EPA 660/2-74-061
June 1974
                         Environmental Protection Technology Series
     Shrimp Canning Waste
     Treatment Study
                                   Office of Research and Development
                                   U.S. Environmental Protection Agency
                                   Washington, D.C. 20460

-------
             RESEARCH REPORTING SERIES
Research  reports of the  Office  of  Research   and
Monitoring,   Environmental  Protection Agency,  have
been grouped into five series.   These  five  broad
categories  were established to facilitate further
development   and  application   of   environmental
technology.    Elimination   of traditional grouping
was  consciously  planned   to  foster   technology
transfer   and  a  maximum   interface  in  related
fields.   The five series are:

   1.  Environmental Health Effects Research
   2.  Environmental Protection Technology
   3.  Ecological Research
   H.  Environmental Monitoring
   5.  Socioeconomic Environmental Studies

This report  has been assigned to the ENVIRONMENTAL
PROTECTION   TECHNOLOGY   series.    This   series
describes   research   performed  to  develop   and
demonstrate    instrumentation,     equipment     and
methodology   to  repair  or  prevent environmental
degradation  from point and  .non-point  sources  of
pollution.  This work provides  the new or improved
technology  required for the control and treatment
of pollution sources to meet environmental quality
standards.
                   EPA REVIEW NOTICE
This report has "been reviewed "by the Office of Research and
Development, EPA, and approved for publication.  Approval
does not signify that the contents necessarily reflect the
views and policies of the Environmental Protection Agency,
nor does mention of trade names or commercial products
constitute endorsement of recommendation for use.
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.O. 20402 - Price $2.05

-------
                                 EPA-660/2-74-061
                                 June 1974
            SHRIMP CANNING

          WASTE TREATMENT STUDY
                    By

              A. Frank Mauldin
                A. J.  Szabo
           Program Element 1BB037
            Roap/Task 21 ALG-33
              Project S 800904
               Project Officer

               Robert L. Miller
        Environmental Protection Agency
             1600 Patterson Ave.
             Dallas, Texas 75201
                Prepared for
   OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
         WASHINGTON, D.C. 20460

-------
                                  ABSTRACT

This study reports on work which characterized shrimp canning plant wastewaters
and operated a pilot dissolved air flotation treatment system during the 1972 and
1973 canning seasons.

The Gulf Coast shrimp canning plants all utilize the same processing equipment
from which the wastewater flows occur.  The waste  handling is complicated by:
(a) the relatively small size of the plants; (b) the lack of prior industry experience
in treating wastes; (c) the plants operate only during short, intermittent seasons
regulated by State conservation agencies and by the economical availability of
shrimp; and (d) the BOD of the effluent waste is approximately 5 to 6 times as
strong as domestic waste.

A wastewater survey was conducted at several canneries during  the  1972 summer
season,  the 1972 fall season and the 1973 summer season.  Water used for each
process within the plant was metered and the wastewater was tested for biological,
chemical and physical characteristics.

Pilot screening studies were conducted at the study  plant during the fall, 1972
and the summer,  1973, seasons.  Tangential, rotary and vibrating screens were
tested.  All the screen types performed approximately the same  with each having
its  advantages and disadvantages.  The tangential screen removed heads and shells
from peeler wastewater very efficiently; however, blinding did occur with the
total composite wastewater.

A 272 m3/day (SOgpm) dissolved air flotation pilot plant with chemical addition
and pH control was tested at the study plant during  the fall, 1972, and summer,
1973, seasons.   Jar tests were performed during the summer, 1972, season and an
effective coagulant, polymer and pH  combination was established.

A pilot basket centrifuge was tested for concentration of skimmings.  The volumes
were reduced to 25 percent of the original and the concentration by percent dry
solids was increased  approximately 2 times.

This study demonstrated that: (1) the waste poundage discharged per pound of raw
shrimp processed is similar in most Gulf Shrimp Canning plants,  (2) screening re-
moval of heads and  shells can  be performed efficiently and with few operational
problems, and  (3) air flotation showed  promise as a wastewater treatment method.
When performing properly, air flotation treatment efficiencies were good; how-
ever, the operation was sensitive and treatment efficiencies that can be expected
on a plant scale remain to be demonstrated.
                                      ii

-------
                                 CONTENTS

                                                                 Page

Abstract                                                           ii

List of Figures                                                     iv

List of Tables                                                      v

Acknowledgements                                                 vi

Sections

I       Conclusions                                                 1

II      Recommendations                                            2

III     Introduction                                                 3

IV     Literature Review                                            9

V      Study Objective and Approach                               31

VI     Wastewater Surveys                                         33

VII    Bench Scale Studies                                         52

VIII    Pilot Scale Studies                                          69

IX     Discussion   '                                               97

X      Bibliography                                               108

XI     Appendices                                               110

       Appendix A      Bench Scale Test Procedures                111
       Appendix B      Laboratory Analytical  Methods             115
       Appendix C      Cost Study Data                           120
       Appendix D      Member Canners
                          American Shrimp Canners Association     127
                                     • * >
                                     in

-------
                              LIST OF FIGURES
No.
Page
  1       General Process Schematic                                  5
 2       Photograph of Shrimp Peeler and Cleaner                      6
 3       Photograph of Shrimp Batch Blancher                          6
 4       Photograph of Shrimp Deveiner - Vein Removal Drum           7
 5       Photograph of Shrimp Deveiner - Vein Exposing Razors         7
 6       Air Flotation Arrangements                                 25
 7       Plant A Layout - Wastewater Drainage System                35
 8       Wastewater Sample Stations - Raw Shrimp Processing          36
 9       Wastewater Sample Stations - Shrimp Canning                37
10       Wastewater Sample Stations - Raw Receiving                 38
11       Water Flow Stations - Wastewater Survey                     40
12       Acid  Precipitation of Shrimp Canning Wastewater             54
13       Bench Scale Flotation Bomb                                 56
14       Sludge Drainage - DAF Run 5 Preliminary                    59
15       Sludge Drainage - DAF Run  12 Preliminary                   60
16    "^Sludge Drainage - DAF Run 5                               61
17       Sludge Drainage - DAF Run 8                               62
18       Sludge Drainage - DAF Run  12                              63
19       Sludge Drainage - DAF Run  13                              64
20       Sludge Drainage - DAF Run  14                              65
21       Sludge Drainage - DAF Run  15                              66
22       Photograph of Pilot Tangential  Screen                       71
23       Photograph of Pilot Rotary Screen                            71
24       Photograph of Pi lot DAF Unit                               81
25       Photograph of Pilot DAF Unit                               81
26       Photograph of Pilot DAF Unit                               82
27       Photograph of Pilot DAF Unit                               82
28       Dissolved Air Flotation - Pilot Schematic                     83
29       Pilot  DAF Plant Evaluation - Solids Loading Optimization     92
30      „Pilot  DAF Plant Evaluation - A/S Optimization           -    94
31       Proposed Treatment Schematic                              106
                                     IV

-------
                                LIST OF TABLES
No.                                                                      Page

 1       Characteristics of Shrimp Processing Wastewater                       10
 2       Composition of Shrimp Wastes                                        11
 3       Amino Acid Composition of Spray Dried Shrimp Waste Protein           12
 4       Waste Distribution in Shrimp Processing - Raw Peeling                  13
 5       Waste Distribution in Shrimp Processing - Peeling After Steaming        14
 6       Wastewater Material Balance - Gulf Shrimp                           15
 7       Gulf Shrimp Canning Wastewater                                     16
 8       Screening of Alaskan Shrimp Canning Wastewater                      17
 9       Screening of Salmon Processing Wastewater                            18
10       Screening of Salmon Processing Wastewater                            19
11       Screening of Salmon Canning Wastewaters                            19
12       DAF Treatment of Shrimp Canning Wastewaters                       26
13       DAF Treatment of Shrimp Canning Wastewaters                       26
14       Treatment of Salmon Canning Wastewater by DAF                     27
15       Treatment of Salmon Canning Wastewater by DAF                     28
16       Average Wastewater Characterization - Plant A                      42
17       Pollutant Material Balances - Processing                             43
18       Pollutant Material Balances - Washdown                             45
19       Pollutant Material Balances - Processing Plus Washdown               47
20       Comparison of Composite Discharge  Pollutants                        49
21       Comparison of Raw Peeler Wastewater Pollutants                      50
22       Comparison of Washdown Wastewater Pollutants                       50
23       Comparison of Deveiner Wastewater Pollutants                        51
24       Shrimp Canning Plants - Code Identification                          51
25       Vibrating Screen Evaluation                                         74
26       Centrifugal Screen Evaluation                                       74
27       Rotating Screen Evaluation                                          75
28       Tangential  Screen Evaluation                                       76
29       Pilot Screen Evaluation  - Peeling Wastewater                        78
30       Pilot Plant  Operating Conditions -  Preliminary                        85
31       Pilot Plant  Efficiency - Preliminary                                   86
32       Pilot Plant  Operating Conditions -  Series I                            88
33       Pilot Plant  Efficiency - Series I                                       89
34       Pilot Plant  Evaluation - Series I                                      90
35       Pilot Plant  Efficiency - Series II                                      91
36       DAF Pilot Plant Evaluation - Flotation Sludge                         93
37       Pilot Centrifuge Evaluation                                          96
38       Waste Treatment Economics                                         107

-------
                           AC K NOWLEDGEMENTS
This study was financed by the American Shrimp Canners Association, a non-profit
trade organization composed of 22 member shrimp canning plants, and the Office
of Research and Development, Environmental Protection Agency.  The cooperation
of both groups is sincerely appreciated.

Special  thanks are extended  to Ray, Alan and Kenneth Robinson of Robinson
Canning Company for their interest, patience and extra effort in assisting this
project.

The cooperation of Mr. Paul Selley, Mr.  Anthony Cuccia, Mr. Vic Blereau, Mr.
C. J. Reuther, Mr. Huey Authement,  Mr. Elmer  Hall, Mr. Glynn Williams and
Mr. Larry Authement of the American Shrimp Canners Association is gratefully
acknowledged.

Thanks are extended to Mr. Irving (Dit) Snider, Mr.  Buddy Reine,  Mr.  Dudley
Stadler, Mr.  Paul McQueen and Dr. A.  J.  Englande, Jr. for their expertise and
technical  assistance throughout the study.

The support of the project by the Office of Research and Development,  Environ-
mental Protection Agency, and assistance provided by Mr.  Kenneth Dostal and
Mr. Robert L.  Miller, the Project Officer, is acknowledged with sincere apprecia-
tion.
                                      VI

-------
                                  SECTION I

                               CONCLUSIONS
1.    The wastewater characteristics for shrimp canning and processing wastes have
      been generally established.  Significant variations in wastewater concentra-
      tions and flows were found between plants.  The total discharged pollutants
      in kg /100 kg of raw shrimp processed, however, were similar between
      plants.  The variations in  hydraulics and wastewater  concentrations between
      plants will  make wastewater surveys of each individual plant necessary.

2.    The peeling area generates the  largest volume of wastewater and the greatest
      weight of wastewater pollutants in a shrimp processing plant. The  deveining
      operation also contributes significant amounts of wastewater pollutants.  Some
      operations such as retort cooling, can cooling and conveyer spray water have
      very low concentrations of pollutants but contribute significantly to the total
      wastewater flow.

3.    Water conservation for most of the shrimp processing  and canning operations
      is feasible.  The use of dry conveying equipment; the metering and control
      of water to process units; and the control of washdown water flows  will
      materially affect total  water use and total pollutant discharge.

4.    Several screen types were evaluated with peeler wastewaters.  All  the test
      screens  performed well without  blinding with the correct screen openings.
      No ideal screen was found,  however, because all of the test screens exhibit-
      ed advantages and disadvantages.

5.    Dissolved air flotation  was demonstrated to be an effective treatment opera-
      tion in treating shrimp  canning  wastewaters.  Screening is a necessary pre-
      treatment operation, however,  in order to reduce the total settleable mate-
      rial.  An efficient coagulant system and good pH control  are required for
      good degrees of pollutant  removal.  Removals of 40% to 50% protein, 75%
      BOD, 70% COD,  90% suspended solids and 90% oil  and  grease were demon-
      strated on a pilot scale.

6.    The dewatering of  screenings and dissolved air flotation sludge will be
      necessary for economical disposal.  Centrifugation of dissolved air flotation
      sludge was demonstrated to decrease the volume 4:1 and increase the total
      solids dry weight by 2:1.

-------
                                  SECTION II

                             RECOMMENDATIONS
 1.    Those in plant changes which will affect reduction in water use should be in-
      stituted.  These will consist of substituting dry conveyors, where feasible, for
      flume conveying; installation of meters and adjustable flow control valves to
      eliminate excessive water flow to unit processes; and changes in the  washdown
      procedures to use vacuum units and to otherwise control the volumes  of water
      used for washdown purposes.

2.    The shrimp canning industry should press  the process  equipment supplier to
      develop more water-efficient equipment  and to provide guidance in the most  -
      efficient operation with the minimum quantity of water.  Changes in the
      process equipment should also be made to reduce the soluble fraction of
      pollutants discharged in order to  increase the treatability of the wastewater.

3.    Each water supply should be metered and the flow of water to process units
      should be metered to enable fine control  of flow rates and  to allow record-
      keeping of water  use which should be instituted on a continuing basis.

4.    The possibility of water reuse should be pursued with the other federal
      regulatory agencies.  A demonstration of a suitable method would be helpful
      in effecting acceptance of the reuse of water in peeling  and other non-con-
      taminating plant uses.

5.    There should be a continuing program to  collect water use  and analytical data
      in an attempt to detect variations with types of plant operation and the types
      and age of shrimp being processed.

6.    The successful DAF treatment method demonstrated on a pilot plant basis should
      be operated on a  plant scale with good control  and collection of complete
      performance data to establish the feasiblity and efficiency of these units on a
      plant scale.

7.    Additional studies should be made to determine best  methods for handling sepa-
      rated solids and sludge developed from the DAF process.

8.    Studies should be continued to find a marketable value and a practicable way
      of developing marketable quantities of waste protein from cannery waste treat-
      ment facilities.

-------
                                 SECTION III

                               INTRODUCTION
The American Shrimp Canners Association sponsored'this study for the purpose of
seeking to develop an economical, practicable method of effectively and effi-
ciently treating the waste waters from shrimp canning plants.  This association
consists of twenty-two member firms as listed in Appendix D .   The joint efforts
through the association were aimed at accomplishing that which many small, indi-
vidual canners could not individually do.

The shrimp canning plants generally process  from 9,000 to 18,000 kilograms (10 to
20 tons) of raw shrimp per day on a single shift basis.  The largest plants are capa-
ble of processing up to 55,000 kilograms (60 tons) per  day with a two-shift operation.
These plants receive their raw shrimp from small commercial fishing vessels of two
or three-man size and a few larger vessels headquartered between Key West,  Florida
and Brownsville, Texas.  In the central  Gulf area alone, there are more than „
10,000 registered small  commercial fishing boats.  These represent the livelihood of
more than 30,000 families.  The canning plants themselves employ more than  4,000
workers during the peak operating season. Some of these fishermen and plant workers
live in remote coastal areas where  canneries represent the principal or, in some
cases, the only employment available.  These  shrimp canning plants, many of which
are remotely located, are now discharging wastes into rivers,  bayous,  bays or other
adjacent waterways.  Some are located  in communities where public sewer systems
receive the wastewater.
                                                        j*

The shrimp fishery  in the Gulf of Mexico has been for  years one of the most valua-
ble in the United States.  Raw shrimp production in  Louisiana alone has increased
from approximately 4.5  million kilograms (5,000 tons) at the turn of the century to
over 453 million kilograms (500,000 tons) annually in  some recent years.  Catches
of white shrimp (Periaeus setiferous), brown shrimp (Penaeus aztecus), and pink
shrimp (Penaeus duorarum) are the most common in the Gulf of Mexico.  These
shrimp spawn during the spring and summer.  Eggs are deposited directly into the
waters where they drift with tides and currents.  The eggs hatch into tiny creatures
similar to mites or ticks  which grow to about one quarter of an inch size and begin
to move into the shallow waters of the bays and bayous.  These inside waters serve
as nursery grounds for the young shrimp.  They grow rapidly as the water begins to

-------
warm and migrate to larger bodies of water, eventually reaching the Gulf of
Mexico and/or the  Atlantic.  Because of this continuing cycle, the size of the
individual shrimp in a catch varies constantly,with the larger sizes occurring in
the outside waters.

Shrimp are caught primarily in coastal waters using trawls drawn on the floor of the
water body.  Most of the shrimp are dead when brought to the surface,and the re-
mainder die shortly thereafter.  Continued refrigeration (usually with ice) is necess-
ary to preserve this very perishable commodity.  Of necessity, the raw shrimp pro-
cessor must be located close to the fishing grounds and must be able to process the
catch rapidly when it  is docked.  Much  of the Gulf Coast catch is handled as a
raw  product directly to markets and consumers, some is processed and frozen, and
up to 50 percent is canned.

The  canning  of shrimp was first successfully done in  1867 by George W. Dunbar,
an enterprising New Englander who settled in New Orleans and operated a can-
nery after the Civil War.  From this difficult and  trying beginning, an industry
has developed which consists of approximately 70 shrimp canners in the United
States,  25 of which are  located on the  coast of the  Gulf of Mexico.   The Gulf
Coast Canneries  are primarily in Louisiana and Mississippi on bays, or bayous or
within short  trucking distance of the docks. These canning plants have for many
years been,and most remain,family enterprises.  The canneries compete for the
available supplies of raw shrimp and generally obtain and process the smaller
sizes.  Therefore, the economical operating period is generally during the short
spring and fall seasons when shrimp may be taken  in the regulated coastal waters.
Because of the controlled seasons,  the variables of supply and the market price,
the  competition for the raw shrimp is great and no plant is assured that it will
operate on a continuous schedule.   Nevertheless, each plant which operates must
be able to handle its perishable raw shrimp supply in a short time.  Therefore,
plants have developed along the same, most efficient mechanical operating basis.
Most of the equipment is of the same or similar manufacture,and the wastes created
by the operating units have very similar characteristics.

The operations in a shrimp cannery are basically the same the world over,as shown
in Figure 1.  Raw shrimp are first thoroughly washed and separated from debris
or trash and  unsuitable materials.  The raw shrimp are peeled and deveined with me-
chanical  devices developed especially for the shrimp industry.  Figure 2 shows a
Laitram Model A peeler. Heads and hulls are removed, pieces of shell and legs are
separated and the remaining tail meat is separated from the  waste.  In the deveiner,
shown in Figures 4 and 5, the back of the shrimp is split by  a unique razor edge  de-
vice, the back strip or vein is removed and the cleaned, peeled and deveined meat
is again inspected.  Following this, there is a pre-cooking or blanching process  for
3 to 5 minutes in boiling brine solution,which curls the meat,  extracts moisture and

-------
Figure I  GENERAL PROCESS SCHEMATIC
              SHRIMP CANNING

             1

B=
f
•v
RECEIVING
4
WEIGHING
1
PEELING
4
CLEANING
I
SEPARATING
1
DEVEINING
1
IKI CJDCT^Tl^M
N5P&CTKJN
4
BLANCHING
4
GRADING
4
FINAL
INSPECTION
4
CANNING
4
RETORTING
4.
COOLING
|
PACKING
rion , uconiD, WMicn ^
^
HEADS , SHELLS , WATER fc
I
SHELL MATER lAL.WATER^
1
SHRIMPMEAT, VE*|NS,WATEF
I
DEBRIS, SHRIMPMEAT
1
SHRIMPMEAT, SALTWATER ^
^
SHRIMPMEAT, DEBRIS
^
SALTWATER
w
\ HOT WATER
^
WATER
w
\
•
» PRODUCT FLOW
h. VUACTF PI d\u

-------
Figure 2.  SHRIMP PEELER AND CLEANER
  Figure 3. SHRIMP BATCH BLANCHER

-------
 Figure 4. SHRIMP DEVEINER VEIN REMOVAL DRUM

Figure 5. SHRIMP DEVEINER VEIN EXPOSING RAZORS

-------
solubles and develops the pink or red color of the finished product.  A batch blanch-
er is shown in Figure 3.  After cooling and drying and further inspection and grad-
ing, the shrimp are packed, on a scaled weight basis into the appropriate size can,
then mechanically sealed and retorted for 12 minutes at  121° C (250° F).  After
cooling, the cans are labeled and are ready for shipment to market.

The waste from washing, peeling, and deveining constitutes  the strongest and lar-
gest wastewater volume.  This,combined with  the other process and plant wash-
down water,constitutes the wastewater flow from the cannery.

