello, Idaho 83201
ROBERT J. BURM
Pacific Northwest Water Laboratory
FWQA
200 S.W. 35th St.
Corvallis, Or. 97330
ROY E. CARAWAN
North Carolina State University
Department of Food Science
Raleigh, North Carolina 27607
BOB CARLILE
Washington State University
Box 2454 C.S.
Pullman, Washington 99163
DR. DALE A. CARLSON
School of Civil Engineering
University of Washington
Seattle, Washington 98105
JAY A. CEDERGREEN
Evergreen Frozen Foods
P. 0. Box 151
Snohomish, Wa. 98290
DAVID B. CHARLTON
Charlton Laboratories
P. 0. Box 1048
Portland, Or. 97207
ART CHRISTIANSEN
Stayton Canning Co.
Stayton, Oregon 97383
LEE R. CHUGG
239 Heritage Court
Woodburn, Or. 97071
DR. BROOKS D. CHURCH
North Star Research Inst.
3100 38th Avenue South
Minneapolis, Minn. 55406
DR. J. PETER CLARK
USDA - ARS - WURDD
800 Buchanan Street
Albany, Ca. 94710
WILLIAM F. CLYDE
Lamb-Weston, Inc.
12977 S.W. 66th St.
Portland, Or. 97229
LEE G. CORDIER, JR.
Campbell Soup Company
43rd & Franklin Blvd.
Sacramento, Ca. 95807
DAVID L. CUMMINGS
Tri-Aid Sciences, Inc.
3017 Plank Road, RD #2
Macedon, New York 14502
JOHN A. DASSOW
Bureau of Comm. Fisheries
2725 Montlake Blvd. E.
Seattle, Wa. 98102
ROGER A. DeCAMP
National Canners Association
1600 S. Jackson St.
Seattle, Wa. 98144
THOMAS M. DONOGHUE
Air & Water News
330 W. 42nd St.
New York, New York 10036
KENNETH A. DOSTAL
Pacific Northwest Water Lab.
FWQA
200 S.W. 35th St.
Corvallis, Or. 97330
392
-------
RALPH E. DREWITZ
ITT Continental Baking Co.
39 Fara Drive
Stamford, Conn. 06905
LARRY ESVELT
1100 6th St.
Albany, Ca. 94706
JOHN FILBERT
Cornell, Rowland, Hayes &
Merryfield
1600 S.W. Western Blvd
Corvallis, Or. 97330
WILLIAM F. FILZ
North Pacific Canners & Packers
5200 S.E. McLaughlin Blvd.
Portland, Or. 97202
LEO A. FITZSIMON
Aquatair West
700 5th St.
Huntington Beach, Ca. 92646
E. D. FLEMMING
General Foods Corp.
Birdseye Division
P. 0. Box 36
Woodburn, Or. 97071
TOM H. FORREST
Chicago Pump, FMC Corp.
2008 Bennett Avenue
Evanston, 111. 60201
JAMES H. FISCHER
Beet Sugar Dev. Foundation
P. 0. Box 538
Fort Collins, Colorado 80521
CHARLES L. FRAZIER
Allen & Hoshall
65 McCall
Memphis, Tenn. 38116
JAMES FULTON
Stevens, Thompson & Runyan
700 Plaza 600
Seattle, Wa. 98101
R. PURER
Dole Company
P. 0. Box 351
Salem, Or. 97308
JACK GANOE
Pronto Pacific, Inc.
P. 0. Box 1029
Moses Lake, Wa. 98837
L. C. GILDE
Campbell Soup Company
375 Memorial Avenue
Camden, N. J. 08110
DR. DAVID R. V. GOLDING
Dole Company
P. 0. Box 3380
Honolulu, Hawaii 96801
EUGENE GOLDMAN
Bechtel Corp.
50 Beale Street
San Francisco, Ca. 94115
JOHN E. GOMENA
Lamb-Weston, Inc.
2017 Lloyd Center
Portland, Or. 97232
DR. WILBUR A. GOULD
Ohio State University
Columbus, Ohio 43210
JOHN L. GRAHAM
Cornell, Howland, Hayes & Merryfield
1600 S.W. Western Blvd.
Corvallis, Or. 97330
ROBERT P. GRAHAM
USDA - ARS - WURDD
800 Buchanan Street
Albany, Ca. 94710
LLOYD M. GRAMES
The Eimco Corp.