In seeking a solution for the wastewater disposal problems being faced by these
plants which have a very high interest in an unpolluted water environment, there
are several factors which are recognized as being unique. Some of these are:

      a.    There is a lack of prior industry experience  in treating shrimp
            canning wastes.

      b.    The plants operate during short,  intermittent periods as a result of
            state conservation regulations and the  availability of the raw shrimp.

      c.    The combined plant effluent is five to six times as strong as domestic
            waste.

      d.    There are no known toxic materials in the waste discharge, but these
            wastes do contain shrimp protein which furnishes nourishment to the
            marine life in the receiving waters.

The study of necessity included a review of literature seeking the available infor-
mation on shrimp processing, shrimp waste analyses, shrimp waste treatment and
waste treatment processes applied to similar wastewaters.

It was an objective of the study to characterize the shrimp canning plant waste-
waters,  to review canning process operations with regard to water conservation,  to
perform pilot scale  studies, to develop design and operational criteria for a suitable
treatment method and to review and develop economic data on solid waste recov-
ery and disposal.

This report seeks to set forth the findings of the study and to  present data for eva-
luation and  application to the design, construction and operation of wastewater
treatment systems for the shrimp canning industry, particularly on the Gulf Coast.
                                         8

-------
                                SECTION IV

                            LITERATURE REVIEW
GENERAL

A literature review is presented here of shrimp canning wastewater characteristics
and similar wastewaters plus treatability by screening and dissolved air flotation.
The reviews of other  investigators will only include the part of their work of inter-
est to this study.
WASTE CHARACTERISTICS

The FWPCA (1  ) in  1968 conducted a survey of seafood processing plants in West-
wego,  Louisiana to  determine their impact on an existing domestic sewage treat-
ment plant.  Two of the plants surveyed were shrimp canning plants.  The plants
surveyed by the FWPCA were also surveyed for this study and are plants A and B
which are^defined in Section VI of this report.  The FWPCA report made the follow-
ing general comments about the characteristics of shrimp canning wastes:

    "The seafood waste waters appear to be about  10 times as strong organically as
    normal domestic sewage.  This additional strength is principally related to
    dissolved organic material.  While the suspended solids concentration in the
    seafood wastewater is about twice  that measured in the domestic sewage, it
    is within the range of that expected from a strong domestic sewage.   There
    was  no indication of materials toxic to treatment organisms in the  seafood
    wastewaters.

    The  seafood wastewaters tended to spoil rapidly and become odorous even
    when stored under refrigeration. While this was quite troublesome in the
    laboratory  and  interfered with the  performance of some tests, it was an indi-
    cation of a rather rapid breakdown of the organic material in the wastewaters.
    An aerobic K  factor  of 0.18/day  to the base  10 was obtained  using the
    bottle technique on a composite sample from Plant A. There was no indica-
    tion of toxicity during the test.  There was an  indication of  nitrification

-------
    beginning with the third day of the test."

The wastewater characteristics of the composite discharge from Plants A and B are
shown in Table 1.  No plant production data was available to compute a pollutants/
raw shrimp relationship.  Therefore, the wastewater data is difficult to compare
directly.  Table 1  indicates that the wqstewater from Plant A was much more con-
centrated than from Plant B.  Also, the flow rate was higher from Plant A making
the weight/time value from Plant A approximately twice that of Plant B.

     Table 1. CHARACTERISTICS OF SHRIMP PROCESSING WASTEWATER
     FEDERAL WATER POLLUTION CONTROL ADMINISTRATION DATA (1)
                       WESTWEGO, LOUISIANA 1968
Parameter
TSS
VSS
Set. S.*
BOD
COD
OSG
Tot. -N
NH3 -N
Org. -N
NO3 -N
Flow
Plant
mg/l
-
-
4.50
1878
3874
42
278
45
233
4.6
2,790
A
kgAr.
27.6
24.0
-
217.7
444.0
5.0
32.2
5.4
27.2
0.5
m /day
Plant
mg/l
239
207
2.70
1188
1536
106
160
22
138
3.1
1,
B
kgAr.
17.7
16.3
-
92.1
119.7
8.2
12.2
1.8
10.9
0.2
875 m3/day
   "in
I/I
                                        10

-------
The USDI  (2) describes the chemical composition of shrimp wastes in terms of
protein, chitin and calcium carbonate.  This also compares the composition of
wastes from hand picking and mechanical picking operations.  In this publication
the percentage of protein and chitin (a complex polysaccharide, not readily bio-
degradable) are lower in mechanical picking wastes but the percentage composi-
tion of calcium carbonate is approximately twice that of hand picked waste.  These
percentages cannot be  logically explained.  The results of that study are:
              Table 2.  COMPOSITION OF SHRIMP WASTES (2)
     Source
Composition (%)

Hand Picking
Mechanical Picking
Protein
27.2
22.0
Chitin
57.5
42.3
Calcium Carbonate
15.3
35.7
 However a study by CRESA (3) reported a different composition for mechanically
picked Alaskan shrimp wastes.  They reported a composition of 37% protein, 28%
chitin, and 35% calcium carbonate and other, which indicates a much larger per-
centage of protein in the wastes.

CRESA also reported the amino acid composition of spray dried shrimp waste protein,
The results are shown in Table 3 where  the shrimp protein composition is compared
to the protein composition of  casein.  In this study feeding tests were conducted
on rats with shrimp waste proteins. The protein showed a marked deficiency in
sulfur containing amino acids which could be  corrected by supplementation with
either cystine or methionine.  Otherwise the shrimp waste proteins were found to
be equal to casein in nutritional value  and there were no toxic effects noticed.

CRESA also reported an average COD value of 1.37 mgO2/mg organic matter for
shrimp wastes and an average COD value of 1.24 mgO2/mg organic matter for the
shrimp chitin  alone.  Also, an average BOD-5 value  of 0.653 mgO2/mg of organic
matter for shrimp wastes.  This would make the BOD-5/COD ratio equal to 0.48.
                                       11

-------
         Table  3. AMINO ACID COMPOSITION OF SPRAY DRIED
                      SHRIMP WASTE PROTEIN (3 ).
Ami no Acid
Lysine *
Histidine *
Arginine *
Aspartic Acid
Three-nine *
Serine
Glutomic Acid
Proline
Glycine
Alanine
Cystine a
Valine *
Methionine *
Isoleucine *
Leucine *
Tyros ine
Phenylalanine *
Tryptophan *
Total
Shrimp Protein
%
I
8.34
2.97
8.06
8.63
3.91
4.69
17.8
4.54
7.52
7.14
5.62
2.60
5.17
8.14
3.61
5.05
0.73
104.52
/
Casein Protein
%
6.02
2.31
2.41
4.45
3.81
5.88
21.90
15.71
1.16
1.47
7.91
2.75
3.91
11.07
2.72 ,
5.46
1.00
99.94
        * essential
        a-not determined
CRESA compared the shrimp waste distribution when using raw peelers (Laitram
Model A) and peelers using steam treated shrimp.  The results  indicate that a con-
siderably higher fraction of waste organics are lost to the wastewater when using
the raw peelers.  All of the Gulf Coast canners use the raw peelers at present.
The results are shown in Tables 4 and 5.
                                      12

-------
Table  4 . WASTE DISTRIBUTION IN SHRIMP PROCESSING (3 )
                     A. Raw Peeling

                        Analysis
Sample
Raw Shrimp
Washed she! Is
Cooked meat
Cooking and
cooling water
Peeling and
washing water
Weight
grams
1,815
1,292
439
1,046
6,248
Total
Solids
%
25.7
18.3
19.8
4.26
1.81
Suspended
Solids
%



0.43
0.49
C.O.D.
mg/l
304, 000
173,000
267,000
36,300
23,900
5 Day
B.O.D.
mg/l



18,000
9,800
N
mg/l



5,850
2,180
                       Distribution
Sample
Raw shrimp
Washed shell
Cooked meat
Cooking and
cooling water
Peeling and
washing water
Solids
grams
467
236
87
44.5
113
Total
%
100
50.5
18.6
9.5
24.2
Waste
%

60.0

11.3
28.7
C. O. D.
grams
552
224
117
38
130
Waste
%

57

10
33
                            13

-------
Table 5  .  WASTE DISTRIBUTION IN SHRIMP PROCESSING (3  )
                 B.  Peeling after Steaming

                         Analysis
Sample
Raw Shrimp
Washed shells
Cooked meat
Cooking and
cooling water
Peeling and
washing water
Weight
grams
1,818
1,325
470
2,563
5,865
Total
Solids
%
25.7
20.1
21.4
1.58
1.24
Suspended
Solids
%



0.37
0.63
C.O.D.
mg/l
304, 000
173,000
267,000
19,200
17,500
5 Day
B.O.D.
mg/l



9,600
7,800
N
mg/l



2,500
1,680
                        Distribution
Sample
Raw Shrimp
Washed shells
Cooked meat
Cooking and
cooling water
Peeling and
washing water
Solids
Grams
467
267
101
40.5
72.8
Total
%
100
57.2
21.6
8.7
15.6
Waste
%

70.2

10.7
19.1
C.O.D.
Grams
505
254
136
34
84
Waste
%

70

10
23
                             14

-------
Soderquist (4) reported the wastewater discharged from the unit operations in Gulf
Coast Shrimp Canneries.  Those results are shown below in Table 6:
             Table 6.  WASTEWATER MATERIAL BALANCE (4)
                    GULF COAST SHRIMP CANNERIES
Unit Operation
Peelers (Model A)
Washers0
Separators
Blancher
De-icing
Cooling & Retort
Washdown
% of Average Flow
58.1
8.8
6.9
1.6
4.2
12.1
8.3
Range, %
42.1 - 73.0
8.0 - 9.9
5.1 - 9.2
.006 - 2.5
.005 - 7.4
8.0 -19.5
6.9 - 9.6
QCleaners


As can be seen above, a large percentage of the wastewater from a shrimp canning
plant originates at the peelers.  A considerable volume of water is used for can
cooling and retorting but this water is relatively unpolluted chemically and biolo-
gically.

Soderquist also reported the average wastewater characteristics of four Gulf shrimp
plants.  This data is shown in Table 7.  Data was reported in terms of concentra-
tion and weight of pollutants/weight of raw shrimp processed.  This latter method
is the only reasonable method of comparing the discharges from different canneries.
In  comparing the coefficient of variation for COD the concentration value is 28%
of  the mean and the kg/lckg value is only 18% of the mean.

Soderquist also reported a product material balance.  He reported that the food
product was approximately 20% of the raw product, the by-products such as dried
                                       15

-------
Table 7- GULF SHRIMP CANNING WASTEWATER (4 )
   AVERAGE CHARACTERISTICS OF FOUR PLANTS
Parameter
Flow Rate, cu m/day
(mgd)
Flow Ratio, lAkg
(gal/ton)
Settleable Solids, ml/I
Settleable Solids Ratio, lAkg
Screened Solids, mg/l
Screened Solids Ratio, kgAkg
Suspended Solids, mg/l
Suspended Solids Ratio, kg/kkg
5 Day BOD, mg/l
5 Day BOD Ratio, kgAkg
COD, mg/l
COD Ratio, kgAkg
Grease and Oil, mg/l
Grease and Oil Ratio, kgAkg
Organic Nitrogen -N, mg/l
Organic Nitrogen -N Ratio, kgAkg
Ammonia- N, mg/l
Ammonia- N Ratio, kgAkg
PH
Mean
788
(0.208)
46,900
(11,000)
13.9
520
—
—
802
37.7
1,081
46
2,296
109
258
11.0
196
7.6
12
0.51
6.7
Standard
Deviation
92.7
(0.0215
9800
(2350)
5.3
470
--
--
459
15.2
216
__
653
20
169
9.8
62
7.7
5.4
0.12
—
Coefficient
of Variation
% of Mean
12
12
21
21
38
90
--
—
57
40
20

28
18
66
88
32
102
46
24
—
Range
695
(0. 184 -
33,000
(7900
5.4 -
184

__
483
15.9 -
1,008
/
43
1 , 975
86
148
5.4 -
39
1.9 -
7
0.22 -
6.5 -
905
0.239)
57,000
14,000)
31
• 978
—
--
1,100
50.1
1,432
61
2,658
122
759
36.4
290
13.4
14
0.47
7.0

-------
shells were 65% of the raw product and waste was 15% of the raw product.
TREATABILITY BY SCREENING

The pre-treatment of shrimp canning wastewaters by screening has been demonstra-
ted by Peterson (5) with a 5.49 meter (72 inches) wide 1.0 mm (0.040 inch) opening
tangential screen. The duration of the run was 1.5 hours and the flow was approxi-
mately 0.568 mvsec.  (^00 gpm).  The data follows in Table 8 below:

Table 8.  SCREENING OF ALASKAN SHRIMP CANNING WASTEWATER (5)
Parameter
COD
Total Solids
Total S.S.
Settleable Solidsb
Turbidity0
Before
Screening
mg/l
2734
2680
1160
50
215
After
Screening0
mg/l
2360
1900
720
6
193
%
Removal
13.7
29.1
37.9
88.0
10.0
a Bauer "Hydrasieve" -1.0 mm Opening Tangential Screen
bml/l
CJTU
Peterson suggested that flow over the screen was extremely important because suf-
ficient flow was required to wash the solids across the screen but the flow could not
be excessive so that solids were carried off the end of the screen.  He felt that the
screens  tested were only efficient in removing settleable solids.  It was obvious
from his findings that most of the solids were in the dissolved and finely suspended
states.

Other studies have investigated screening of salmon and tuna wastes.  Although
there are probably many differences in the screenability of shrimp  wastes and other
fish wastes enough similarities  exist to justify a review.

Peterson (5 ) also investigated screening of salmon wastes with the same type screen
                                       17

-------
as used in his shrimp waste screening study.  The salmon study used a 0.46 m (18
inch) wide pilot model tangential screen (Bauer "Hydrasieve") where the shrimp
study used a 5.49 m  (72 inch) wide plant scale tangential screen.  For the salmon
study the flow rate was 246 rrr/day (45 gpm), the duration of the run was 2 hours
and the screen opening was 1.0 mm (0.040 inch).  The results of the salmon test
are as shown in Table  9 below:

       Table  9.  SCREENING  OF SALMON PROCESSING WASTEWATER (5).
Parameter
COD
Total Solids
Total S. S.
Settleable Solids b
Before
Screening
(mg/l)
42,256
24,915
17,400
290
After
Screening0
(mg/D
19,075
11,425
7,580
145
%
Removal
54.9
54.1
56.4
50.0
  1.0 mm Opening Tangential Screen.
 bml/l    .

 Peterson comments that blinding of the screen was a problem and a periodic spraying
 consisting of a few seconds every minute near the top of the screen relieved the
 problem.  He further comments that salmon wastes are extremely variable and heter-
 ogenous in nature and the results above are only indications of what could be expec-
 ted over a  long time period.  It does seem obvious from the results of this study,
 however, that salmon  wastes are much more concentrated than shrimp wastes and the
 percentage reduction over the screen are much greater for all test parameters.

 Dehn and Holz (6) also performed a study on salmon processing wastewater where
 they investigated three types of screens: a tangential screen (Dorr-Oliver "DSM"),
 a vibrating screen (Sweco "Vibro-Energy Separator") and a spinning cylindrical
 screen (Sweco "Wastewater Concentrator").  A summary of their results are shown
 in Table 10 below.
                                      18

-------
    Table 10.  SCREENING OF SALMON PROCESSING WASTEWATER (6 ).
          TANGENTIAL, VIBRATING AND CYLINDRICAL SCREENS
Parameter
fag/0
COD
Total Solids
Total S.S.
Settleable Solids0
% Removal
Tangential
Screen
0.149mm
13
8
15
35
Vibrating
Screen
0.42 mm
30
17
31
14
Cylindrical
Screen
0.05 mm
36
58
34
100
a
 ml/I

Dehn and Holz commented that none of the above screens had all the ideal charac-
teristics of producing good effluent quality, high solids concentration, high screen-
ing capacity and trouble free operation.  The tangential screen was somewhat trouble
free and the vibrating screen concentrated solids the best while the cylindrical
screen obtained the best effluent quality.  The conclusions of these investigators
were somewhat noncommittal because of lack of knowledge in ultimate disposal of
screenings.

Claggett ( 7 ) also investigated screening of salmon wastewaters by different types
of screens.  A summary of his results  is shown in Table 11 below:

     Table  11.  SCREENING OF SALMON CANNING WASTEWATERS (7 ).
                   TANGENTIAL AND ROTARY SCREENS
Sample Location
Tanqential Screen (Dorr-Oliver "DSM" - 40 mesh)
Feed
Undersize
Oversize
Remova 1
Rotary Screen (North Sewage Screen - 34 mesh)
Feed
Undersize
Oversize
Remova 1
Total Solids (g/l)

4.5
2.5
164.0
44.4%
4.2
2.4
105.1
42.9%
                                   19

-------
Claggett commented that both screens performed well on salmon canning wastewater
but a gradual  loss in capacity occurred with time with both screens.  However, he
said this was readily corrected by the use of water sprays.

From the above investigations it appears  that no special type of screen was given an
overwhelming  recommendation.  It appears, however, that one could expect a large
reduction in settleable solids, good reductions of total and suspended solids, and a
small reduction in COD when screening shrimp canning waste waters.  Reductions
expected in COD while screening shrimp wastes would be small because the major
screenable portions of the waste such as chitin has generally a small COD per gram.
TREATABILITY BY DISSOLVED AIR FLOTATION

The work performed by  Claggett and Wong (7) on salmon wastewaters indicated
that DAF could effectively treat fish product wastewaters. Also, the meat packing
industry (7) has many plant scale DAF  treatment units in operation and they are
experiencing good performance. It  was recognized, however, that many differences
existed between salmon, meatpacking  and shrimp canning wastewaters.  Therefore,
development of an effective  coagulant system would be needed for shrimp canning
wastewaters.

There are basically four size groups  of particles in wastewaters (8).  They are: course,
which are the  larger suspended solids;  fine, which are the smaller suspended solids;
colloidal; and molecular, which are the particles in solution or dissolved.   Of this
range of sizes, coagulation can only aggregate the colloidal and larger particles.
The suspended solids are relatively easy to coagulate, usually, but efficient coagu-
lation of colloids is quite difficult.

Colloidal particles are  electrically charged.  Depending upon the wastewater the
charge is either positive or negative.  The colloids are also very small and the ratio
of surface area to their mass  if large, therefore,  their behavior is influenced greatly
by surface phenomena.  The  combination  of large surface area to mass and  its
electrokinetic property makes the colloid very adsorptive and, therefore, capable
of being aggregated or  coagulated by adsorbing  to another colloid of opposite sign.
This continues until the particle is heavy  enough to settle.

Shrimp canning wastewater has a very  large concentration of protein.  Protein is
an organic compound and is considered a  wastewater pollutant because of the oxy-
gen demand created in  hydrolysis and deaminization degradation of protein  to free
fatty acids.  The protein molecule contains approximately 54% carbon, 7% hydrogen,
23% oxygen and 16% nitrogen with  occasional trace amounts of sulfur and  phospho-
 rus (8). The protein molecules are very large and are colloidal in size (9).
                                        20

-------
Proteins are made up of pepHde chains with a structure as follows:
                          H             H
                     ^   I     I         It
                    •^•^N -C — C—N-C- C  ^xx.
                               I    II         I    II
                                   0             0
A peptide is made up of a mi no acid residues with an amide  linkage.  To ever/ third
atom of the peptide chain is attached a side chain.  Its structure depends upon the
particular amino acid residue involved such as CH3 for alanine.  Some of the side
chains contain basic groups such as -  NFig in lysine and some of the side chains
contain acidic groups such as - COOH  in glutamic acid.  Therefore, because of the
acidic and basic side chains, there are  positively and  negatively charged groups
along the piptide chain (9).  The structure is as follows:
H
1
N-



0

CH - C/^x
1
CH2
C00~
H.
1
1 N





- CH -
1
(CH2)<
+NH,
The electrostatic behavior of protein is determined by the relative numbers of its
positive and negative charges which is affected by the acidity of the solution. At
the isoelectric point the positive and negative  are balanced, the migration forces
are zero and the solubility is at a minimum.  In a solution on the acid side of the
isoelectric point, positive charges exceed negative charges and, if on the basic side,
negative charges will exceed  positive charges.

The trivalent aluminum and  iron salts are good  primary coagulants with protein
wastewaters (7).  The stoichiometric reaction of hydrated aluminum sulfate (alum)
with water alkalinity is:

             AI2(S04)3 • I8H20 + 3Ca(OH)2—-2AI(OH)3 + 3CaS04+ 18 H20

Under acidic conditions:

                 2 AI(OH)3  +  I8H20  -~AI203  •  I8H20  + 3H*

The solubility of the aluminum hydroxide floe is as follows:

                                           = KS  - ICT33

Therefore, the solubility of  aluminum decreases with increasing  hydroxide concen-
tration or increasing pH.  The solubility of aluminum at pH  4.0  is 51.3 mg/l and
at pH 9.0 is 10.8 mg/l (14).  The aluminum hydroxide floe  is least soluble at pH
7.0 and becomes only highly soluble at pH values less than  5.0.  Therefore, it

                                      21

-------
follows that the effect of trivalent aluminum ions on coagulation is not brought
about by the ions themselves, but by their hydrolysis products.  For a given alumi-
num ion concentration, the rate and efficiency depends on the  pH of the medium
04).

Polyelctrolytes are very effective coagulant aides that form bridges between colloids
and charged floe.   The polyelectrolytes are actually high molecular weight polymers
which contain sites where adsorption can readily take place.  Polymers are not
affected by pH as significantly as the metal salts. There are basically three kinds
of polymers (10):  a cationic which adsorbs on a negative colloid-or floe particle
and which can be used as a coagulant; an anionic,  which replaces the anionic
groups on colloidal particles and permits hydrogen bonding between  the colloid  and
the polymer; and a nonionic, which adsorbs and flocculates by hydrogen bonding
tween the solid surfaces and the polar groups in the polymer.

The principle of dissolved air flotation is based on Henry's Law which states that the
partial pressure of a gas in equilibrium with a  solution  is proportional to its concen-
tration in the solution. Therefore, if a pressure greater than atmosphere is applied
to  a liquid the dissolved air will be proportional to  the pressure applied.  When the
pressure is released, the dissolved  air will come out of solution in  the form of
extremely tiny bubbles.  The bubble will preferentially form on a finite nucleus or
particle.  The chemical coagulation system has created an enlarged  particle or floe
 in  the water so the bubble either forms or attaches itself to this floe. The bubble
gives the floe added buoyancy and it floats to the surface.

The purpose of a DAF  treatment system is to separate and concentrate suspended and
 colloidal particles in the feed wastewater.  Larger particles of the settleable solids
size should be removed prior to DAF treatment by screens and cyclones  if high
density particles are present. Separation of small suspended and colloidal  solids
depends more on their structure and surface properties than on their size and density.
Therefore, DAF treatment plants cannot be designed theoretically or rationally by
mathematical equations but by the use of laboratory (bench scale) and pilot scale
studies.  Factors of greatest importance in designing DAF plants are  as follows:
        1.   Chemical coagulants.
        2.   Feed solids concentration.
        3.   Quantity  of  pressurized air used.
        4.   Overflow rate.
        5.   Retention Time.
        6.   Recycle/Pressurization Mode.

Feed solids concentration and air quantities are usually grouped in a dimension less
air/solids ratio. For a given plant, overflow  rate and  retention time are propor-
tional to the depth of the flotation tank.  The results for this study will include
both values.

                                       22

-------
The air/solids ratio can usually be related to effluent suspended solids.  Solids in
this dimensionless ratio is the weight of solids per unit of time in the influent. Air
is the weight of air per unit of time released to atmospheric pressure in the flotation
cell. The weight of air that can be released is proportional to the saturation of air
in water, and saturation of air in water is directly proportional to the absolute pres-
sure and inversely proportional to the temperature.  Saturation is also dependent
on a good mix of air and water.  Therefore, the saturation is dependent on the
machinery used  for this purpose.  Eckenfelder (11) proposes the following equation
for air saturation:

                 S=Sa  QfP/1.035HJ
                                                          «3
        Where:   S =air released at atmospheric pressure (cm /liter)

                 Sa =air saturation at atmospheric pressure and given
                      temperature (cm^/liter)

                 f  = fraction of saturation obtained in saturation tank

                 P = absolute pressure  (KgF/crrr)

        Therefore the air/solids (A/S) ratio would be:

                 A/S = CSQ/Ss
                                     n
        Where:   C = air density (g/cnr)

                 Q =feed flow (liter/sec.)

                 Ss = suspended solids  (g/sec.)

The performance of the DAF system depends upon having sufficient air present to
float all of the suspended floe.  An insufficient quantity of air will result in only
partial flotation of the solids and excessive air will give no improvements of sus-
pended solids removal so, therefore, an optimum air/solids ratio for  each applica-
tion exists.

Overflow rate is a function of influent and recycle flow, and the  surface area of
the flotation cell.  Overflow rate  is usually expressed in m /sec/m  . This term
can also be  reduced to an equivalent m/sec. relationship,  so, therefore, it also
expresses rise rate in an  upflow tank.

The recycle/pressurization arrangement can be of three basic types:
                                       23

-------
                 1.   Total Pressurization.
                 2.   Partial Pressurization.
                 3.   Recycle Pressurization.

Selection of one of these arrangements should ideally be based on the application
of the  DAF unit.  It is usually,  however, governed by the proprietary position of
the manufacturer of the system.  All three variations operate under the same physi-
cal laws and relationships as described earlier and differ only in piping arrangements.
The three types are shown schematically in Figure 6.

Total pressurization is straight through with a single pass.  The recycle shown in
Figure 6 is only for protection of the pump suction during periods of low flow and
is not  related to recycle pressurization. The advantages of this system includes low
capital investment and the ability to inject a high air/solids ratio.  The disadvantages
are: possible short circuiting; the possibility of shearing delicate  chemical floes;
and the possibility of emulsifying oils.

With the partial pressurization system only part of the  influent stream is pressurized
and the remainder enters the flotation cell under gravity conditions.   Again the
recycle  is only for pump protection  during low flow conditions. The  advantages are:
relatively low capital costs; and the ability to operate under extremely variable
flow conditions.  The disadvantages are generally the same  as full flow pressuriza-
tion plus a poor ability to inject a high air/solids ratio.

The recycle pressurization system offers an alternative to the previously discussed
systems because it represents a significant deviation.  With  this system the influent
flows to the flotation cell by gravity.  A portion of the effluent is recycled, pres-
surized, and released as  influent to the flotation cell.  The advantage of this
system is the ability to incorporate a flocculation tank separate from the flotation
cell.  This allows an increased  detention time for slow developing delicate floes
to form.  The floe is then discharged by gravity to the flotation cell and not sheared
by a pressurization pump as in total pressurization. Therefore, if delicate floe is a
characteristic of the coagulated wastewater, recycle  pressurization offers a tremen-
dous advantage.  The major disadvantage to this system is high capital and operating
costs.

Peterson  (5 ) treated Alaskan shrimp canning wastewaters with a 50 gpm pilot dis-
solved air flotation (DAF) treatment unit.  The unit was a Model 50 Pacific  DAF
system.  Peterson operated the plant with the average conditions shown  in Table 12
below:
                                       24

-------
      Figure  6   AIR FLOTATION ARRANGEMENTS
              RECYCLE / PRESSURIZATION
Influent


   Surge
   Tank
                               Effluent
        Retention
           Tank       Flotation Cell

TOTAL PRESSURIZATION
  Surge
  Tank
                   Pump
                  Retention
                    Tank
                                                 Effluent
                              Flotation
                                Cell
                 PARTIAL  PRESSURIZATION
      Surge Tank
Influent
                        Control   Air
                        Valve
             Retention
                Tank
                             Pump
                            Effluent
                            Flotation
                              Cell
                 RECYCLE  PRESSURIZATION
                             25

-------
    Table 12.  DAF TREATMENT OF SHRIMP CANNING WASTEWATERS (5 )
                          OPERATING CONDITIONS
Pre -treatment
Screening with 1.0 mm Opening Tangential Screen
Flow
Pressurization
Recycle
Run Times
pH
Coagulant^
Polymer
Alkalinity
   109 to 328 m3/day (20 to 60 gpm)
   3.16kgf/m3 (45 psig)
   1:1 at 272 m3/aay influent
   1.0 to 4.0 hours
   3.0 to 7.0
- Alum
-  High  molecular weight anionic
-  Excess with Na OH
Peterson obtained the average results shown in Table 13 below:

     Table 13. DAF TREATMENT OF SHRIMP CANNING WASTEWATERS (5 )
                               AVERAGE RESULTS
Parameter
COD (mg/l)
Total S. S. (mg/l)
Settleable Solids (ml/1)
Protein %
Turbidity (JTU)
After
Screening
4227
1090
22.3
0.201 .
500
After
DAF
1123
252
2.5
0.114
100
Reduction
73.4
76.9
88.8^
43.3
80.0
 The conditions occurring within the shrimp plant while the pilot DAF work was in
 progress are unknown for the above study. There were, however, six peelers with
 four being the pre-cook type (Laitram Model PCA) and two being the raw type
 (Laitram Model A). The plant was processing small pink shrimp and using a high
 salinity process water.

 Peterson comments on the relative insensitivity of pH.  He observed little difference
 in treatment efficiencies from pH 5.0 to pH 6.0. He also observed a low sludge
 solids concentration and settleable solids carryover in the DAF effluent. He attri-
 buted this problem to the high salinity of the process water and secondary reactions
                                     26

-------
with the salt and alum.

Claggett and Wong  (12) studied the treatment of salmon canning wastewater by DAF.
Two seasons of operation using two different pilot systems were reported.  In Part I
of their study a Pacific DAF system which utilize  total flow pressurization was
studied.  Coagulants were injected with a metering pump just prior to entering the
flotation cell.  When pH was adjusted, sulfuric acid was metered in at the same
point and was controlled manually using a portable pH meter. The operating condi-
tions for all runs were:  flow rate, 0.0032 nrr/sec (50 gpm) operating pressure,
289.7 k N/rrr g (42 psig); air injection, 2% by volume.  The results are shown below
in Table 14.

Table 14.   TREATMENT OF SALMON CANNING WASTEWATER BY DAF (8 ).
                          PACIFIC DAF SYSTEM
Coagu lant
Alum
Alum
Alum
Alum
Alum
Alum
X-37-1 a
FeCI3
FeCI3
FeCI3
FeCI3
F-flokb
F-flok
F-flok
F-flok
47 mg/l
47mg/l
47 mg/l
47 mg/l
80 mg/l
60 mg/l
5 mg/l
133 mg/l
60 mg/l
50 mg/l
60 mg/l
1 000 ppm
1000 mg/l
1000 mg/l
500 mg/l
pH
6.1
6.0
6.0
6.0
5.1

5.8
4.1
5.5
5.5
5.5
4.0
4.0
4.0
4.0
DAF
Influent
Total Solids
(mg/l)
5,400
2,290
2,800
2,600
4,180

3,390
1,800
2,860
6,390
5,580
5,900
4,200
1,300
3,500
DAF
Effluent
Total Solids
(mg/l)
1,560
1,200
2,400
2,200
5,000

3,000
1,180
950
1,200
2,400
1,200
1,600
1,400
1,800
%
Removal
71.1
47.6
15.3
15.4
-19.6

11.5
34.4
66.8
81.2
57.0
79.7
61.9
-7.7
43.6
a anionic polyelectrolyte
  a lignosulphonic acid derivitive

As can be seen from the above the  results are scattered but promising, especially
with ferric chloride and F-flok as coagulants.

The investigators commented that ferric chloride when used as the coagulant reduced
considerably the value of any by-products produced from captured solids.  Ferric
                                      27

-------
chloride was also observed to be quite corrosive and would require special materials.
F-flok had the disadvantage of requiring very low operating pH's.  This would also
necessitate special construction materials.

The investigators observed that although the DAF system improved clarity of the
wastewater considerably, the runs were plagued with considerable carryover of floe
and a resultant high settleable solids concentration in the effluent.  No  final con-
clusions were drawn by the investigators from their results.  They felt more testing
under different conditions was required.

In Part II of their study Claggett and Wong used the Favair DAF system which had
a rectangular cell and compressed air injection.  The unit also had the versatility
of allowing effluent recycling and partial  pressurization of the influent stream.

Basically four coagulant combinations were tested: alum and sodium hydroxide;
alum, sodium hydroxide and zetal A (activated animal glue); alum and lime; and
F-flok. Results are shown in Table  15 below.

Table 15.  TREATMENT OF SALMON CANNING WASTEWATERS BY DAF (7 ).
                             FAVAIR DAF SYSTEM
Stream
Influent
Effluent
% Removal
Influent
Effluent
% Removal
Influent
Effluent
% Removal
Suspended Solids Protein BOD Total Solids
mg/l mg/l mg/l (mg/l)
Coagulants: Alum (375 mg/l) and Na OH (75 mg/l)
640
180
70
1,440
485
65
Coagulants: Alum, NaOH
697
200
66
1,020
505
50
1,775
475
73
2,685
1,505
44
and Zetal A (1.0 mg/l)
1,275
381
70
Coagulants: Alum and Lime
1,993
397
73
1,982
830
57
2,833
633
79
2,441
1,625
34

4,268
2,162
46
 The investigators computed the standard deviation for the runs with a coagulant
                                        28

-------
system series.  The variations were from approximately 20% to 75% of the mean
values. The three coagulant systems shown above all gave very similar results.  The
investigators commented that any of the above three systems could be used in proto-
type operation.

When using  F-flok as the coagulant, the investigators found pH adjustment below
the protein isoelectric point (pH 3.8 to 4.2)  was required.  Dosages were heavy
and had to be  proportional to total solids in the  influent feed or solids carryover
resulted.  A distinct plus for F-flok,however, was the large percentage of protein
captured (approximately 70%).
BY-PRODUCT RECOVERY

CRESA (3) investigated the recovery of high grade protein from shrimp shells and
meat by caustic extraction.  Their laboratory studies showed 85% to 90% recovery
of protein by this technique as compared to approximately 55% recovery of protein
by isoelectric precipitation alone.  They proposed a protein recovery system consis-
ting of a caustic extractor, then isoelectric precipitation of protein,  then a screw
press for concentration of remaining solids. They reported that expected yearly oper-
ating  and  depreciated capital costs for this system for the Kodiak Alaska area would
be approximately 1.25 million dollars.  Expected average income would be  approxi-
mately 1.5 million dollars/year or 0.25 million dollars/year profit before taxes.
The economic analysis assumed shrimp protein would sell at $0.33 per kg.  At the
present a  much larger price could be obtained.  This system appears economically
feasible for the  Kodiak Alaska area.

An alternative to processing wastes to high grade protein is to make a crude waste
meal.  This can be done  by individual canners or by a rendering plant (13).  How-
ever, there are  many problems in using  this approach,  including economic and
technical. Crude waste  from shellfish cannot provide the major source of protein
in livestock feeds  because it  would contribute an excessive proportion of calcium
(14).  Digestibility in animals depends on non-protein components of the diet,  in-
cluding calcium (14).  Novak (14) reports a new method for decalcifying meal  by
using  a screening technique.  Therefore, for some application refining of the crude
meal to a  usable meal may be feasible.

Shrimp wastes in the past have been sold as mink food  (13),  as a food supplement
supplying  red pigment for flamingos and trout  (15) and as a flavoring  for  human
food in fish flours  (16).  These methods  of utilization are, however, declining
because of development of synthetic substitutes (17).

Mendenhall (13) reports that  fertilizer and livestock feed is apparently the only


                                      29

-------
established market for crude shrimp waste meals. Also, Mendenhall reports this
market is not large and  crude meal for this purpose seems to offer little, if any,
profit.  Also, the protein available from crude shrimp waste meal is very small in
quantity compared to fin fish protein.  Therefore, the availability of fin fish protein
will control the market  price and would make the economic feasibility of crude
shrimp meal very uncertain.

A refined derivitive of chitin,  chitosan, has many  commercial uses (13).  However,
the commercial  feasibility of refining chitosan from shrimp shells would be poor,
because of, (a)  the small  supply available at any location and, also,  (b) the seasonal
nature of the shrimp processing industry.

It appears that,  although  isolated  uses and markets for crude shrimp waste meal can
be found, the overall economics appear discouraging for a new installation. How-
ever, the economic  feasibility  of refined protein appear good in certain areas, such
as Kodiak, Alaska.
                                      30

-------
                                SECTION V

                     STUDY OBJECTIVE AND APPROACH
The objective of this study was to demonstrate on a pilot scale a wastewater treat-
ment plant for ifie Gulf Coast shrimp processing  industry.

This general objective was considered necessary  to help solve the wastewater pro-
blem of the Gulf Coast shrimp processors.  No canning plant on the Gulf Coast
had any experience  in the treatment of wastewaters, with the  exception of screen-
ing.

Because of the basic similarities  in the processing and operation of the various
shrimp canning plants/  it was decided to make a thorough study of one plant which
would be considered typical. The typical data and results from this operation could
then  serve as a guide for applying the information gained to each of the similar
processors.

Although some data on the characteristics  of the wastewater of shrimp canning
plants was found to  be available, as mentioned in the preceding section,  the volume
of data was quite limited. A further objective of the study was to accumulate and
add to the available data on this type of wastewater.
                                           \
The following specific tasks were established as objectives in  this study:

      1.    Characterize in a physical, chemical, and biological sense the waste-
           water from a typical shrimp canning  plant.

      2.    Measure the  wastewater stream flows from a typical shrimp canning
           plant.

      3.    Evaluate and recommend technical changes in the canning process
           operations  to reduce wastewater flows.

     4.    Perform pilot scale studies on  typical shrimp canning wastes in order  to
           develop specific design and operational criteria for selected  treatment


                                      31

-------
            methods.

      5.     Determine the economics of various alternative procedures to capture
            and dispose of the solid wastes.

After selecting a typical plant (Robinson Canning Company, Inc., Westwego,
Louisiana, shown in the report as Plant A), the study was conducted  at this plant
site during  the summer and winter seasons of 1972 and during the summer of 1973.
A portable  laboratory was installed at the site, a portable pilot scale treatment
facility was installed to treat a portion  of the waste discharged from the canning
plant, and  study personnel monitored  the plant operations.

The initial  season was used to measure wastewater flows and to sample and analyze
the waste streams.  Volumes of water  wasted from each process were measured by
metering the source of water to the process and/or by measuring waste flow streams
in channels.  A continuous record of water consumption was kept during the period
that this study was conducted.

The  characteristics of the waste  streams were established by multiple analyses
during the  period of the study.   Also, waste flows from various other plants in the
New Orleans area, in the lower Terrebonne Parish  area of Louisiana and in the
Biloxi area of Mississippi  were sampled  to compare  to the "typical"  plant waste
discharge.   Results are compared and  discussed in succeeding sections of this
report.

It was not  possible to obtain the pilot plant until October,  1972, missing the first,
summer season.  The shortage of the shrimp supply to the typical plant during the
fall  of 1972 hindered the  accomplishment of many pilot runs during that period.
The preliminary jar testing, chemical selection, shakedown operations and general
preparation was done during this time,  however. During the summer 1973  season
the pilot plant was operated as much as possible, varying operating conditions to
permit pilot scale demonstration of chemical dosages, pH control, air ratio, recir-
culatipn ratio and other plant variables.  Performance data was collected and ana-
lyzed, and is presented in later sections of this report.
 The alternative procedures now used for capturing and disposing of solid wastes
 were reviewed at the typical plant and at other shrimp canning plants.  Various
 types of screens were tested and evaluated for efficiency in the removal of settle-
 able solids.  Solid waste handling techniques in use and other available methods
 were reviewed.
 The techniques and approach used in each objective or task of this study are more
 fully discussed in the subsequent sections of this report.  All data developed are
 available from EPA in a separate Appendix.
                                      32

-------
                                SECTION VI

                           WASTEWATER SURVEYS
GENERAL

Shrimp canning wastewaters were characterized physically, chemically and bio-
logically  at six different plants in the Gulf Coast area.  Most of the work was
performed at Plant A (the  location of the pilot plant studies).  Wastewater sur-
veys were performed a total of 18 days at this plant during the summer and fall
1972 canning seasons and  the summer 1973  canning  season.   This plant was lo-
cated in the New Orleans,  Louisiana metropolitan area. An additional five days
of surveys were performed at three other plants (Plants B, C and D) in the New
Orleans area during the summer 1973 canning season.   A one day survey was per-
formed at a plant (Plant E) in the Houma,  Louisiana area in the summer of 1973,
and^a three day survey was performed at a plant (Plant F) in the Biloxi, Mississ-
ippi area.