P. 0. Box 300
Salt Lake City, Utah 84110
DON E. GRAY
Gray & Osborne
228 S. 2nd Street
Yakima, Wa. 98901
ALEX GRINKEVICH
Hunt-Wesson Foods, Inc.
1389 Rosborough Drive
Placentia, Ca. 92670
ED GRODY
Jewel Co.
1955 North Avenue
Melrose Park, 111. 60160
393
-------
GILBERT GROFF
Clark & Groff Engineers, Inc.
3276 Commercial St. S.E.
Salem, Or. 97302
KRISTIAN GUTTORMSEN
Ruskin Fisher & Assoc.
17962 Midvale Ave. North
Seattle, Wa. 98133
BILL HARRIOTT
University of Arizona
P. 0. Box 631
Mesa, Arizona 85202
HERBERT H. HART
Snokist Growers
2506 Terrace Heights Rd.
Yakima, Wa. 98901
HARRY HATCH
Carnation Co.
5045 Wilshire Blvd.
Los Angeles, Ca. 90036
ROBERT E. HATTEN
Gray & Osborne
228 S. 2nd St.
Yakima, Wa. 98901
C. R. HAVIGHORST
Food Engineering Magazine
1762 (B) Foothill St.
South Pasadena, Ca. 91030
RAY HEIDENRICH
Eugene Fruit Growers Assn.
675 E. 39th PI.
Eugene, Oregon 97405
DR. DAVID W. HILL
Southeast Water Laboratory
FWQA
College Station Road
Athens, Georgia 30601
BERNARD S. HORTON
Abcor, Inc.
341 Vassar St.
Cambridge, Mass. 02139
PETER HOULE
Swift & Co., R&D Center
1919 Swift Drive
Oakbrook, 111. 60521
PAUL V. HOURIET, JR.
Allis-Chalmers Manuf. Co.
2351 North Mayfair Road
Wauwatosa, Wise. 53226
ROGER W. HUIBREGTSE
The Larsen Company
P. 0. Box 1127
Green Bay, Wise. 54305
ROBERT ISAAC
Blue Lake Packers, Inc.
1192 Ruge St., N.W.
Salem, Oregon 97304
M. L. JACKSON
University of Idaho
Coordinator of Research
Moscow, Idaho 83843
FRED JERMANN
Bumble Bee Sea Foods
P. 0. Box 60
Astoria, Oregon 97103
DR. JAMES J. JEZESKI
Montana State University
Dept. of Botany & Microbiology
Bozeman, Montana 59715
E. L. JOHNSON
Food Chemical & Research Labs., Inc.
4900 9th N.W.
Seattle, Wa. 98107
FRANK R. JONES
Pennwalt Corporation
2901 Taylor Way
P. 0. Box 1297
Tacoma, Wa. 98401
DR. RICHARD H. JONES
Environmental Engineering, Inc.
2324 S.W. 34th St.
Gainesville, Florida 32601
394
-------
ALLEN M. KATSUYAMA
National Canners Assoc.
1950 6th Street
Berkeley, Ca. 94710
H. GEORGE KEELER
Industrial Pollution Cont. Br.
Federal Water Quality Admin.
Washington, D.C. 20242
WILLIAM K. KITCHIN
Canners League of California
9 1st St., Room 720
San Francisco, Ca. 94105
DR. GEORGE 0. KOHLER
USDA - ARS - WURDD
800 Buchanan St.
Albany, Ca. 94710
WILLIAM F. KOHLER
Conservation Engineering, Inc.
27205 Marine View Drive South
Kent, Washington 98031
WILLIAM J. LACY
Industrial Pollution Cont. Br.
Federal Water Quality Admin.
Washington, D. C. 20242
ROBERT LANGHOFF
Blue Lake Packers, Inc.
963 Kingwood Dr. N.W.
Salem, Oregon 97304
DR. DON H. LARSEN
Brigham Young University
Microbiology Dept.
Provo, Utah 84601
ARNOLD LOCKHART
Ocean Spray Cranberry
Markham Star Route
Aberdeen, Wa. 98520
DON R. LONG
Aquatair West
700 5th St.