The objective of this work was to establish  a typical wastewater characterization
profile for the Gulf Coast shrimp canning industry.  The characterization of each
of the 22 shrimp canning plants was beyond the resources of this study; however,
due to the similarity of process equipment used in the industry it was assumed that
a well  operated plant would produce a "typical" wastewater.  A plant was cho-
sen that would offer good facilities for pilot testing and be well established in or-
der to ensure good production during the study periods.  This was Plant A and most
of the characterization effort was expended at this plant.  The possibility existed
that this plant was not typical  so check surveys were performed at three other
plants in the  New Orleans area  and one plant in each of the other principal shrimp
canning areas; the Houma, Louisiana  area (Bayou Grand Caillou) and the Biloxi,
Mississippi area (Biloxi Bay and  Mississippi  Sound).
SURVEY METHODOLOGY

The wastewater survey at Plant A characterized all wastewater discharges and in-

                                      33

-------
ternal unit processes as to flow and physically, chemically and biologically.

Ten internal sample stations (See Figures 7 thru 10) were established at Plant A
where composites were collected on each of the survey days.  The stations were
located so that all internal unit processes in the plant were  characterized. Compo-
site sampling techniques are included in Appendix B.  A list of sample stations and
locations follows:
         Sample Station
            Plant Process
               1
               2
               3
               4
               5
               6
               7
               8
               9
              11
Screened peeler and separator wastewater
Raw Peeler and Separator wastewater
Cleaner Flume Water
Separator Wastewater
Pre-Inspect ion Flume Water
Deveiner Wastewater
Blanch Tank Wastewater
Canning Room Floor Wash
Processing Room Floor Wash
Rece iving Wash Wastewater	
 Three external sample stations were established and composites of wastewater dis-
 charge were collected on each of the survey days.  Two separate discharges existed
 at Plant A.  The main discharge served all the heavily polluted wastewater dis-
 charged from the plant and received the mapr sampling emphasis.  A minor discharge
  (Sample Station 12) served only can cooling  water discharges.   The main discharge
 was further broken down into a process wastewater discharge (Sample Station 10)
 where a composite sample was collected during shrimp processing times only and a
 washdown  wastewater discharge (Sample Station 10A) where a composite sample
 was collected during washdown times only. The breakdown of the main discharge
 into two samples was necessary because of an extreme variation of total process
 time from day to day.  If a composite sample  had been collected of the main
 discharge from start of process to the end of washdown on each day, the washdown
 discharge which was fairly uniform from day to day would have  had a varying im-
 pact on the total discharge.  Therefore, in order to compare daily discharges on a
 pollutant/product basis,  a separation of this main discharge was necessary.

 A "typical" process day was defined as eight  hours of shrimp processing plus  wash-
 down.  On days where this occurred at Plant  A a total composite (Sample Station
 10 T) was collected at the main discharge from start of process to the end of wash-
 down.  During the planning for this study  the eight hour process day was thought to
 be typical and would occur most days. This,  however, turned out to be wish-
ful thinking because widely varying but usually short process  times occurred
(one to three hours),  hence the necessity for separating the process and wash-

                                      34

-------
Figure 7  PLANT A LAYOUT
  WASTEWATER DRAINAGE  SYSTEM
           CANAL
                   CAN COOLING DISCHARGE
      O SAMPLE STATION
                  35

-------
  Figures  WASTEWATER  SAMPLE  STATIONS
                WASTEWATER SURVEY
                       PLANT  A
                RAW SRIMP  PROCESSING
CANNING
 ROOM
               SHRIMP 8 PLUMING WATER
        DEVEINER
        DEVEINER
                                     RAW  PROCESSING
                                         ROOM
        SAMPLE STATIONS
                         SHELLS
fEFFLUENT
                            48 VIBRATING SCREEN
                             36

-------
Figures  WASTEWATER SAMPLE  STATIONS
                 WASTEWATER SURVEY
                         PLANT  A
                 SHRIMP  CANNING  AREA
                                                    FLOOR DRAINS


                                                   RAW PROCESSING
                                                      ROOM
                                   COOLING DISCHARGE
                           RETORT COOLING
                              TANK
                             CANS IN
                             BASKETS
       CAN  SEALING
          AREA
                     CANS  IN
                     BASKETS
CAN  FILING
  AREA
BLANCHING
  TANK
      SHRIMP IN TRAYS
                                 FLOOR DRAIN
                                  ( BELOW)
                                  SHRIMP IN BUCKETS
                                  FOOD PUMP
                                  WATER ADDED
      HAND  GRADING AREA
                               FLUMING
                               WATER TO
                               DRAIN	'
                                              SHRIMP a FLUMING WATER
    CANNING   ROOM
                                         PLUMED .
                                         WATER /
                                             WASTEWATER FLUME
   SAMPLE STATION
                               37

-------
CD
                 Figure 10 WASTEWATER  SAMPLE  STATIONS
                              WASTEWATER  SURVEY
                                     PLANT  A
                                      CANAL

WASH DOWN 	 «-o— — — —
1
1
1 *
1 >
' 
-------
down samples.  A list of discharge sample stations and locations follows:
         Sample Station
               10
               IDA
               10T
               12
               Plant Discharge
Main Discharge - Process Wastewater
Main Discharge - Washdown Wastewater
Main Discharge - Process and Washdown
Can Cooling Water Discharge
The wastewater flows, as in any plant, were difficult to measure.  At Plant A sever-
al methods were necessary in order to determine the wastewater flows at the point
of the sample station.  Flow determinations at the sample stations were necessary
in order to compute pollutant poundages; however, direct wastewater  flow mea-
surements at most of these locations were for the most part impractical.  The
wastewater flows were confined to gravity flumes, gravity pipes, or gravity col-
lection hoppers discharging to floor drains or wastewater pumps.  Any  method of
measuring flows at these locations would have been highly approximate and time
consuming since an hourly measurement at each station was necessary in order to
detect fluctuations and to compute material balances.  The flow measurement
method employed at most flow stations was to measure with a water meter the water
entering  the process.   The water meters were read hourly during processing and
washdown and theflowsat the sample  stations were computed.

The main discharge wastewater was measured by a different method because of
economic constraints.  The process water entered the plant from two sources; from
a municipal treated supply which was metered and from a deep well which was
not metered.  The well water feed line required a 6 inch meter for flow measure-
ment  and expensive repiping work.  This expense was beyond the capability of the
project so an alternate main discharge flow measurement method was chosen which
involved installing pressure gauges and time clocks on the main discharge pumping
stations.   The time clocks and pressure gauges were read hourly and the flow rate
was then determined from  the appropriate pump curves.  This method of flow mea-
surement was approximate, but a check was available for comparing with the sum of
internal water uses contributing to this discharge.  The water used and, therefore,
the discharged wastewater for the can cooling was measured with a water
meter.

Following is a listing of flow measurement stations (See Figure 1 l),the  flow mea-
surement method, the plant process measured and the process units at Plant A:
                                       39

-------
    Figure II   WATER  FLOW   STATIONS
                  WASTEWATER  SURVEY
                            PLANT   A
              TO WEIGHING
              BUILDING
  3"RISER TO  - -I-,
  RETORT COOLING
  TANK
                                                          I RISER
                                                          FLUME WATER
                                                          a CLEANER
I RISER
FLUME WATER
a CLEANER
   RISERS
CONV SPRAY'" 1/2"o.
             0 H.
       PLUMING
       WATER J,
                                                         I RISER
                                                      CLEANER ONLY


1 O.H.
SEPARATORS


                    CD  ,/<•• >TZ]
                       f*a unco rtkti v<
                       GRADER ONLY
                        •3/4" TRAVELING SPRAY
                                         I" RISER
                                   FLUME WATER,GRADER
                                   a SEPARATOR
                                                        FREEZER'ROOM
                                                        AIR COMPRESSER
                                                        ROOM
                                      2 O.H.
                                                                    I 1/3"
                            RISER
                                                              2 RISER
                      ^
                        BOOSTER PUMP

                FLOW STATIONS
                                    40

-------
Flow
Station
A
B
C
D
E
F
G
H
1
J
Pump Station


Process
Flume make-up
Peeler
Separator
Flume make-up
Deveiner
Conveyer Spray
Pump Make-up
Conveyer Spray
Raw Receiving
Can Cooling
Main Wastewater Discharge


Units
Plant A
2
8
4
1
1
1
1
3
1
1
1


Method of
Measurement
Water Meter
Water Meter
Water Meter
Water Meter
Water Meter
Water Meter
Water Meter
Water Meter
Water Meter
Water Meter
Time Clock-Pres.
Gauge & Pump
Curve
RESULTS AT PLANT A

Table 16 shows characterization results at Plant A in terms of pollutants (kg per 100kg
raw shrimp processed).  These computations were  made in order that dilution water
would not be a factor in comparing other plants to Plant A and also in comparing
different processing days at Plant A.  The first four columns give characteristics of
internal plant processes.  The last two columns give the combined plant discharge
broken down  into processing time only and washdown time only. The last two col-
umns should be summed  to obtain total combined plant discharge.  Results indicate
that approximately 5 kg BOD-5 are discharged as waste per 100 kg of raw shrimp
processed.

Pollutant material balances were computed for Plant A on days where sufficient data
existed.  For each pollutant parameter the sum of the weights produced  internally
should equal  the sum of the weights discharged.   For the main plant discharge during
the processing period  internal pollutant weights were computed and compared to
external discharged pollutant weights.  The results are shown in Table 17.  This was
also computed for the washdown period, as shown in Table 18 and processing plus
washdown in  Table  19.

Table 17 shows the  results of  material balances of pollutants at  Plant A during shrimp
processing only.  The composite wastewater samples were started when shrimp peel-
ing started in the plant  and the composite was completed when the unit operation
tested,such as deveining, stopped.  The internal wastes were the summation of wastes
discharged at Station  1  (peeler and separator wastewater), Station  6 (deveiner^
wastewater),  Station 7 (blanching wastewater), Station 11 (unloading and de-icing
                                      41

-------
                            Table 16.  AVERAGE WASTEWATER CHARACTERIZATION
                                                PLANT A
                                       Values in kg/100 kg raw shrimp
Wastewater
Parameter
BOD5
COD
O&G
Tot. N (as N)
Tot. Alk.
Ortho P (as P)
TS
TVS
TSS
VSS
Raw Peeler
& Separator
Station 2
* s
4.89 1.01
10.80 1.74
2.91 1.74
0.80 0.95
2.05 0.46
0.25 0.09
10.64 2.66
7.12 2.33
2.63 1 .32
2.43 1.34
Deveiner
Station 6
x s
0.51 0.03
1.08 0.27
0.15
0.10 0.07
0.11 0.02
0.03 0.01
0.99 0.24
0.75 0.19
0.45 0.22
0.40 0.26
Blanch
Tank
Station 7
x s
0.15 0.04
0.30 0.08
0.27

0.03 0.01
0.01 0.01
3.51 1.00
0.84 0.49
0.19 0.13
0.09 0.11
Receiving
& Raw Washing
Station 1 1
x s
0.66
0.87 0.24
0.19 0.06
0.02
0.22 0.12
0.013 0.003
1.04 0.36
0.59 0.26
0.25 0.19
0.12 0.06
Total Discharge
Washdown Only
Station 10A
* s
0.40 0.11
1 .64 0.47

0.014 0.007
0.82 0.38
0.07 0.05
6.35 2.10
1.95 1.23
0.34 0.29
0.33 0.27
Total Discharge
Processing Only
Station 10
x s
4.61 0.32
8.14 0.05
3.58 1.63
0.19 0.12
2.22 0,31
0.23 0.03
15.22 2.50
6.38 2.87
0.83 0.49
0.61 0.53
N)
    x = Mean
    s = Standard Deviation

-------
                                  Table 17.  POLLUTANT MATERIAL BALANCES
                                      DURING PROCESSING AT PLANT A
Date
Sampled
7/13/72
6/29/72
7/15/72
7/7/72
7/15/72
6/29/72
7/13/72
6/29/72
7/13/72
7/13/72
Parameter
COD
Tot. Alk.
Tot. Alk.
Ortho Pas P
Qrtho Pas P
TSa
TSa
TVSb
TVSb
TSSc
Weight (kg)
Internal
Station
1
1,151
313
182
31
26
1,633
1,629
1,130
1.022
149
Station
6
128
21
14
10
4
164
125
134
97
70
Station
7
52
6
3
2
2
734
805
326
85
9
Station
11
124
40
12
3
2
196
157
124
96
27
Misc.
150
38
21
5
4
273
272
171
130
25
Sum
1,605
418
232
51
38
3,000
2,988
1,885
1,430
280
External
Station
10
1,351
399
239
48
.25
2,884
2,678
1,181
1,111
237
External
Sum of
Internals
0.84
0.95
1.03
0.95
0.66
0.96
0.90
0.63
0.78
0.84
a Total Solids D Total Volatile Solids c Total Suspended Solids
CO

-------
wastewater) and miscellaneous discharges such as fturning water and conveyer spray
water.  The external wastes were the total combined plant discharge wastes at
Station  10.  For a material balance the sum of the internals should be equal to the
external.

On July 13,  1972 sufficient data was collected and analyzed in order to compute a
material balance for COD. The balance had approximately a 16% error.  The error
was due to cumulative errors in sampling, flow measurement and analysis in the
laboratory.  Material balance  computations for other parameters had a range  of
errors from 3% to 37%.

Table 18 shows material balance computations during washdown at Plant A. For this
period composite wastewater samples were started at Sample Station 9 when washdown
started in  the raw process section of Plant A and the composite was completed when
washdown in  this section of the plant was completed.  The same procedure  was used
for Sample Station 8 in the canning section of the plant.  During this same time a
total composite which included washdown from both sections of the plant was col-
 lected at  Sample Station 10A.

The material  balances shown in Table 18 indicate poor correlation for all parameters
with the exception of solids.  These results were expected, however, because of the
extreme difficulty  in obtaining  representative samples during washdown.  For exam-
ple, when the blanch tanks were dumped and cleaned an extremely concentrated
slug of water was released. The floor washings were also quite variable.  A great
deal of meat  fragments accumulated on  the floor during the washdown of the equip-
ment.  When this material was washed from the floor to the drain flumes representa-
tive samples were difficult to obtain.

Table 19 shows material balance computations for a total plant day,  including pro-
cessing and washdown.  The external discharge station 10 T was only composited on
days when the shrimp supply allowed the plant to process approximately 8 hours.
The errors in  material balances  varied from approximately 62% to 33%.  These
results were inferior to  "process only" balances because the washdown period was
included in the composites.  As previously discussed, obtaining representative com-
posite samples of washdown water was most difficult.
COMPARISON OF PLANT A TO OTHER PLANTS
                                                         n
A comparison of Plant A discharged pollutants to five other Gulf Coast shrimp can-
ning plants is shown in Table 20.  The comparison was made on a kg pollutants/
100 kg raw shrimp basis.  For BOD-5, Plant A was the lowest of 4 plants that were
tested.  Plant A was only slightly lower than Plants D and E, but considerably

-------
Table 18.  POLLUTANT MATERIAL BALANCES
    DURING WASHDOWN AT PLANT A
Date
Sampled
7/6/72
7/13/72
7/13/72
7/15/72
6/29/72
7/6/72
7/13/72
7/15/72
7/6/72
7/13/72
7/15/72
Parameter
BOD5
COD
Tor. N as N
Tot. N as N
Tot. Alk.
Tot. Alk.
Tot. Alk.
Tot. Alk.
Ortho Pas P
Ortho Pas P
Ortho Pas P
Weight (kg )
Internal
Station
8
6.3
131.3
1.8
0.5
2.3
28.1
77.5
38.1
0.5
2.7
1.8
Station
9
3.2
46.2
0.9
0.5
4.1
31.3
58.0
35.3
0
0.9
0.5
Sum
9.5
177.5
2.7
1.0
6.4
59.4
135.5
73.4
0.5
3.6
2.3
External
Station
10A
73.4
293.5
1.4
1.8
68.0
88.8
182.0
116.9
6.3
20.8
6.3
External
Sum of
Internals
7.73
1.65
0.52
1.80
10.63
1.49
1.34
1.59
12.60
5.78
2.74

-------
Table 18 (continued).  POLLUTANT MATERIAL BALANCES
         DURING WASHDOWN AT PLANT A
Date'
Sampled
7/ 6/72
7/15/72
7/6A2
7/15/72
7/15/72
Parameter
TSa
TSa
TVSb
TVSb
TSSC
Weight (kg )
Internal
Station
8
105
164
58
50
19
Station
9
529
143
417
17
4
Sum
635
307
475
67
23
External
Station
10A
933
366
324
52
59
External
Sum of
Internals
1.47
1.19
0.68
0.78
2.57
a Total Solids D Total Volitile Solids c Total Suspended Solids

-------
Table 19. POLLUTANT MATERIAL BALANCES
  DURING PROCESSING PLUS WASHDOWN
               PLANT A
Date
Sampled
6/30/72
7/10/72
6/30/72
7/10/72
7/10/72
7/10/72
Parameter
BOD5
COD
Tot. Alk.
TSa
TVSb
TSSC
Weight (kg)
Internal
Station
1
1,755
2,991
611
3,823
2,640
, 503
Station
6
185
439
45
450
351
117
Station
7
42
71
8
1,301
347
64
Station
i
8
59
57
84
124
53
35
Station
9
21
10
134
43
1
35
Station
11
230
227
125
255
98
39
d Total Solids ° Total Volatile Solids *• Total Suspended So
Misc.
229
392
101
600
349
79
Sum
2,517
4,187
1,108
6,596
3,840
872
External
Station
10T
1,420
3,529
827
8,788
2,590
329
External
Sum of
Internals
0.56
0.84
0.75
1.33
0.67
0.38
ids

-------
lower than Plant F.  For COD, Plant A again was the lowest of six plants tested at
8.14 kg COD/100 kg shrimp.  Plant C was in the same range as Plant A at 8.27 kg
COD/100 kg shrimp; however, Plants B, D, E and Fall had considerably  larger
discharges with Plant F at 23.65 kg COD/100 kg shrimp.  Total solids discharged
shows Plant C the lowest and again Plant F the  highest with Plant A approximately
mid-way.  Total suspended solids discharged shows a large amount of scatter in the
testing with Plant A again having the  lowest discharge and Plant F the highest.

Plants A, B,  C and D all screen  their peeler wastewaters, but Plants E and F do not.
This partially accounts for the larger discharges at Plants E and F.  Deviations could
have also occurred  because different labs analyzed the wastes from each of the
Plants E,  F and remaining 4 plants. Comparing only Plants A,  B C and  D and con-
sidering a 20 to 30% scatter in the data, it appears Plant  A has a slightly lower but
similar discharge to the other three plants.

Table 21 shows peeler wastewater pollutants compared at five plants. The values
shown are raw peeler wastes prior to screening.  The peeler wastewaters at Plant A
appear very similar to the other plants with the exception of Plant E. Again the
analysis of the wastewaters at  Plant E was performed at a different lab than the other
four plants.  Total suspended solids at Plant A, however,  was much lower than at the
other plants.  This  is a significant difference and it is not known whether the cause
was a sampling and testing error or a processing difference at Plant A.  Total solids
showed  less scatter and were within the expected deviation.

Table 22 compares washdown at Plants A and C.  Considering the difficulties in
obtaining representative washdown composites,the  values are extremely  close and
the two tested plants appear very similar.

Table 23 compares deveiner wastewaters at Plants A and D.   Plant A has a slightly
higher discharge than Plant D, but within the expected deviation.

In summary, the comparison  data showed considerable scatter, but did indicate that
Plant  A had a similar discharge to the other tested plants.  The code identification
for  the tested plants is shown in Table  24.
                                       48

-------
     Table 20. COMPARISON OF COMPOSITE DISCHARGE
                 WASTEWATER POLLUTANTS
                     Processing Time Only
Plant
A°
Bb

Cc
Dd
Ee
Ff
Quantity
kq/lOOka Raw Shrimp
^BOD-5
4.61



5.27
5.36
13.90
COD
8. 14
16.16

8.27
12.60
19.28
23.65
Total
Solids
15.22
12.25
\
8.50
11.95
29. 14
32.35
Total '
Susp. Solids
0.83
3.67

2.19
4.39
1.64
9.20
a  average of 5 values - screened peeler wastes
b  average of 2 values - screened peeler wastes
c  one value          - screened peeler wastes
d  average of 2 values - screened peeler wastes
e  one value          - raw peeler wastes
f  average of 3 values - raw peeler wastes
                              49

-------
               Table 21.  COMPARISON OF RAW PEELER
                     WASTEWATER POLLUTANTS
Plant
Aa
Bb
Cc
Dd
Ee
Quantity
kg/lOOka Raw Shrfmo
BOD-5
4.22


4.71
3.26
COD
7.38
9.21
6.67
8.75
16.15
Total
- Solids
8.85
7.49
10.13
10.90
24.01
Total
Suspended Solids
1.65
3.20
5.51
4.63
3.05
a - average of 6 values, b - one value, c - one value,  - one value
^*  ^KBA *^ • Mfm I • • «^
  - one
value
               Table 22. COMPARISON OF WASHDOWN
                     WASTEWATER POLLUTANTS

Parameter
COD
Orfho Phosphate
Total Solids
Total Suspended Solids
Quantity
ka/1 00 ka Raw Shrimo
Plant
A
1.64
0.07
6.35
0.34
Plant
C
1.15
0.03
6.18
0.27
                              50

-------
Table 23.  COMPARISON OF DEVEINER
      WASTEWATER POLLUTANTS
Parameter
BOD-5
COD
Ortho Phosphate
Total Solids
Total Suspended Solids
Quantity
ka/100 ka Raw Shrimo
Plant
A
0.51
1.08
0.03
0.99
0.45
Plant
D
0.24
0.80
0.02
1.31
0.32
 Table 24. SHRIMP CANNING PLANTS
             Code Identification
Plant
Robinson Canning Company
Cutcher Canning Company
Reuther Seafood Company
Southern Shellfish Company
Calvin Authement Canning
Plant
DeJean Packing Company
Location
Westwego, La.
Westwego, La.
New Orleans, La.
Harvey, La.
Dulac, La.
Biloxi, Miss.
Code
A
B
c
D
E
F
               51

-------
                                SECTION VII

                            BENCH SCALE STUDIES
GENERAL

Bench scale studies with a flotation bomb and jar tests were performed in order to
determine treatability by the DAF process.  Gravity drainage of DAF skimmings
was also performed in order to  evaluate the effect of chemical dosages on the nature
of the skimmings and to evaluate further concentration by gravity thickening.

Buchner funnel tests on the DAF plant sludge were performed in order to evaluate
the feasibility of vacuum filtration.
JAR TESTS

Jar tests were performed only as a method for screening coagulants and coagulant
aides in order to have an effective chemical system prior to the start of DAF pilot
plant testing. Jar tests are nothing more than a method of optimizing a coagulant
system.  The methods used for this study are explained in Appendix A.  The results
of jar tests, however, divulged considerable basic information about the wastewater,
such as the relative size of the  particles in the wastewater, as discussed in Section
IV.

The treatability of shrimp canning wastewater by the DAF process was first evaluat-
ed in the fall of 1972 by conventional jar tests.  Selected combinations of coagu-
lants and polyelectrolytes were  tested in order to qualitatively determine a work-
able chemical combination.  The composite main discharge  at Plant A was tested
with a variety of coagulants and coagulant aides.   The following chemicals were
tried by using all reasonable combinations and a wide range of dosages:

            pH Control                  Sulfuric Acid

                                       Caustic Soda

           Water  Conditioners          Sodium Carbonate
                                       Zinc Sulfate
                                      52

-------
            Coagulants                 Aluminum Sulfate
                                       Ferric Sulfate
                                       Clay
                                       Magnifloc 503C

            Coagulant Aides            Betz  1130
                                       Magnifloc 835A
                                       Magnifloc 837
                                       Magnifloc 905N
                                       Magnifloc E323

The best floe was obtained with an acid, alum  (aluminum sulfate), and magnifloc
835A combination.  Alum and acid alone formed a light, fluffy, slow settling floe.
When the  Magnifloc 835 A was added the floe  formed much faster, was much larger,
tighter and settled better.  None of the other coagulant aides in the list above
performed as well as Magnifloc 835 A.  No success was obtained  by substituting
a polymer or clay for alum.  Good settling was obtained with alum dosages between
75 and 150mg/l.  Zinc sulfate and sodium carbonate conditioning seemed to have
no effect on the settleability.  The raw water at Plant A did contain a fairly high
alkalinity, however, so water conditioning  may be necessary at other plant locations.
Using NaOH for pH adjustment when using alum and polymer was definitely inferior
to acid  pH adjustment.  Good results were obtained with pH values between 4.8 and
5.2.

Figure 12  shows the results of H2  $04 titration on  the raw shrimp  canning wastewater.
A regular  jar test was performed by adjusting a sample to a desired pH, rapid mixing,
allowing the sample to settle and then  measuring the turbidity of the supernatant.
The titration curve shows protein  precipitation  occurring only in a narrow pH range
with maximum efficiency at about pH 4.4.   From this data  the isoelectric point was
assumed to be pH 4.4.

The best chemical coagulant combination found in the jar testing  was extremely
efficient at optimum pH values stated above.  However,  since the isoelectric point
of protein was pH 4.   protein precipitation and protein removal at pH 5.0 was
below maximum efficiency.

Many trial runs were made with various chemical combinations to try to eliminate
alum as  the primary coagulant. It was believed at this stage of the project that
elimination of alum would greatly simplify sludge treatment and would increase the
value of any by-product.  Little success was achieved for these efforts.  It appeared
that alum would be  required as the primary coagulant.

As described  in Section VIII, Pilot Plant Studies, the chemical combinations dev
                                      53

-------
en
                                   Figure   12  ACID PRECIPITATION
                                 OF SHRIMP CANNING WASTEWATER
                                                    Least Soluble
                                                    8 Isoelectric  Point
                                                    = pH 4.4
                                              PH

-------
eloped in the above described tests were used in the first pilot plant phase.  An
excellent correlation was obtained between the pilot plant results and the jar
tests.
AIR FLOTATION BOMB

Prior to the start of the second pilot plant phase during the summer of 1973 a more
thorough evaluation of chemical dosages was believed to be warranted.  With the
chemicals and dosages used in the first phase studies, an effluent barrier of about
1000 mg/l  of COD developed that appeared impossible to break  through.  It was
believed a completely different combination of chemicals could  be found that
could possibly give much better results.  Information about the skimmings was
also desirable. This could not be obtained  from jar tests so a bench scale air flo-
tation bomb was used for further testing.  Additional jar tests were also performed
for correlation.

Figure 13 shows thebaftch  scale air flotation bomb used.  The operational  proce-
dure used in these tests is explained in Appendix A.  The air flotation bomb
utilizes the dissolved air flotation (DAF) method for producing a  bubble for floe
flotation.

With the pressure bomb/ combinations of coagulants and polyelectrolytes were
tested to determine optimum pH and dosages.  System performance was primarily
based on effluent quality as measured by total  organic carbon (TOC).  The effects
of various dilution rates, direct versus pressurized recycle, variation in operating
pressure, and alum efficiency at increased alkalinity were also investigated.  In-
herent in this study was,of course/ a comparison of bench scale versus pilot plant
performance.

The first pilot plant phase had determined that pH was an extremely important
variable in DAF treatment of shrimp canning wastewater.  The wastewater had a
high concentration of protein  with an isoelecrric point of 4.4.  The pilot plant,
however,  performed rather poorly at  pH 4.4.  This was due to the inactivity of
the alum coagulant.  An optimization of pH was required with each coagulant
used.

The results of the jar testing and air flotation bomb testing for this period are
included in a supplementary appendix available from EPA.  A summary of the
results and the investigator's conclusions are as follows:

      1.    The optimal pH as obtained by  jar tests and  prototype performance was pH
           5.0, closelycorresponding to  the isoelectric point of the protein present,
                                      55

-------
Figure 13 BENCH SCALE FLOTATION BOMB
         FROM ECKENFELDER
               PRESSURE
                GAUGE
OMPRESSED VALV
AIR V 	 ^r
SOURCE CHECK n
Ml IT r**
AIR ^^
PRESSURE
CHAMBER SUPPORT
ASSEMBLY i
m\ •***%
!(£

{

rn
V-

c

!r~^
^AIR BLEED
^ VALVE
fe-

n
svv
/ \
-'"—
-—
c


i r\ f\r\ i
1000 ml.
GRADUATE
CYLINDER
=*

       PRESSURE CHAMBER
                 56

-------
       which was pH 4.4.  At pH 5.0 a weak floe formation commenced with-
       out coagulant addition.

 2.    The most effective combination of chemicals for successful flotation
       operation consisted of alum and Magnifloc 835 A polymer at respective
       dosages of approximately  100 mg/l and 2 mg/l.  While jar tests results
       indicated that alum dosages could be substantially reduced and possibly
       eliminated,  flotation experience required alum to provide necessary
       floe size and strength.

 3.    Increasing the alkalinity for the alum runs did not appreciably affect
       system performance.

 4.    For the alum and Magnifloc 835 A runs, the jar tests gave comparable
       results (on a TOC removal basis) to the flotation bomb.  The jar test,
       therefore, was used as  a screening device for other coagulant and
       polymer combinations.   Little information concerning floe strengths,
       however, was obtainable  and good results with the jar tests did not
       imply equally good results with the flotation bomb.

 5.    It was observed that system performance could be correlated  to color
       content - the clearer the  effluent, the better the quality in terms of
       TOC.

 6.    Results indicate that precipitation by coagulation,  flocculation, and
       sedimentation could provide a feasible alternative-to dissolved air flo-
       tation (perhaps without alum addition).

 7.    Varying dilution rates for direct pressurization did not affect system
       performance appreciably.

 8.    A very clear effluent and dense float solids concentration was observed
       when pressurized recycle  was employed with the bench-scale flotation
       device.

 9.    Variation in operational pressure of the dissolved air flotation bomb
       from 30 to'60 psi  only  improved TOC removals from 27% to 30%.

10.    Bench-scale testing can satisfactorily simulate pilot plant performance
       based  on effluent quality  as measured by total organic carbon.  Labora-
       tory TOC removals from jar tests and the flotation bomb (when using
       alum and Magnifloc 835 A polymer) averaged 34% and 29%  respectively.
       A study at the plant site on a fresh sample, however, yielded a reduc-
                                 57

-------
            tion in TOC of 66% which was very near the average pilot plant perfor-
            mance of 61%.  Thus it appears that for the shrimp wastewater, a
            fraction of the colloidal TOC solubilized or protein ammonified during
            shipment to the laboratory.  It is therefore necessary that future bench-
            scale testing be conducted on site for absolute numerical comparisons of
            pilot plant or prototype results.   In the completed study, however,
            successful screening of coagulants and flocculants was accomplished
            by the methodology employed.

     11.    Chemical precipitation or dissolved air flotation offers effective remov-
            al of organic constituents from shrimp cannery wastewaters with the
            possibility of protein recovery.
SLUDGE DRAINAGE

During most of the pilot plant DAF runs,  a sludge drainability test in a 1000 ml
graduated cylinder was performed on the  sludge skimmings.  Information about the
drainage rate as a function of the DAF plant operating variables was desired.  It
was hypothesized the drainage rate would be a function of the air/solids ratio and
the coagulant chemicals used with the DAF plant.   The bench tests would  indicate
if gravity drainage would be  feasible  as a sludge concentration operation on a
prototype scale.

Figures 14 through 21 show the results of  drainage tests during the  Fall 1972 pilot
plant phase  and  the Summer 1973 pilot plant phase.  The drainage rates in ml/
minute was computed for the  straight line segment of the curve in order to quanti-
tatively compare the runs.  The shape of  all the curves are generally the same ex-
cept the ones shown on Figures 14 and 15.  These tests were the first done and,
as can be seen,  were prematurely ended.  The other curves show a fast drainage
linear drainage rate for about one hour; then a drastically decreased drainage rate
for about two hours; then another fast linear rate for about two hours; then an
abrupt leveling off (Figure  18).

It was hypothesized prior to the drainage studies that alum concentration would
have a very large effect on the drainage  rate.   Alum floe is highly hydrated and
it was believed that high concentrations of alum would hold water in the sludge
and retard drainage.   The drainage studies do  indicate a relationship between
drainage rate and alum feed concentration.  For example the drainage studies
shown in Figures 18 and 19 had the same  operating conditions with only a  slight
difference in alum concentration and  the curves are very similar with the drainage
rates being almost identical.  However, comparing  Figures 16 and 19 where alum
concentrations are in a 2:1  ratio the curves indicate a much faster initial drainage
                                      58

-------
o
o
o
o
o

I-
o
CD

5
O
LL)
O
                 Figure 14   SLUDGE  DRAINAGE

                            DAF  RUN 5P

                             Fall, 1972
CO

Q
UJ
O
or
MJ
I-
•Rote

:8.60ml/min.
          AIR = 1.5% by Volume

          PH = 5.0

          ALUM = 150 mg/l

          POLYMER =5mg/l 835A

          SLUDGE FLOW = 30.52m3/doy

                       (5.6 gpm)
                  100
            150   200  250   300  350  400
                       TIME  (MINUTES)
                           59

-------
o:
o
J
5
o
o
CD
O
tr
UJ
O
UJ
o

if
DC
UJ
400



380



360


340



320



300


280



260


240


220



200


 180



 160


 140



 120


 100



  80


  60



  40



  20


   0
                Figure 15   SLUDGE DRAINAGE

                           DAF RUN I2P

                            Fall, 1972
                     Rate

                    = l.88ml/min.
AIR = 1.5% by Volume

PH = 5.0

ALUM = 75mg/l

POLYMER =5mg/l 835A

SLUDGE  FLOW = 22.35m3/day

             (4.1 gpm)
             50
              100    150  200   250   300   350  400
                       TIME  (MINUTES)
                           60

-------
o
J
5
o
o
o
o

o
CD
O
CL
U.
LJ
O
LU
o
a:
LJ
                Figure  16    SLUDGE  DRAINAGE
                           DAF RUN 5
                         Summer, 1973
                                            Rate

                                            0.74 ml/min.
            = 1.89 ml/min.
AIR =6.0% by Volume

PH= 5.0

ALUM = I00mg/l

POLYMER = 4mg/l 835A

SLUDGE FLOW=22.45m3/day

             (4.!2gpm)
            50    100   150   200  250   300   350  400
                       TIME (MINUTES)
                          61

-------
             Figure  17  SLUDGE DRAINAGE
                       DAF RUN 8
                       Summer, 1973
DC.
O

J
O
O
O
O
CD

5
O
o:
LJ
O
UJ
o
J2
o:
400


380


360


340


320


300


280


260

240


220


200

 180


 160


 140


 120


 100


  80


  60


  40


  20

   0
              'Rote
              = l.96ml/min.
AIR = 6.0% by Volume

PH = 6.2

ALUM = 300mg/l

POLYMER=4mg/l835A

SLUDGE FLOW=6.05m^day
             (l.llgpm)
             50   100   150   200   250  300   350  400
                       TIME  (MINUTES)
                           62

-------
                Figure  18    SLUDGE  DRAINAGE

                           DAF RUN 12

                           Summer, 1973
o:
o
o
o
o
o
CD
o
o:
u.


LJ
o
o
UJ
o

(2
tr
LJ
i-
         AIR = 6.0% by Volume

         PH = 5.0

         ALUM = 50mg/l

2.53mi/min.  POLYMER = 4mg/l 835A

         SLUDGE FLOW = 29.48m3/doy

                     (5.4lgpm)
                                  250   300  350  400
                       TIME  (MINUTES)
                          63

-------
a:
o
J
5
O
o
i-
o
OD
O
tr
UJ
o
cc
UJ
                Figure 19   SLUDGE  DRAINAGE
                           DAF  RUN 13
                           Summer, 1973
                                             Rote
                                            = 0.86 ml/min.
            •Rote
             2.64 ml/min.
    AIR = 6.0% by Volume
    PH = 5.0
    ALUM = 60mg/l
    POLYMER = 4mg/l 835A
    SLUDGE FLOW = 24.80m3/doy
                (4.55 gpm)
             50
100
150   200  250   300  350   400
                        TIME (MINUTES)
                           64

-------
a:
o
J
5
o
5
O

H
O
CD
O

-------
tr
o

J
5

O
o
O
o
03
O
o:
UJ
o
UJ
o

if
cc
UJ
400



380



360


340



320



300


280



260


240


220



200


 180



 160


 140



 120



 100



  80


  60



  40



  20


   0
                Figure  21   SLUDGE DRAINAGE

                           DAF RUN 15

                           Summer, 1973
                                 Rote
                                = 0.68ml/min-
         AIR=6.0% by Volume

         PH=5.0

-Rote      ALUM=75mg/l

= i.97mi/min.  POLYMER = Img/l 835A

         SLUDGE FI_OW=2l.69m3/day

                      (3.98gpm)
             50    100   150  200   250  300   350  400
                        TIME  (MINUTES)
                           66

-------
rate for the smaller alum dose and a slightly faster rate in the second straight sec-
tion of the curve.  A disturbing factor, however, for the lower alum dosages
(Figures 18 and 19) is the long lag period between the straight sections of the
curve. The lag period is much shorter for the runs with a higher alum dosage
(Figures 16 and 17).  As a consequence the absolute drainage in ml  is greater
after 4 or 5 hours for the runs with a higher alum dosage.  In summary it appears
that alum feed concentration does have an affect on sludge characteristics, but
a simple relationship between alum concentration and sludge drainage does not
appear to exist.

The effect of polymer feed dosage oh sludge drainability was also investigated.
Figures 19, 20 and 21 show drainage tests where conditions are similar,except
polymer feed dosage  varies from  1 mg/l to 4 mg/l.  The three curves show very
similar shapes, drainage rates and absolute drainage at 5 hours.  It appeared that
polymer dosage had very little effect on drainage rate.

Injected air was another operating variable that could affect gravity sludge drain-
age.  However,  because of the limited number  of pilot runs, little information
about this variable was developed.   Figures 14  and 15 show results of runs in the
Fall of 1972.  All of these runs were made with air injected at  1.5% by volume.
The other sludge drainage curves were from runs in the Summer of 1973 where
all runs were made with air injected at 6.0% by volume.  Direct comparisons
cannot be made,  however,  because of differences in water temperature and
sludge removal rates.   The average water  temperature for the  Fall of 1972 runs
was 18° C and for the Summer of 1973 runs was 26° C. Also, during the Fall,
1972 runs two scraper blades were used where for the Summer,  1973 runs only one
scraper blade was used. Since sludge was removed at a slower rate during  the
Summer,  1973 runs,more drainage occurred in the flotation tank.  Figures 14 and
15 do indicate better drainage during the  Fall,  1972 runs as compared to the
Summer,  1973 runs.  However, because the variables were not  isolated, no con-
clusions can be drawn.

Overall it appears that neitheralum nor polymer dosages affect sludge drainage  rates
to an  observable degree.
BUCHNER FUNNEL

For DAF pilot plant runs  10,  13,  14 and  16 during the Summer of 1973, a Buchner
funnel  test of the sludge  skimmings was attempted.  The procedure in Appendix A
was followed.

Basically, an attempt was made to find a filter cloth and a filter aide that would


                                      '67

-------
optimize filtrate quality and filtration rate.  However, the few attempts that were
tried ended in failure.  The filter cloth had a tendency to blind very quickly ini-
tially.  However, if the sludge was allowed to stand in the funnel for several
minutes prior to filtration a considerably larger volume of filtrate could be obtain-
ed.  This was due to gravity drainage in the funnel.  The captured solids in ihe
sludge seemed to be very sticky and,  therefore, very difficult to filter.

The sludge used for the filtration test was fresh sludge which was considerably
aerated.  No filtration attempts were made with a deaerated sludge.

Because of the  limited test runs, the Buchner funnel  test results are probably only
an indication of what could possibly be achieved by vacuum filtration.  However,
because of the  nature of the DAF sludge,  vacuum filtration does not appear
promising  as a sludge concentration operation.
                                      68

-------
                                 SECTION VIII

                             PILOT SCALE STUDIES
GENERAL
Screening and centrifugal-ion of DAF sludge was pilot scale tested at Plant A during
the summer, 1973 canning season.  Preliminary pilot screening and DAF treatment
testing was performed during the fall,  1972 canning season.

The original intent of the pilot plant work was to demonstrate the treatability of
shrimp canning wastewaters by the DAF process and to develop specific criteria for
prototype design.  The scope of the project also included the same evaluation as
above for primary treatment such as screening and for sludge concentration.  Several
things conspired, however,  to prevent completion of all  objectives that were pro-
posed .

The most limiting variable not foreseen during the project conception was very
short plant runs due to a limited raw shrimp supply at Plant A.  The problems Plant
A experienced during this period were generally felt by most of the Gulf canners.
First, the shrimp catch was below average because of adverse weather  conditions.
this  caused tremendous competition for  the available supply, and severe inflation
of raw shrimp prices occurred.   This,  coupled with price freezes  in 1973 for the
finished product but not for the raw supplies, caused economic uncertainty. The
result was very poor canning seasons at Plant A during the study.   As discussed in
Section III, because of the perishable  nature of shrimp, by necessity the shrimp
plants must possess a very fast process capability.  This meant an  extremely short
plant run when the raw supply was meager.