Huntington Beach, Ca. 92646
K. R. MAJORS
U. S. Dept. of Agriculture
1711 E. Maple Ridge Dr.
Peoria, 111. 61614
DR. GEORGE MALLAN
Garrett R&D Co., Inc.
1855 Carrion Road
La Verne, Ca. 91750
F. M. MARQUIS
Albany Frozen Foods, Inc.
P. 0. Box 609
Albany, Or. 97321
TED McCAFFRAY
National Fruit Canning Co.
Box 9366
Seattle, Wa. 98109
DR. PERRY MCCARTY
Stanford University
Stanford, Ca. 94305
R. J. McCULLOCH
University of Wyoming
Division of Biochemistry
Laramie, Wyoming 82070
ALTON McCULLY
Eugene Fruit Growers Assn.
1638 Orchard St.
Eugene, Or. 97403
JOHN H. MCDONALD
WATCOA
N.W. 31st
Portland, Oregon
COLIN McKAY
R. J. Reynolds Foods, Inc.
Glenwood Road
Weston, Conn. 06880
ROBERT E. MEANS
Bouillon, Christofferson & Schairer
505 Washington Bldg.
Seattle, Wa. 98101
395
-------
ALAN MEFFORD
Pronto Pacific, Inc.
P. 0. Box 1029
Moses Lake, Wa. 98837
FRED MELHASE
Allen Fruit Co., Inc.
P. 0. Box 352
Newberg, Oregon 97132
WALTER A. MERCER
National Canners Association
1950 6th Street
Berkeley, Ca. 94710
J. K. P. MILLER
Garrett Res. & Dev.
1855 Carrion Road
La Verne, Ca. 91750
Co.
HOWARD P. MILLEVILLE
Oregon State University
Dept. Food Science & Technology
Corvallis, Or. 97331
WILLIAM A. MONGE
Libby, McNeill & Libby
520 S. El Camino Real
San Mateo, Ca. 94402
A. J. MONTA
Welch Foods
Westfield, N.Y.
14787
DAVE MOODY
Prosser Packers, Inc.
1001 Bennett Avenue
Prosser, Wa. 99350
WILLIAM B. MOORE
Moore, Wallace & Kennedy, Inc.
1915 First Avenue
Seattle, Wa. 98101
ROY MOSER
University of Hawaii
Honolulu, Hawaii 96822
DANIEL V. NEAL
Washington State WPCC
321 Button Bldg.
Spokane, Wa. 99204
MICHAEL E. NEARY
Birdseye Division, General Foods
2101 Crawford Dr.
Walla Walla, Wa. 99362
J. L. NELSON
Green Giant Co.
509 Lee St.
Dayton, Wa. 99328
MORTON NEMIROFF
Dole Company
P. 0. Box 3380
Honolulu, Hawaii 96801
DR. H. P. NICHOLSON
Southeast Water Laboratory
Federal Water Quality Admin.
College Station Road
Athens, Georgia 30601
JIM GATES
J. R. Simplot Co., Inc.
P. 0. Box 1059
Caldwell, Idaho 83605
DOUG OLESEN
Battelle-Northwest
P. 0. Box 999
Richland, Wa. 99352
CARL OLSON
Stokely Van Camp, Inc.
P. 0. Box 456
Mt. Vernon, Wa. 98273
DAVID PAHL
Northwest Food Processors
2828 S.W. Corbett
Portland, Or. 97201
ROBERT E. PAILTHORP
Cornell, Howland, Hayes & Merryfield
1600 S.W. Western Blvd.
Corvallis, Or. 97330
396
-------
WARREN G. PALMER
FMC Corporation
1185 Coleman Avenue
Santa Clara, Ca. 95052
GEORGE R. PANKEY
Del-Pak Corp.
2245 Elmwood Dr.
Corvallis, Or. 97330
BOB PARODI
Tri Valley Growers
P. 0. Box 948
Modesto, Ca. 95352
QUINTIN P. PENISTON
Food Chem. & Res. Labs.
4900 9th N.W.
Seattle, Wa. 98107
CHARLES 0. PERKINS
New England Fish Company
Pier 89
Seattle, Wa. 98119
G. G. PERKINS
Dole Company
P. 0. Box 351
Salem, Or. 97308
GRANVILLE PERKINS
Artichoke Industries, Inc.
11599 Walsh St.
Castroville, Ca. 95012
ED PITKIN
Eugene Fruit Growers Ass'n.