Another important limiting  variable was the pilot plant operational inflexibility.
The pilot plant that was chosen had one mode of operation and could not easily be
changed.

The most frustrating variable has probably been experienced by all pilot plant ^in-
vestigators.  This was inexperience with the equipment being tested.  By the time
                                        69

-------
the investigator has learned to cope with the equipment idiosyncraeies and fre-
quent malfunctions, the studytime is over.  For this study the investigators certain-
ly had their quota of these problems.  This will also be discussed later in  the
section.

Because of these problems, only preliminary design criteria have  been developed.  We
feel the pilot plant studies have been positive, however.  The treatability of
shrimp canning wastewaters by the DAF process has been demonstrated. Also,
a sludge concentration method has been demonstrated.  Finally, from the  pilot
screening  studies a fairly specific design can be recommended.
PILOT SCREENING STUDIES
                                                 \                   r-  -
A total of six different screens were pilot tested during the Fall,  1972 and
Summer, 1973 seasons.  The screens were tested individually and in series
arrangements.  A tangential and a rotary screen are pictured in Figures 22 and 23.

The purpose of  the screening test was to evaluate  the efficiency and ease of opera-
tion of several  different types of screens.  Several of the larger canners had obtain-
ed experience in screening the shells and heads from their peeler wastewater with
vibrating screens.  These screens operated satisfactorily performing this function,
but were quite  expensive to operate and maintain.  None of the canners,  however,
had experience in screening the total wastewater  from a plant, which it was hypo-
thesized would be harder to screen than the peeler wastewater.  The total waste-
water would contain shrimp veins,  meat fragments and small shell fragments which
would  tend to blind a screen much quicker  than the larger shells and heads.  It
was felt that ease of operation and economical maintenance should be a prime con-
sideration in evaluating the pilot screens.

The screening test objectives were to be the following:

      1.    Evaluate  the most effective screen for performing a specific solids
            removal function.

      2.    Evaluate  the operating conditions of the screens such as mesh  size and/
            or rotational speed for the  rotating  screens.

      3.    Evaluate  series arrangements and compare pollutant removal efficiencies
            with other forms of treatment.

Most of the screening studies were performed during the summer,  1973 season.  A
                                      70

-------
Figure 22. PILOT TANGENTIAL SCREEN
  Figure 23.  PILOT ROTARY SCREEN
      SCREENING PEELER WATER


-------
plant scale vibrating screen,  however, was evaluated during the Summer, 1972
season, and a series arrangement consisting of a vibrating primary screen and a
secondary tangential screen was evaluated during the Fall,  1972 season.  The
screens were generally evaluated  by collecting composite samples of raw influent
and screen underflows during  the screening run.  Some of the screening runs were,
however,  tested by collecting grab samples.  This*was necessitated because  of
several pilot tests operating simultaneously which was required because of an
anticipated short canning season and short day runs.

A description and evaluation of each of the test screens follows:

Vibrating Screen.  This screen was a 1.22 m. (48") diameter Sweco Vibro-Energy
      Separator with an 0.84 mm  opening screen fabric.  This screen was circular,
      mounted  on coil  springs, and wastewater entered from the top.  The under-
      flow passed through the screen and the screened solids were vibrated with
      a spiral rolling motion  to the sides of the screen where they were discharged
      through two ports 180 degrees apart.   The vibrations were caused by an
      electric motor whose shaft was eccentrically loaded.  This screen was a
      permanent installation  at Plant A.  Wastewater from eight peeling machines
      and four  separators was discharged by centrifugal pumps to the screen. With
      eight peeling machines operating (the usual practice)  the flow to the screen
      was approximately 2725 mvday  (500 gpm).  The average results for this screen
      are shown in Table 25.

      This screen removed suspended solids very effectively; the  removal efficiency
      approached 40%.  The screen, however, was not nearly as efficient in re-
      moving settleable solids; the removal was less than 60%, leaving  a mean
      settleable solids  residual of approximately 20 ml/1  in the underflow.  BOD-5
      and total solids removal appeared to average at around 15% removal. The
      screened  solids were fairly dry with an average value  of 16.2% solids.

      At Plant A the screen surface was periodically sprayed throughout the day
      with a high pressure hose to prevent blinding.   The waste did tend to  par-
      tially blind the screen,  but not totally. The amount of spraying required
      for optimum operation is not known.  Since this unit was not a pilot unit
      but a necessary plant scale  operation,  no changes were made in  the screen
      during the study  period. Controllable variables,  such as degree of eccen-
      tricity of the shaft motor size, stiffness of the spring mounts, cleaning devices
      and screen mesh  were not evaluated.

Centrifugal Screen.  This screen was a 0.305 m. (12 in.) diameter Sweco centrifugal
     Wastewater Concentrator. With this unit, wastewater was pumped to the middle
                                       72

-------
     of a spinning cylindrical screen.  The liquid was spun through the screen and
     was removed as effluent.  The solids too large to pass through the screen
     dropped out and were removed as concentrated solids.  The manufacturer
     claims that the screen  rotational velocity in combination with the impinge-
     ment velocity of the influent results in a vector velocity that allows the
     screen to remove particles smaller in size than  the wire openings.

     The pilot screening unit tested was  operated and set up by Sweco personnel.
     Sampling and wastewater  analysis was done by  study personnel.  The unit
     was tested on total  composite discharge wastewater during plant  processing.
     The peeler and separator wastewater part of this discharge had been pre-
     screened with a 0.84 mm  opening vibrating screen.  Therefore, most of the
     large peeling solids had been removed,  but this wastewater did contain
     additional solids from deveining, blanching, inspection and canning opera-
     tions which were not pre-screened.

     The operating variables available were:  interchangeable 0.035 mm opening
     and 0.097 mm opening screens and  flow rate.   Seven test runs were made in
     which operating conditions were varied on each.  Average results are  shown
     in Table 26.

     Settleable solids removal  with this screen was excellent with the mean value
     being 89% removal.  BOD-5 and suspended solids  removals were 8.6% and
     17.2% respectively.  For the seven runs the concentrate flow averaged
     14.4% of the influent flow to the screen with a maximum of 36.4% and a
     minimum of 5.4%.   The higher concentrate flows generally occurred with
     the use of the 0.035 mm opening screen.  Removals of suspended solids
     were generally better using the 0.035 mm screen; however,  removal of BOD-5
     was observed to be better using the 0.097 mm screen

Rotating Screen.   This screen was a HydrocycIonics Rotostrainer.  The screen had
     a diameter  of 0.64 m.  (25 in.) and a length of 0.61 m. (24 in.). The unit
     had a screen opening of 0.5 mm.  The cylindrical  screen had the appearance
     of well screen with a wedge wire grid.   The unit was equipped with a  weir
     influent box for even influent distribution to the screen.  The water passed
     through the  screen openings on  the  top of the screen, fell through the  center
     of the cylinder andpassed through the screen openings again on the bottom,
     thus backwashing any solids trapped in the screen.  The solids were carried
     on the top surface  of the screen to a scraper bar where the solids were removed.

     The operational  variables  available for testing  this unit were: rotational
     speed of the screen and influent flow rate.  The unit was tested with raw
     peeler wastewater with shells.  The unit was tested for three runs.  The only
                                     73

-------
       Table 25.  VIBRATING SCREEN a EVALUATION
           POLLUTANT REMOVAL EFFICIENCIES
Parameter
BOD-5
Total Solids
Suspended Solids
Settleable Solids
Mean
Removal
%
16.1
15.1
39.9
57.1
Maximum
Removal
%
35.9
39.0
81.3
75.7
Minimum
Removal
%
1.8
11.8
23.8
41.9
a  Sweco 0.84mm opening
      Table 26.  CENTRIFUGAL SCREEN a EVALUATION
          POLLUTANT REMOVAL EFFICIENCIES
Parameter
BOD-5
Suspended Solids
Settleable Solids
Mean
Removal
%
8.6
17.2
89.0
Maximum
Removal
%
16.7
37.6
93.3
Minimum
Removal
%
0.0
3.4
84.7
a  Sweco Wastewater Concentrator
                          74

-------
     operational variable tested was the rotational speed of the screen. Consi-
     derable pump clogging problems were experienced in pumping the tested
     wastewater to the screen influent, therefore,  the flow rate variable was not
     evaluated.  The average results from evaluating the screen are shown in
     Table 27.

                Table 27.  ROTATING SCREEN °  EVALUATION
                    POLLUTANT REMOVAL EFFICIENCIES
Parameter
Suspended Solids
Settleable Solids
Mean
Removal
%
26.7
47.8
Maximum
Removal
%
48.5
60.6
Minimum
Removal
%
6.1
36.3
     a     Hydroclonics Rotostrainer

     The removal of suspended and settleable solids was somewhat less for this
     screen than for the vibrating screen even though the screen opening for the
     rotating screen (0.5 mm) was less than for the vibrating screen (0.84 mm).
     The screened solids, however, were fairly dry.  One sample was tested at
     22% dry sol ids.
                               >
Tangential Screens.  Tangential screens from three different manufacturers were
     evaluated.  They were:  the Bauer Hydrasieve,  the HydrocycIonics  Hydro-
     screen and the Dorr-Oliver 45° DSM screen.   The Hydroscreen and the
     DSM screens were tested with raw peeler water only and the Hydrasieve was
     tested with raw peeler water and total composite process wastewater (peeler
     water pre-screened).

     The Bauer Hydrasieve test unit was 0.457 m wide and 0.85 m high.  The test
     unit was supplied with screens with four different openings: 0.50 mm, 0.75 mm,
     1.0 mm and  1.5 mm.

     The Hydrasieve had a headbox and an influent weir for even influent distri-
     bution and had a mechanism to feed the wastewater on the screen tangential ly.
     The screen bars were wedgewire and were transverse across the screen.   The
     wedgewire bars curved downward between the  vertical supports to cause the
     flow to divide into separate streams between the vertical supports.   The
                                     75

-------
                                                                    X
manufacturer claims this helps prevent clogging and blinding.

This unit was used for pretreatment of wastewaters for the DAF pilot plant.
The wastewater screened at this location was total composite process waste-
water.  Therefore, the tangential screen was operating as a secondary
screen in series with the vibrating screen for peeler wastewaters and opera-
ting as a primary screen for the remainder of wastewaters produced  in the
plant (deveining, fluming, blanching, canning and raw receiving).  The
average results are shown  in Table 28.

          Table 28.  TANGENTIAL SCREEN a EVALUATION
                POLLUTANT REMOVAL EFFICIENCIES  b
Parameter
COD
Total Solids
Suspended Solids
Settleable Solids
Mean
Removal
%
4.4
4.7
0
55.6
Maximum
Removal
%
36.0
40.0
47.0
80.0
Minimum
Removal
%
0
0
0
39.0
 a    Bauer Hydrasieve (1.0 mm Screen opening) at 272 m //day (50 gpm)
 b    From Total Composite Process Wastewater
 The results indicate  very poor removal efficiencies of COD, total solids
 and suspended solids but fairly good removal of settleable solids.  The
 complete results for settleable solids  indicate a residual average  under-
 flow concentration -of 8.8 ml/1.  The average results above are all
 from runs using a 1.0 mm screen  opening.  Trials were made with a 0.5 mm
 opening screen but severe blinding resulted.

 Blinding was somewhat of a problem screening this type of wastewater.  The
 peeler wastewater had been pre-screened and, therefore, the only solids left
 in the i-nfluent water were small  shell fragments, shrimp veins and small meat
 fragments.   The small shell fragments tended to build up in layers between
                                76

-------
the transverse screen wedgewires.  This type of clogging usually started
at the top of the screen and caused a water jump or uneven water distribu-
tion which resulted in  a considerable amount of water in the screenings.  A
periodic hosing of the  screen (every 30 minutes) solved this problem.  The
screen also tended to completely blind over with a slimy material.  This was
impossible to remove with a hose and a wire brushing was required about
once an hour.

The Hydrasleve was also tested as a  primary screen on raw peeler wastewater.
All the screen openings available were tested at 272  m3/day (50gpm).  The
evaluation was limited, however, because only one short run was made with
each screen opening.  The results are shown in Table 29.   These results
indicate that the 0.50 mm opening screen produced the best results.  This
screen,  however, tended to blind fairly quickly with a slime build-up.  This
unit with  a 0.75 mm opening screen performed excellently during the short
test run.  Residual settleable solids in the underflow was only 14 ml/I.  The
other screen openings  (1.0 mm and 1.5 mm) also performed without blinding
problems,but solids removal was inferior to the 0.75 mm opening screen.  The
screened solids were extremely wet when first leaving the screen but tended
to gravity drain very quickly.  The screenings were 8% dry weight solids at
the point  of leaving the screen.  This was due probably to a noticeable amount
of water continuously  trickling from the end of the screen.  The test  unit was
probably several years old and the seals between the  sides of the wedge wire
screen and frame were worn causing water  to channel down the inside walls.
This was probably the major cause of wet screening solids.

The HydrocycIonics Hydroscreen was also tested with raw peeler wastewater.
This screen was similar in design to the Hydrasieve but with several differ-
ences, which include: the screening surface was actually three separate
screens, all at slightly different angles to the vertical; the influent weir did
not direct the water tangentially to  the screen but was actually a small  jump;
and the screening surface of the Hydroscreen test unit was about one foot
longer than the Hydrasieve.  The  Hydroscreen unit was also apparently new
and in excellent condition.

The differences in the  screen design were apparently  significant.  Table 29
shows the  solids removal  results for the Hydroscreen.  As can be seen, the
residual settleable solids in the underflow was 22 ml/1. This was considerably
higher than the Hydrasieve.  However, screened solids from the Hydroscreen
were approximately 18% dry solids when leaving the  screen.   This was due
to the solids staying on the screen much longer, and,  also, no noticeable amount
of water was  observed  trickling from the end of the screen.  Only one test  run
                                 77

-------
Table 29.  PILOT SCREEN EVALUATIONS
        272 m3/day Influent Flow

Manuarcturer
Hydras! eve
by "Bauer"






Hydroscreen by
"Hydrocyc Ionics'
DSMby
"Dorr-Oliver"




Rotostrainer by
"Hydrocyc Ionics'




Screen
(mm) Conditions
0.50

0.75

1.00

1.50

0.50
ii
0.50

0.71

1 . 00

0.50 10RPM
i
5RPM

2RPM

a Water tested - raw peeler water with
b ml/1
c mg/l



Test
Set. solids
Susp. solids0
Set. solids
Susp. solids
Set. solids
Susp. solids
Set. solids
Susp. sol ids
Set. Solids
Susp. sol ids
Set. solids
Susp. solids
Set. solids
Susp. sol ids
Set. solids
Susp. sol ids
Set. solids
Susp. solids
Set. solids
Susp. sol ids
Set. solids
Susp. sol ids
shells


Screen
Influent
70
1250
42
1240
50
1910
70
1065
70
2250
33
660
33
660
33
660
33
660
33
660
33
660



Screen %
Effluent Removed
12 82.9
1290 0
14 66.7
358 69.5
15 70.0
910 52.4
33 52.9
645 39.4
22 68.6
855 62.0
Screen Failed
by blinding
13 66.7
650 1.5
18 45.5
680 0
21 36.3
490 25.6
18 45.4
340 48.5
13 60.6
620 6. 1



                  78

-------
was made with this screen and a 0.5 mm screen opening was used at 272 m3/
day.  No blinding problems were observed during the test run.

The Dorr-Oliver 45° DSM screen was also tested with raw peeler wastewater.
This screen is also similar in design to the Hydrasieve.  Important differences
are: the DSM screen has a continuously curved surface; the DSM test unit
was approximately twice as high as the Hydrasieve; and the DSM test screen
was only approximately 0.305 m (12 inches) wide where the Hydrasieve was
0.457 m wide.  The manufacturer describes the operation of DSM screen as:

      The DSM screening device employs a concave wedge bar type of
      screen. In operation, the feed to the screen is introduced tangen-
      tial ly at the proper velocity and flows across the concave screen sur-
      face at right angles to the openings between the wedge bars.  Each  of
      the wedge bars acts as a knife on the underside of the passing slurry.
      A layer of liquid and undersize solid particles strikes the sharp lead-
      ing edge of the bar and is sliced off and directed downward through
      the slot into the screen box.  The layer peeled off by this unique
      slicing action produces undersize particle separations smaller than the
      size of the wedge bar openings,  thus preventing blinding of the screen.
      The oversize material continues to  flow across the screen being conti-
      nually dewatered as it moves along and is discharged at the far end
      of the  screen.  The volume of undersize that passes through the open-
      ings between the wedge bars is largely determined by how well the
      slurry is held in contact with the screen surface.  The continuously
      curved surface design and the velocity across the surface provide this
      holding action by the centrifugal force created.

In the investigators opinion, this screen has a superior design and it is regret-
able that a thorough evaluation could not be performed. Test runs  with screen
openings of 0.5 mm,  0.71 mm, and  1. 0 mm were made.  Results are shown in
Table 29.  The velocity across the face of the screen was very fast, and, as a
consequence, a slight blinding of the 0.5 mm screen caused a complete fail-
ure because of water discharged at the end of the screen. With the 0.71 mm
opening screen, residua I settleable solids of only  13 ml/1 in the underflow
was tested.   The 1.00 mm opening had a residual  settleable solids of 18 ml/I.
No indication of blinding was observed with these two screens.   The screened
material had  approximately 18% dry solids  content when leaving the screen.

The advantages and disadvantages of the tested screens will be discussed in
Section IX.
                                  79

-------
PILOT DISSOLVED AIR FLOTATION STUDIES

A pilot DAF treatment unit was evaluated for two canning seasons using the total
composite discharge from Plant A.  The first test season, the Fall of 1972 canning
season, was intended to be of a preliminary nature where equipment bugs could
be worked out and chemicals could be evaluated.

For the second season, the Summer of 1973  canning season, objectives were to
intensely evaluate the operational variables of the DAF process while treating
shrimp canning wastewaters.  Because of a multitude of problems these objectives
were not completely accomplished, but a considerable amount of operational data
was collected.

The first major decision facing the project was to select the type of DAF system to
pilot test.  Preliminary jar tests were performed in  the early Summer, 1972, can-
ning season.  An extremely good coagulant system  was found that produced a very
tough floe.  With this knowledge and knowing  that the Gulf shrimp canning
industry required an economical treatment system both in capital and operating
costs,  and, because of land shortages a  very compact system, the  decision was
made to pilot test the total pressurization system.

The Pacific DAF system manufactured by the Pollution Control Division of
Carborundum Company was selected.   This pilot system offered the  features:  a
screening and surge tank module, a chemical addition module and automatic pH
control with the flotation cell  module. The DAF pilot plant was set up and made
operational in the early Fall of 1972.  The testing program during this season was
titled the Preliminary Pilot Plant Phase.   The pilot DAF unit is pictured in Figures
24  through 27.
                                                                             s

The objectives for this project phase were:

            1.      Set up and de-bug the operation of the mechanical pilot plant.
            2.      Optimize chemical dosages.

This was believed to be a realistic program  that could be accomplished during a
two month canning season.