Eugene, Oregon 97401
DR. KAREL POPPER
USDA - ARS - WURDD
800 Buchanan St.
Albany, Ca. 94710
JOHN C. POSTON
Gray & Osborne
228 S. 2nd St.
Yakima, Wa. 98901
E. G. PUGSLEY
Colorado Department of Health
Water Pollution Control Division
6902 S. Jackson Way
Littleton, Colorado 80120
JACK W. RALLS
National Canners Association
1950 6th St.
Berkeley, Ca. 94710
GARY H. RICHARDSON
Utah State University
Dept. of Food Science
Logan, Utah 84321
LYNN B. RICHMOND
Twin City Foods, Inc.
Box 647
Lewiston, Idaho 83501
GLENN A. RICHTER
Cornell, Rowland, Hayes & Merryfield
1600 S.W. Western Blvd.
Corvallis, Or. 97330
CHRIS D. ROBERTS
Contadina Foods
P. 0. Box 29
Woodland, Ca. 95695
J. ROBERTS
Washington State University
Dept. of Agricultural Engineering
Pullman, Wa. 99163
C. DAN ROBISON
Washington State WPCC
P. 0. Box 829
Olympia, Wa. 98501
ROBERT M. ROODEN
Stanford Research Institute
333 Ravenswood Avenue
Menlo Park, Ca. 94025
CHARLES J. ROGERS
U. S. Public Health Service
Bureau of Solid Waste Management
5555 Ridge Avenue
Cincinnati, Ohio 45213
397
-------
WALTER W. ROSE
National Canners Association
1950 6th St.
Berkeley, Ca. 94710
C. FARRELL RUPPERT
Beech-Nut, Inc.
Church Street
Canajoharie, New York 13317
PAUL RUSSELL, JR.
Harnish & Lookup Associates
1319 Summit Dr.
Newark, New York 14513
JOHN SACKLIN
Nalley's Fine Foods
3303 S. 35th St.
Tacoma, Wa. 98409
HERBERT A. SCHLESINGER
Lockwood Greene Engrs.
200 Park Avenue
New York, N. Y. 10017
LOWELL SENTER
Clermont West, Inc.
666 E. Main St.
Hillsboro, Ore. 97123
ROBERT D. SHANKLAND
Pacific Northwest Water Lab.
FWQA
200 S.W. 35th St.
Corvallis, Or. 97330
EVAN P. SHEA
Midwest Research Institute
425 Volker Blvd.
Kansas City, Mo. 64110
GERALD L. SHELL
The EIMCO Corp.
Box 300
Salt Lake City, Utah 84110
ROGER SHERWOOD
Department of Environmental Quality
State of Oregon
1400 S.W. 5th
Portland, Or. 97201
HOWARD H. SHOCKEY
National Fruit Product Co., Inc.
P. 0. Box 609
Winchester, Va. 22601
TRAVER J. SMITH
Magnuson Engineers, Inc.
1010 Timothy Dr.
San Jose, Ca. 95133
WALTER L. SMITH
Stayton Canning Company Coop.
930 W. Washington St.
Stayton, Or. 97383
WILLIAM G. SMITH
Sun-Rype Products Ltd.
1165 Ethel St.
Kelowna, B.C.
MICHAEL R. SODERQUIST
Oregon State University
Dept. Food Science & Tech.
Corvallis, Oregon 97331
MILTON STAACKMANN
URS Research Co.
1811 Trousdale Dr.
Burlingame, Ca. 94115
MARVIN P. STEINBERG
University of Illinois
109 S. Wing Hort Field Lab.
Urbana, Illinois 61801
DONALD STEPHENS
Libby, McNeill & Libby
200 S. Michigan Ave.
Chicago, 111. 60604
RICHARD W. STERNBERG
National Canners Association
1133 20th St. N.W.
Washington, D.C. 20036
TREVOR G. STEVENS
Canadian Canners Limited
1101 Walkers Line
Burlington, Ontario, Canada
CHARLES A. STEVENSON
C-B Foods
P. 0. Box 670
Rochester, N.Y. 14602
398
-------
HAROLD STONE
Libby, McNeill i Libby
200 S. Michigan
Chicago, 111. 60604
DR. JURGEN STRASSER
FMC Corporation
1185 Coleman Avenue, Box 580
Santa Clara, Ca. 95052
LEALE E. STREEBIN
University of Oklahoma
Norman, Oklahoma 73069
GARY STROH
Stokely Van Camp, Inc.