Figure 28 shows a schematic of the pilot plant set-up and sampling  stations.  The
sludge concentration operation and sampling were not part of this project phase.
The pilot plant was nominally rated at 272 rrP/day (50 gpm) and this flow was
used for this project phase.  Influent  flow was pumped from a wet well that con-
tained the total plant  composite flow to the top of a tangential screen.   The screen
underflow drained to a surge tank beneath the screen where the coagulant and  acid
                                     80

-------
Figure 24.  PILOT DAF UNIT

 Figure 25.  PILOT DAF UNIT

-------
      Figure 26.   PILOT DAF UNIT
PressurizaHon Pump, Tank and Air Injector
     Figure 27.  PILOT DAF UNIT
       Automatic pH Controller
                   82

-------
    Figure  28 DISSOLVED  AIR  FLOTATION
                PILOT PLANT SCHEMATIC
                AND SAMPLING STATIONS
  SCREENED
            TANGENTIAL
            SCREEN
SOLIDS


^ SCREEN INFLUENT
/I D\
                 PILOT
                 PLANT
                 INFLUENT
     SCREEN
     TANK
  ACID  CD
   COAGULANT
o
                                      SLUDGE
                                      COLLECTION
                                      TANK
                                                   PRESSURIZATION
                                                   TANK
                                                   PRESSURE
                                                   RELEASE
                                                   VALVE
                                                     POLYMER
SAMPLING
STATIONS
SLUDGE
BASKET CENTRIFUGE
SLUDGE CONCENTRATION
    RAW CANNERY WASTEWATER
    AFTER SCREENING THROUGH
    20 MESH VIBRATING SCREEN
                               83

-------
were added with a chemical addition pump.   The contents were then mixed with a
mechanical mixer and pumped to another surge tank which served as the wet well
for the pressurizing pump.  A recycle from the flotation cell  effluent was also dis-
charged to this tank.  The recycle flow was controlled with a float valve that was
controlled by the water level in the surge tank.  The recycle was strictly a method
of providing a continuous flow for the pump.  With a wastewater influent of
272 m3/day (50 gpm) the recycle would also provide 272m3/day (50 gpm) for a
constant pressurized flow of 544 m3/day (100 gpm) at 40 psig (2.81 kgf/cm2)
pressurization.  Air was injected into the pump suction through a rotometer imme-
diately downstream from the surge tank.  The wastewater was then pumped to a
non-packed, 1  minute detention time, pressurization tank. The wastewater then
passed through a pressure release valve and then  to the bottom of a coagulation
tube which was a small cylindrical tank at the center of the air flotation cell.  The
flow was upward in this tank because of air bubble buoyancy.  The wastewater dis-
charged from the top of the coagulation tube radially into the main  body of the
flotation cell.  Solids were scraped from the top  of the flotation cell into a sludge
hopper and effluent supernatant traveled from near the  bottom of the tank through
three standpipes to a top effluent weir where a portion  was recycled and a portion
discharged as effluent.

A total of 12 runs were made with the DAF treatment system during this phase of the
study. The constant operating conditions for all  runs were:

            1.      Flow Influent - 272 m3/day (50 gpm).
            2.      Injected Air -  0.17 m3/hr. (6 cu.  ft./hr.)
            3.      Pressurization - 2.81 kgf/cm2 (40 psig).
            4.      Acid Addition - Screen tank.
            5.      Alum Addition - Screen  Tank.
            6.      Polymer Addition - Flotation Cell Influent.

The variable conditions  for each of the runs were:

            1.      Alum coagulant dosage.
           2.      Polymer dosage.
           3.      pH.

      Pilot plant dimensions of interest were:
                   Diameter of Flotation Cell - 2.29 m,(7.5 ft.).
                   Volume of Flotation Cell - 8.24 m3. (290 ft.3)
1.
2.
3.     Side Water Depth - 2.01 m. (6.6 ft.).
4.     Surface Area of Flotation Cell - 4.11 m2 (44.2 ft.2)
                                      84

-------
           5.      Volume of Pressurization Tank - 0.38 m^ (100gal.).
           6.      Volume of Coagulation Tube - 0.38 m^ (TOO gal.)

     Computed design variables of interest were:

           1.      Flotation Cell detention  time - 22 minutes.
           2.      Overflow rate - 132.3 m/day (2.26 gpm/ft.2).
           3.      Pressurization Tank detention time -  1 minute.
           4.      Flocculation Tank detention time - 1 minute.
           5.      Assumed fraction of saturation in pressurization tank - 0.5.
           6.      Average Air/Solids Ratio = 0.0440.
           7.      Average Cell Solids Loading = 1.30 Kg/hr./m2 (0.267 Ibs./hr/ft.2)
Mechanical problems hampered all of the runs in the preliminary pilot plant phase.
The automatic pH control equipment malfunctioned from the start and manual pH
adjustment was required.  This was accomplished by using a constant discharge
acid feed pump, periodically checking pH values with a portable meter and making
continuous adjustments of the feed pump  discharge.  This method did not work well
and pH control was poor.

Alum coagulant dosages ranged from 150 mg/l down to 75 mg/l and  polymer dosages
from 5 mg/l to 0 mg/l.  Control of pH at 5.0 was attempted, ;but actual values
fluctuated from 3.0 to 6.7.  Table 30 shows the testing conditions for this testing
phase.

             Table 30.  PILOT PLANT  OPERATING CONDITIONS
                          Dissolved Air Flotation Unit
                          Preliminary Pilot Plant  Series
                                  Fall 1972
Run
1
2
3
4
5
6
7
8
9
10
11
J2 	
Date
11/13/72
11/14/72
11/17/72
1 1/20/72
1 1/20/72
11/21/72
12/01/72
12/05/72
12/05/72
12/05/72
12/05/72
12/05/72
Coagulant a
mg/l
150
150
150
150
150
150
150
150
150
100
100
75
.. 	 .^_^_w*_^~^
Polymer^
mg/l
0
0
5
2
5
5
5
2
5
5
2

PH
6.5
3.0
3^k
.0
6.7
6**9
.7
5n
.2
4f\
.9
4 "7
.7
40
.8
5/\
.U
5 A
. U
c 0
»/. v
— "•
a - aluminum sulfate b - magnifloc 835 A
85

-------
Best treatment was observed at alum and polymer dosages of 75 and 5 mg/I, respec-
tively, and with pH controlled at 5.0.  Trie control pf pH was observed to be extreme-
ly important for good treatment.  When the pH was at the optimum value of 5. 0 the
floated sludge had the appearance of chocolate malted milk, but when the pH wan-
dered from the optimum as much as ±0.5 the new sludge would be white and far
less dense.  The visual quality of the effluent would also almost immediately
worsen.

With the DAF plant operating correctly, the effluent was almost crystal clear, but
with a small amount of white haze and a small  amount of floe carryover. When  the
pH or one of the coagulants was not at optimum, the white haze intensified and, if
corrections were not made, turned to a turbid pinkish color which was the same color
as the  influent.  Table 31  shows the results of this testing phase.

                     Table 31.  PILOT PLANT EFFICIENCY
                            Dissolved Air Flotation
                         Preliminary Pilot  Plant Series
                                  Fall, 1972
Run
1
2
3
4
5
6a
7
8
9
10
11
12
COD
Influent
mg/l
• M
	
	
3450
2350
1900
3250
3050
3050
3050
3050
3050
Effluent
mg/l
2400
2200
1370
1800
1500
1400
1100
900
950
1300
1350
900
%
Removal
•••
—
—
48
36
26
60
71
69
57
56
71
Total
Suspended Solids
Influent
mg/l
— ^
—
—
825
490
118
354
220
220
220
220
220
Effluent
mg/l
..
456
220
350
355
158
82
60
74
64
68
32
%
Removal
«M

	
58
11
—
66
73
66
71
69
86
a   Plant processing frozen shrimp

The study was hampered by lack of processing at Plant A and short plant runs when
the plant did process fresh shrimp.  Raw shrimp supplies were very scarce and the
plant resorted to processing peeled frozen shrimp.  Run number 6 was made with the
wastewater from this product.  The DAF pilot plant was able to treat this wastewater
                                      86

-------
to about the same residual values as for raw shrimp.  One problem encountered was
foaming during  pumping.

The conclusions of the preliminary DAF pilot plant phase were:

            1.      The system was extremely sensitive to pH and this variable
                   would have to be controlled very precisely.
            2.      Alum dosages as low as 75 mg/l  provided good treatment.
            3.      Polymer dosages in the range of 2  to 6 mg/l would be required
                   for efficient treatment.

During the spring of 1973,  repairs were made to the pH control system and to other
minor mechanical problems.  The Pilot Plant Phase was started in May,  1973.
The objectives  of this phase were to:

            1.      Continue the chemical optimization.
            2.      Optimize solids loading rate.
            3.      Optimize air/solids ratio.

A total of 18 runs were made with the DAF treatment system during this phase of
the study.  Sixteen (16) of these runs were made at constant flow and overflow
rate while evaluating chemical  dosage (Pilot Series  I)  and two (2) runs were made
with constant chemical dosages while evaluating overflow rate.  (Pilot Series II).

For Pilot Series 1  the following operating  conditions were used for all runs:

            1.      Flow - 272 m3/day (50 gpm).
            2.      Injected Air - 0.68 m3 hr.
            3.      Pressurization - 2.81   kgf/cm   (40 psig).
            4.      Acid Addition - Surge Tank.
            5.      Alum Addition - Screen Tank.
            6.      Polymer Addition - Flotation  Cell  Influent.

The variable conditions for Pilot Series I, were the same as for the preliminary pilot
plant series.  In Pilot Series I computed design variables were the same as for the
preliminary pilot plant series, except for  the following:

            1.      Average air/solids ratio = 0.1339.                         2x
            2.      Average Cell Solids Loading = 1.61  Kg/Wm2 (0.33 Ibs/hr/ft.  ).

All of the DAF pilot plant runs in Pilot Series I had  generally a very short duration
Most of the runs were between one and two hours, which  was the same as the plant
processing times.  Between 40 and 50 minutes from start-up was required for the
                                       87

-------
formation of a good thick sludge layer and the simultaneous development of an
effluent that appeared well treated.  This time corresponded to two detention
times or two passes through the flotation cell.  Therefore, complete equilibrium
conditions were probably established during the two hour runs, but not for a
one hour run.

Alum coagulant dosages ranged  from  150 mg/l to 50 mg/l and polymer dosages
from 10 to 0.5 mg/l.  Best pollutant  removals were obtained at alum and polymer
dosages of 75 mg/l and 2 mg/l respectively.   A pH of 5.0  ±  0.2 was maintained
for most runs.  A pH of 9.0 was maintained for one run and  extremely poor treat-
ment resulted. Forthree runs, pH values from 6.1  to 6.5 were maintained and poor
treatment resulted.   Table 32 shows the test conditions for this pilot plant phase.

              Table 32.  PILOT PLANT OPERATING CONDITIONS
                          Dissolved Air Flotation Unit
                                 Pilot  Series I
                                 Summer 1973
Run"
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Date
5/29/73
5/30/73
5/31/73
6/01/73
6/04/73
6/04/73
6/04/73
6/05/73
6/06/73
6/07/73
6/07/73
6/1 1/73
6/12/73
6/13/73
6/13/73
6/14/73
Alum
(mg/l)
75
75
150
150
100
150
150
300
600
75
0
50
60
75
75
75
Polymer (a)
(mg/l)
10
5
10
5
4
4
4
4
4
4
4
4
4
2
1
0.5
pH
5.0
5.0
5.0
5.0
5.0
6.5
9.0
6.2
6.1
5.0
5.0
5.0
5.0
5.0
5.0
5.0
pH Control
Chemical
H2S04
H2 S04
H2S04
H2S04
H2SO4
^H ^i
None
Na OH
None
None
H2S04
H2S04
H2SO4
H2S04
H2SO4
H2S04
H2S04
      (a)  Magnifloc 835 A

The effluent with good runs was almost crystal clear with a turbidity of less than
20 units.  A small amount of floe carryover persisted and caused this small amount
of turbidity.  The effluent was visually crystal clear between floe particles.  The
effluent BOD-5 for good runs was below 400 mg/l, the effluent COD was below

-------
1200 mg/l, the effluent suspended solids was below 100 mg/l and the effluent
protein was below 600 mg/l.  Table 33 shows the COD and suspended solids removal
efficiencies for each run.  Table 34 shows the mean removal efficiencies for all
parameters.
                     Table 33.  PILOT PLAN EFFICIENCY
                            Dissolved Air Flotation
                                 Pilot Series I
                                 Summer 1973
Run
1
2
3
4
5
6
7C
8
9c
10

12
13
14
15
16
COD
Influent
ma/la
3400
3400
3400
4000
3600
3100
—
4000
--
3500
mm^
3200
4100
3300
3600
3200
Effluent
1500
1300
1600
1400
1500
1750
—
2000
— •
1250
__
1500
1250
1200
1250
1150
Removal
55.9
61.0
52.9
65.0
58.3
43.5
—
50.0
—
64.3
--
53.1
69.5
63.6
65.3
64.1
Total
Suspended Solids
nfluent
mg/la
510
590
620
427
333
344
-—
1135
— —
664
600
585
613
355
560
625
Effluent
mg/|b
350
110
210
152
68
320
— —
232
«"•"•
97
600
438
87
49
122
110
Removal
31.4
81.4
65.6
64.4
80.0
7.0
—
79.6
~""
85.4
0.0
25.1
85.8
86.2
78.2
	 - fc •
82.4
  a - Sample Station 2 P
  c - Very low treatment efficiency
b - Sample Station 3 P
                                       89

-------
                  Table 34.  DAF PILOT PLANT EVALUATION
                        Pilot Plant Phase, Summer,  1973
                     Pilot Series I - Chemical  Optimization
                         Pollutant Removal Efficiencies
Parameter
BOD-5
COD
Total Solids
Suspended Solids
Protein
Turbidity
Ortho Phosphate as P
Total Organic Carbon
Mean
Removal
%
65.1
59.0
14.9
65.6
52.5
83.0
27.5
61.4
Maximum
Removal
%
80.0
69.5
42.9
85.8
91.1
97.5
38.2
62.8
Minimum
Removal
%
50.0
43.5
0.0
7.0
25.7
61.9
15.4
60.0
Control of pH was again observed to be very critical. When pH varied as .much as
-  0.5 from the optimum of5.0, the sludge and effluent quality decreased signifi-
cantly. Lag time for the pH control system was a problem for the  first two runs.
When the influent pH wandered slightly, the system would over-compensate and
push the pH outside the acceptable range.  It would  take the automatic system 20
or 30 minutes to bring the pH back to a steady pH 5. 0.  The location of the con-
centrated feed was then changed from the screen tank (See Figure 28) to the surge
tank.   The pH control sensor was located immediately downstream from the surge
tank and less lag time and better pH control resulted.

The jar tests and the preliminary DAF pilot tests had  shown that the alum-acid
coagulant was very slowacting, and for this reason the acid was added to the
screen tank. When  the pH lag problem resulted, the alum and acid feeds were
changed to the surge tank and better pH control resulted.  However,  poor floe
formation became a  problem, so the alum feed was changed back to the screen tank.
Then  floe  formation and  treatment efficiencies immediately increased.
                                      90

-------
Pilot Series II was concerned with optimizing the solids loading rate.  From Pilot
Series I the chemical dosages were optimized at alum, polymer and pH values of
75 mg/l, 2 mg/l and 5.0 respectively.  All runs in Pilot Series II used these
optimum dosages.  Table 35 shows treatment efficiencies for this testing.  The
intention was to make several runs with these dosages at different influent flow
rates; however,  this pilot series was abbreviated because of a  premature ending  of
the shrimp season at the test plant.   Three runs, however,  were completed with
influent flow rates of 136  m3/day (25 gpm), 272 m3/day (SOgpm) and 408 m3/day
(75 gpm).  The influent suspended solids concentration for each run was slightly
different.  Therefore, flow and solids loading were  not directly proportional.
The results are shown in Figure 29.

                     Table 35.  PILOT PLANT EFFICIENCY
                           Dissolved Air  Flotation  Unit
                                 Pi lot Series II
                                  Summer 1973
Run &
Date
1
6/22/73





2
6/25/73

^^••KMMMV^^HVK^MM^H
, Test
PH
COD
Turbidity
Phosphate as P
Total Solids
Tot. Susp. Solids
Oi 1 & Grease
pH
COD
Turbidity
Phosphate as P
Total Solids
Tot. Susp. Solids
Oil & Grease
1
Influent0
mg/l
7.5
3600
450
90
4150
490
944
7.9
3200
370
80
3686
390
760
•^
Effluentb
mg/l
5.0
1800
105
75
4000
170
144
5.1
1550
20
70
3578
110
110
•-
%
Removal

50.0
76.7
16.7
3.6
65.3
84.7

51.6
f\ A t
94.6
12.5
2.9
71.8
86.8
_
      a - Sample Station 2 P
      b - Sample Station 3 P
                                        91

-------
       Figure  29  PILOT DAF PLANT  EVALUATION
        SOLIDS  LOADING  VS. TURBIDITY  REMOVAL
    3.50 -i
    3.00 -
cvi
 £
 ^  2.50
 o
 5  2.00 -|
 O
 _J
S  1.50
 UJ
 O
    1.00 -
0.50-_

    75
                .Influent Flow
                =408 m^doy
                                      Influent Flow
                                     - 272 m3/doy
                   I
                  80
                                    Influent Flow,
                                    -136 m3/day
                             I
                            85
 I
90
 i
95
100
                        TURBIDITY (% REMOVAL)
                              92

-------
. 25
218
            (40 gpm).
                             °PHmUm Ce" S°lidS loading is Wo**-""* «/ 1
                             a corresponding influent flow of approximately
                                                             r         7
                                                                        1.22 Kg/
 Pilot Series III was intended to be optimization of air/solids ratio; however, this
 series was not run because of lack of processing at Plant A.  Several values of air/
 solids ratios were computed from similar runs during the Preliminary Pilot Plant
 Series,  Pilot Series I  and Pilot Series II.  The results of these computations are
 shown in Figure 30. where A/S ratios are plotted against removal of suspended
 solids.

 From Figure 30 it appears optimum A/S ratios are within the range of 0. 10 and
 0. 15.   This is from 3  to 4 times literature values for other industrial wastes. With
 this very high optimum A/S ratio a total pressurization DAF system would be
 superior to a recycle  pressurization system.

 The concentration and flow rate of the  flotation sludge was measured for most of
 the pilot runs.  Mean results are shown in Table 36.

                   Table 36.  DAF PILOT PLANT EVALUATION
                         Flotation Sludge Characteristics
Parameter
Dry Solids0
Flow**
Protein
Units
%
m /day
mg/l
Mean
5.0
23.3
15,819
Maximum
6.7
32.5
26,318
Minimum
2.6
6.4
6,963
Attempts were made to find relationships between chemical feed concentration and
sludge characteristics; however, with the relatively small number of runs tested,
none could be found.  It seems certain, however, that sludge volume will increase
with increased chemical dosages.

The conclusions of the DAF pilot plant phase are:

             1.    The optimum chemical dosages were alum - 75 mg/l, polymer-
                   2.0 mg/l and pH-5.0.
                                       93

-------
    Rgure  30 PILOT DA F PLANT EVALUATION
      A/S RATIO VS. SUSPENDED SOLIDS REMOVAL
 0.25-1
  0.20-
  0.15-
V)
  0.10-
  0.05-
  0.00-
     50
 L
60
70
 I
80
 I
90
                SUSPENDED SOLIDS (% REMOVAL)
                         94

-------
             2.    The alum floe was slow in forming and should be fed into the
                   system as early as possible.

             3.    Concentrated acid for pH control should be fed into the system
                   immediately upstream from the pH control sensors in order to
                   reduce response time.

             4.    Polymer should be fed into the system immediately upstream
                   of the flotation cell.

             5.    The retention time of the DAF cell should be designed on an
                   optimum solids loading of  1.22 Kg/hr./m2 (0.25 Ibs/hr/ft2).

             6.    The injected air to the DAF cell should be designed on an
                   optimum A/S ratio of 0.125.

             7.    The influent flow  was concentrated by volume to an average
                   of 5.1 % floated sludge and 94.9 % treated effluent.

             8.    The floated sludge had an average dry solids content of
                   approximately 5. 0%.
 PILOT SLUDGE CENTRIFUGATION STUDIES

 The flotation sludge skimmed from the top of the DAF pilot plant was concentrated
 in a basket type  pilot centrifuge manufactured by DeLaval Separator Company.
 The centrifuge had the following characteristics:

             Method of Feed        :  Batch
             Feed Volume           :  9.5 liters (2.5 gallons)
             Basket Type            :  Solid
             Material Removal Method:  Skimmer

 The centrifuge was operated at  1200 G's for all the test runs.  The flotation sludge
was fed to the centrifuge while spinning at approximately 3.78 liters/min. (1.0 gpm)
from a bucket for a total feed volume of 9.5 liters (2.5 gallons).  The feed sludge
was fresh, having been collected from the DAF unit immediately prior to testing.
After feeding, the sludge was spun from 2  to 10 minutes, after which the material
was removed with a skimmer.  The  liquid phase was removed first and the volume
was measured. The solid phase or centrifuge cake qualitatively was soft and very
skimmable.  The  entire cake was removed from the basket except for a very th.n
                                      95

-------
layer next to the basket.  The volume was measured and the per cent dry solids
was analyzed.

A total of nine test runs were made with the basket centrifuge.  Spin time was
varies from two to ten minutes.  On two of the runs, polymer conditioners were
tested.  Average results are shown in Table 37.