P. 0. Box 98
Emmett, Idaho 83617
DR. DONALD J. STUCKY
Allis-Chalmers Mfg. Co.
P. 0. Box 7247
Milwaukee, Wisconsin 53213
CARL G. SWANSON
Cudahy Co.
100 W. Clarendon Ave.
Phoenix, Arizona 85013
THOMAS T. TAKEOKA
Alaska Packers Assn., Inc.
P. 0. Box AA
Elaine, Wa. 98230
WILLIAM F. TALBURT
Southeastern Agr. Res. Lab.
USDA - ARS
P. 0. Box 5677
Athens, Ga. 30604
DONALD J. THIMSEN
General Mills, Inc.
3020 Atwood Dr.
Minnetonka, Minn. 55343
STEVE THURLOW
Stephen Thurlow Co.
116 W. Harrison St.
Seattle, Wa. 98119
ERNIE TODD
Chef Reddy Foods
P. 0. Bpx 607
Othello, Wa. 99344
DON TRANDUM
Stephen Thurlow Co.
116 W. Harrison St.
Seattle, Wa. 98119
RONALD A. TSUGITA
National Canners Association
1950 6th St.
Berkeley, Ca. 94710
D. M. UPDEGRAFF
University of Denver
Denver Res. Inst.
Denver, Colorado 80210
DIRK M. VAN WOERDEN
Moore, Wallace & Kennedy, Inc,
1915 First Avenue
Seattle, Wa. 98101
CAESAR VAN ZUIDEN
Stauffer Chemical Co.
636 California
San Francisco, Ca. 94030
DR. JOHN W. VENNES
University of North Dakota
Department of Microbiology
Grand Forks, N. D. 58201
ROBERT W. VIVIAN
Stevens, Thompson & Runyan
1984 N.W. Overton
Portland, Or. 97209
FRED VOIT
Patterson Frozen Foods
Patterson, Ca. 95363
DR. JOSEPH R. WAGNER
USDA - ARS - WURDD
800 Buchanan St.
Albany, Ca. 94710
399
-------
EVERETT R. WALFORD
USDA - ARS - WUKDD
Puyallup, Wa. 98371
DAVID J. WARD
USDA - S&ES
Room 331E
Washington, D.C. 20250
ROBERT E. WARD
Tree Top, Inc.
P. 0. Box 248
Selah, Wa. 98942
RAY WARREN
Contadina Foods, Inc.
A Subsidiary of Carnation Co.
5045 Wilshire Blvd.
Los Angeles, Ca. 90036
RALPH N. WATTERS
American Frozen Food Institute
1838 El Camino Real
Burlingame, Ca. 94010
DR. EUGENE WEISBERG
Envirotech Systems, Inc.
100 Valley Dr.
Brisbane, Ca. 94005
W. JAMES WELLS, JR.
Bell, Galyardt & Wells
220 Hillcrest Bldg., Ralston
Omaha, Nebraska 68127
BILL G. WHITE
Early California Foods
P. 0. Box 71
Visazia, Ca. 93278
PROFESSOR J. WAYNE WHITWORTH
New Mexico State University
Agronomy Department
P. 0. Box 3Q
Las Cruces, New Mexico 88001
JOHN R. WIEGMANN
Hercules Incorporated
910 Market St.
Wilmington, Delaware 19899
MILES WILLARD
Consultant
3067 Gustafson Circle
Idaho Falls, Idaho 83401
TED A. WILLIAMSON
Oklahoma State Health Dept.
3400 N. Eastern
Oklahoma City, Okla. 73105
JACK L. WITHEROW
Robert S. Kerr Water Res. Ctr.
Federal Water Quality Admin.
P. 0. Box 1198
Ada, Oklahoma 74820
GEORGE K. YORK
University of California
Davis, Ca. 95616
RALPH A. YOUNG
University of Nevada
Reno, Nevada 89507
LEE H. ZIMMERMAN
Rex Chainbelt, Inc.
5700 N.E. Nassaco St.
Portland, Or. 97213
400
V. s- GOVERNMENT PRINTING OFFICE : 1971 O - 411-754
------- |