                 Table 37. PILOT CENTRIFUGE EVALUATION
                                 Mean Results
Parameter
Feed Sludge
Centrifuge Cake
Centrate
Air
Mean
% Dry Solids
3.36
6.23
1.05
0.0
Mean
Volume
(Liters)
9.5
2.2
3.7
3.6
Using the mean results above the average solids capture was 77.8%.  The volume
reduction considering the cake and the feed sludge was approximately 77%.  The
appearance of the cake was solid but soft and could probably be pumped with a
positive displacement pump.  The separate additions of 10 mg/l of non-ionic and
cationic  polymers appeared to be of no value in increasing the percentage of dry
solids or  reducing the cake volume.

The conclusions of the pilot centrifuge studies are:

             1.     The evaluation was only preliminary because of the limited
                   number of runs.
                                                 \
             2.     The concentration of the cake and solids capture was encou-
                   raging for a preliminary test.
                                     96

-------
                                  SECTION IX

                                  DISCUSSION
GENERAL

This section will discuss, basically, the data and results presented in other sections.
Comparisons of different manufacturer's equipment will be made.  The advantages
and disadvantages, in ouropinion, of how the equipment treats or responds to shrimp
canning wastes will be discussed.   The problems encountered during  the study and
how they could influence the results and conclusions are discussed.  Finally, areas
where further study or research could be of benefit are discussed.
WASTEWATER CHARACTERISTICS

From the start of the study the hypothesis was made that wastewater characteristics
from all Gulf shrimp canning  plants would be similar because of similar processing
methods.  Section VI compares the total discharged wastewaters from six different
plants.  In order to make direct comparisons, weights of discharged pollutants were
computed in terms of weights  of processed shrimp.

The data indicate  that the original hypothesis was inaccurate.  Total COD dis-
charged varied by a three times (3x) factor among the plants tested.  Total solids
varied by a factor of almost four times (4x).   This information indicates that the
study plant (Plant  A) or any of the  Gulf plants does not have a "typical" total dis-
charged wastewater.  Because the total wastewater is composed of wastes from so
many plant operations and has so many variables,  it  is not surprising  that a
"typical" discharge does not exist.

Comparing the wastewaters from one unit operation, the peeling operation, gives a
great deal of insight into conditions that cause the variability.  The  peeling opera-
tion  is performed with the same type of equipment and supplied by the  same manu-
facturer.  The equipment operating variables are feed rate of shrimp per hour per
peeler and the quantity of water  used per unit of time.  Other variables would be
the size, age and species of the shrimp being processed.  Supposedly,  the size,
species and age  of the shrimp  govern the feed rate and the water required for
                                       97

-------
optimum peeling.  Theoretically, when processing the same type of shrimp, peeler
wastewaters should be very similar at different plants.

The results presented in Section VI indicates that peeler wastewaters were very
similar at four out of five tested plants for COD and total solids discharged.   The
fifth plant's peeler wastewater was not typical; however,  it was known that the
shrimp being processed on the test day were relatively old and,  therefore, not
typical.

The processing variables that exist beyond the peeling operation are as follows:

           1.       Method of conveying shrimp varies and  can  be pumping, wet
                   fluming or dry conveying.

           2.       The percentage of processed shrimp that is deveined.

           3.       The method of blanching, whether batch or  continuous.

           4.       Characteristics of the raw water supply.

           5.       Washdown period and methods.

These variables could be significant in establishing characteristics of the total
discharged wastewater, since  there were extreme differences measured at the  test
plants, although the measured peeler wastewaters were uniformly typical.  Because
the study  data gives little insight about the  relative importance  of the  plant process
variables  on wastewater characteristics,  each plant should be tested individually
to determine the characteristics of its total wastewater discharge.


TREATMENT BY SCREENING

A study objective was to examine the problem of screening all the process waste-
water from a shrimp canning plant.   This presented a problem significantly different
from screening just peeling wastewater, which has been the industry practice up to
this time.  The peeling wastewaters were screened fairly efficiently by vibrating
screens at Plants A, B and D and by a tangential screen at Plant  C. The residual
settleable solids in the screened wastewater generally averaged  about  15 ml/I  for
these installations.  The large amount of settleable solids in the peeling wastewater
probably assisted the continued operation of the screens by scouring the smaller
solids that caused blinding and clogging.  The total wastewater,  however,  would
contain shrimp veins and small fragments of  shrimp meat and shells not present in
the peeling wastewater.
                                       98

-------
Considerable data were collected on screening the peeler wastewater alone by
several types of screens.  Also, considerable data were collected at Plant A on
screening  the total wastewater discharge with the peeler water being prescreened
by a 20 mesh (0.84 mm opening) vibrating screen.  However, because of plant
piping limitations no data were collected on primary screening of the total process
wastewater.

On screening peeler wastewaters the following screens were evaluated:

          A.        Tangential Screens

                    1.    Bauer  "Hydrasieve".
                   2.    Hydrocyclonics "Hydroscreen".
                   3.    Dorr-Oliver "DSM".

          B.        Rotating Screen - Hydrocyclonics "Rotostrainer".

          C.       Vibrating Screen - Sweco "Vibro-Energy Separator".

Of the three types of screens the tangential screens produced the best effluent with
the lowest residual settleable solids concentration in the effluent.  These screens
also produced the wettest screening of the three types tested.  The rotating  screen
produced  the driest screenings and the vibrating screen performed midway to the
other two types for both criteria.

The tangential  screens consumed no power, therefore, were best for this category.
The vibrating screen was the worst and the rotating screen, which required only a
fractional horsepower motor, wasmidway.  The rotating screen was  only slightly
behind the tangential screens in this respect because it was much lower than the
tangential screens and the pumping head  required would be lower.

In ease of operation,the rotating screen was best.  During a short evaluation it
showed no tendency to blind or clog.   The vibrating  screen required a frequent
water hosing and was midway in this category.   The tangential screens required
frequent nosings and periodic brushing with a steel brush.

In anticipated operating cost,  the  rotating screen appears to be the best because
no operator would probably be required and maintenance should  be minimal.  The
tangential screen should be midway in this category even though no maintenance
costs are likely and  an operator will  probably be needed.   The vibrating screen,
because of ifc mechanical nature, is last because of expected high maintenance costs
and the need for an  operator.

On screening the total  process wastewater with the peeling wastewaters  pre-

                                       99

-------
screened,the following screens were evaluated:

           1.    Tangential Screen -  Bauer "Hydrasieve".
                                            i
           2.    Centrifugal Screen - Sweco "Wastewater Concentrator".

The centrifugal screen produced the best effluent with a residual settleable solids
concentration of about 1.0 ml/I.  This screen, however, removed only an  average
of 8.6% of BOD-5 and 17.2% suspended solids, so the residual concentrations in the
screened effluent were still very high. The disadvantage of this screen was the very
voluminous concentrate flow. This flow would need to be treated separately.
Treatability of the concentrate flow was not evaluated.

The tangential screen removed settleable solids to a residual of about 10 ml/1,
removed an average of 4.5% of BOD-5 and, on an average, removed no suspended
solids.  The screenings tended to be very wet because of a continual blinding pro-
blem which resulted in water discharged off the end of the screen.  The screen
tended to blind because of a slime layer which could  only be removed with a
wire brush.

In summary,  of the  three screen  types evaluated while screening peeler wastewaters,
the rotating screen  and the tangential screens seemed to be superior over the
vibrating screen.  In screening the total process wastewater the centrifugal screen
performed well,  but the large concentrate flow volume was a disadvantage.

In recommending a screen for total  process wastewater discharge,  the tangential
screen has to be the choice.  Its advantages of simple operation, low maintenance
and adequate screening efficiency far outweigh  its disadvantages.


TREATMENT BY DAF

A study objective was to pilot plant the dissolved air flotation (DAF) process  and to
develop design criteria for future full  scale design.  The first decision point was to
select the DAF type, whether full flow pressurization, partial flow pressurization or
recycle pressurization.  Literature information on protein wastewaters indicated
that either full flow pressurization or recycle pressurization should be chosen.

The full flow pressurization system offered the very large advantage of lower  capi-
tal costs.  The major disadvantage to this system, technically,  was that any floccu-
lation prior to pressurization would be sheared in the pressurization pump.  All
flocculation would then have to  occur in a short period within the flotation cell
and could result in immature floe and  poor separation.  If this occurred,  then this
type of system would not be suitable for treating shrimp canning wastewaters.

                                       100

-------
The recycle pressurization system had the reported disadvantage of high capital
costs.  Its advantage was that a floe could be preformed in a separate flocculation
tank and be fed to the air flotation cell by gravity, thus preserving the fully
matured floe.  This would facilitate good separation of the floe and the remaining
water.

FVeliminary jar testing was performed at Plant A during fhe summer of 1972.  A
coagulant system consisting of pH  adjustment, alum and an anionic polymer was
found to give an extremely  tough, stable  floe.  The decision was then made to
pilot test a total pressurization system for the first season and,  if results were un-
acceptable, then change to a recycle pressurization system.

A total  pressurization DAF pilot system manufactured by the Carborundum Company
was tested during the fall,  1972 canning season.  Mechanical problems throughout
this season prevented a thorough testing.   However, runs where proper controls
could be maintained resulted  in very efficient treatment with an effluent that was
close to sparkling clear.   A small  amount of floe carryover to the effluent did occur
even with the best runs, but there seemed to be no relationship between a small
amount of floe carryover and  effluent COD.  The polymer dosage seemed to have
the greatest effect on floe carryover because when no polymer was used, a great
amount of floe carryover occurred which resulted in much higher effluent COD.

Because of the potential that this unit demonstrated,  it was decided to pilot test
this unit for a second season if the manufacturer would completely  recondition the
unit. The corrective repair work was performed during the winter  and spring of
1973.

The pilot DAF unit performed very well mechanically during the summer,  1973
canning season.   However, Plant A  experienced an extremely poor processing season,
and therefore only about one-half the number of runs required for a thorough eva-
luation  were made.  Most of the runs were also of very short duration, lasting only
for one  or two hours.   It was doubtful  that complete equalization occurred with
the shorter runs.

Since the chemical addition system was not completely optimized during the pre-
vious testing season this testing  series was rerun.  The pH control system and the
chemical addition pumps performed well during this season.  Effluent quality was
observed to be very sensitive  to pH.  The optimum pH was in the range of 5.0 to
5.2.  If the pH was adjusted to  as high as 5.5 or as low as 4.5 the physical appear-
ance of the floated sludge started  to change immediately.  The color of the sludge
during optimum processing was a rich chocolate brown, and in non-opt.mum condi-
tions it  immediately started to turn white.  The white sludge had a very low dens.ty
of about one percent dry solids, and obviously very  little solids separat.on was
occurring.

                                        101

-------
Floe carryover was still a problem with the DAF unit during this period.  There
was again no observable relationship between effluent COD and minor degrees of
floe  carryover.  A major floe carryover problem was observed when runs were made
with polymer dosages of less than 2.0 mg/l.  This was in contradiction to the  jar
tests where good separations were made with  lower polymer dosages.  This pro-
bably demonstrates the economic trade off necessary when using a full pressurization
system; capital costs will be lower but operating costs will  probably be somewhat
higher.

Several preliminary runs or parts of runs were made using no chemicals or pH ad-
justment, but using air flotation only.  The effluent was not tested because there
was  no observable treatment.  The effluent appeared to be  of exactly the same
quality as the influent, and the floated sludge was extremely light and bubbly.
Very few captured solids were observed in this type of sludge.  Flotation bomb
bench  scale tests were made with no chemical but with full air.  Virtually no
reduction in  TOC resulted.  Runs were made with the DAF  pilot plant with in-
efficient coagulant systems and virtually no removals of  COD7 turbidity or
suspended  solids occurred.   These facts made  it very clear that an efficient
coagulant system must be used if significant  pollutant reductions are  to be
 achieved.

Another observed problem with the  DAF unit was the relatively long start up
time required.  The pilot unit was started each  morning  using fresh water. "The
chemical injection pumps and process pumps were started and adjusted prior to the
start up of the shrimp processing plant.  This was done in order  to get the maximum
run time possible and to check the start up  time required.  After the wastewater
entered the flotation cell approximately 40 minutes was required before a relative-
ly clear effluent was observed.  This time was approximately 2.0 retention periods
of the  flotation cell.   Therefore, it is obvious that if a shrimp cannery processes
for only one  or two hours and 40 minutes are required for treatment plant start-up,
a very poor daily average treatment will result.

The  operation and maintenance of a DAF treatment unit should  be relatively simple
for most shrimp canners. The plant system is composed of standard items such  as
pumps, valves and mixers which all shrimp  canners should be able to maintain with
few  problems.  The  DAF unit should operate automatically  and  should only require
periodic spot checks by an operator.
IN PLANT CHANGES

The treatment operations pilot tested during the study had the ultimate capabilities
of removing only insoluble or suspended solids.  This is also true of nearly all
                                      102

-------
other types of available treatment operations.  The exceptions are primarily bio-
logical treatment and reverse osmosis operations which have the ability to remove
solubles.  Therefore, the ultimate efficiency of practical treatment methods for
the shrimp canning industry will depend on the respective percentages of insol-
uble and soluble waste  pollutants.  Because of this, in-plant changes would be
desirable  in order  to decrease the percentage of soluble waste  pollutants in the
discharged wastewater.

A common method of transporting the raw product from one process operation to
the next is by water fluming.  This method of transport will increase the concen-
tration of soluble solids and will also increase the BOD-5 thereby reducing the
ultimate efficiency of the treatment.  The take-up of soluble pollutants by the
transport water will be  proportional to the time of transport and the agitation in
the flume.  Therefore,  it is assumed that if the flumes were eliminated and replaced
by dry conveyers a significant amount of soluble pollutants would not enter the
wastewaters.

During the bench  testing phase it was found that the age of the wastewater had a
profound effect on the treatability. COD removals of 70% and greater were
possible with fresh wastewater but only about one-half this efficiency was possible
with wastewaters as old as 12 hours.  Therefore,  biochemical changes are occurr t
ring, possibly' ammoniafication of the protein, that change the soluble/insoluble
pollutant ratios.

This same phenomena can occur when non-fresh shrimp is processed by the cannery.
This is probably a  major reason why the characterization data showed so much
scatter.  Therefore, the shrimp canneries can probably expect  low wastewater
treatment efficiencies when non-fresh raw shrimp are processed.

The Literature Review,  Section IV, compares soluble/insoluble pollutant ratios of
Alaskan shrimp peeler wastewater using Model A (raw peelers) and Model PCA
(pre-steam) peelers.  The PCA peeler wastewater showed a significantly higher
percentage of insoluble solids.  This should make the PCA peeler wastewater much
more  treatable.

The Gulf shrimp canners at the present use Model A peelers.  The canners should
initiate research into equipment modifications or commission original new designs
that will reduce the wastewater pollutant concentration or increase the msoluble
pollutant  percentage in order to make the wastewater  more treatable.
                                       103

-------
SOLID WASTE DISPOSAL

In this discussion flotation sludge will be considered a solid waste.  Two methods
of dewatering were investigated during this study:  centrifugation and gravity
drainage.  The economics of shell dewatering and drying of dewatered sludge and
shells will be discussed based on laboratory investigations of moisture contents and
unit weights.  The scope of  the project did not include pilot plant investigations
for dewatering shells and drying shells and sludge.

Sludge dewatering by centrifugation was investigated on a preliminary basis.  The
results of this testing are not verified because only one run with each test condition
was made.  Chemical conditioning was not optimized, and no noticeable benefit
was observed with  the two chemical  addition runs made.   The tests did establish
lhat centrifugation could,as minimum performance,  decrease the sludge volume one-
fourth and increase the dry solids content by two fold.  Optimization of this opera-
tion could greatly  increase this performance.

Sludge dewatering by gravity drainage was investigated on a bench scale for a
number of pilot DAF runs.  The sludge drained quite readily by gravity.  Appro-
ximately 4 hours of drainage was required to equal the centrifuge performance in
reducing the volume. This could be an economical method for  sludge dewatering.

Problems were observed, however, which could cause severe problems with a full
scale installation.  The  sludge separated initially into two phases with a  clean
interface.  The drained  water was on the bottom and the lighter sludge continued
to float.  The water was relatively clear with obvious good suspended solids
capture.  This continued for only about one  hour and then small amounts  of the
sludge began to settle.  It was quite obvious at this stage that even small  amounts
of agitation would promote large scale settling  of the sludge.  Because of this,
design of a continuous,  full  scale, gravity unit would be extremely difficult.  From
the batch bench scale studies, a prediction can not be made as to the efficiency
of a full  scale continuous unit.   A full scale batch unit would be difficult to
operate.

At this stage  the most feasible sludge dewatering operation appears  to be centri-
fugation.  From the pilot plant work, this operation can be sized with confidence
for a full scale installation.  The chemical optimization can be performed on-line,
if need be.
TREATMENT ECONOMICS

The estimated  treatment costs are presented in AppendixC, Cost Study Data.  Esti-
mated treatment costs are presented for an 8-peeler size cannery and a 4-peeler

                                      104

-------
size cannery.  Approximate costs for other cannery sizes can be obtained by
straight line interpolation.  The  estimated operation and maintenance (O & M)
costs are presented in Cost/day.  A day is defined as eight hours of operation.
To obtain the  O & M costs for two hours of plant processing, the costs would be
25 per cent of the values shown.

The capital cost  estimate is based on the hypothetical treatment system shown in
Figure 31. The estimate is broken down into the following elements:

     1.     basic treatment system,

     2.     auxiliary operations to basic treatment,

     3.     in plant changes, and

     4.     solid waste disposal.

The basic treatment system is  an  independent system in which water is treated by
DAF and the solid wastes are  dewatered and hauled to a land fill.  The auxiliary
operations are basically chemical additions that will greatly increase the treat-
ment efficiency of the DAF operation. In-plant changes basically involve modi-
fications of the method for transporting raw shrimp from gravity water flumes to dry
conveyers, with some wastewater piping changes included.  The estimates for dried
solid waste disposal include screening, dewatering, drying of screenings and sludge,
grinding and packaging of the solid waste.
                                               *
For an 8-peeler plant the estimated capital costs for a basic treatment system are
$172,800. With chemical addition operations added, the costs increase to
$208,000. When a drying system is added, the total estimated costs are
$287,450.

The O & M costs will vary directly with plant processing times.  The estimated
costs for basic treatment, including a., treatment plant operator and  solid wastes
truck driver, are  $104.00 per 8 hour day.  If a plant operates for a greater or
lesser length of time, the costs will  be approximately proportional to this value.

The O & M costs increase drastically when  chemical addition is added.  The
maximum costs will be associated with maximum possible treatment efficiency.
If a lesser degree of treatment is required for discharge, the chemical  O & M costs
could decrease drastically.  The maximum estimated costs are $51.37/day for an
8-peeler plant.  This would give a combined basic treatment plus chemical  costs
of$155.37/day.

If drying is added  to the basic system, the O & M costs rise only slightly.  Addi-

-------
               Figure  31  PROPOSED
           SHRIMP CANNING WASTEWATER
               TREATMENT SCHEMATIC
                                      OE WATERED

WASTEl GRINOING
« E PACKAGING j
MEAL; a
1 MARKETING









PLANT
w«biti PIPING
^.1 B

--<
SLUDGE
1 1
. | |DEWATERED SLUDGE
| «"""» | SLU^GE DEWATERING
	 1 "' 1
•
+ 1
r— L -i !



t

r i i
.SCREENINGS (SCREENING!
SCREENINGS ^ - / ^ /"
DEWATERING ' "* ^
L , 	 1 1
i *
RECYCLEj T
I SCREENINGS * ^ |
1 RECYCLEj J
1 "
\*

SLUDGE


i
1




P
WET
^ HAULING
*"" TO
LANDFILL


p

T P
WATER! PROCESS
(CHANGES 1






nicruAo^c nc


\



\
p SCREENED
WASTE
VVATFR
^J i
Z
TREATED NEUTRALIZED
WASTE WASTEWATER
WATER DISCHARGE
tA ' hi
£ >
_1
yi
i It
u

r-1— i - — ' —
CHEMICAL ' CHEMICAL
COAG. a
ACID
TO
- CAUSTIC
TO
pH 5.0 I PH 6.0 +
J 	 1 J 	 1

r

c,
^7
MTTDMATIWC -u
SELECTED PROCESS
WASTEWATERS
              LEGEND
                       ADJUSTMENT WITH
                       UNTREATED WASTE-
                       WATERS.
              D
BASIC TREATMENT UNITS
              LJ  ALTERNATIVE TREATMENT UNITS
                          106

-------
Honal costs associated with the dryer are fuel and power costs.  These are sub-
stantial,  but the dryer will replace wet hauling to a land fill.  The estimated
costs for an 8-peeler plant with basic treatment plus chemical addition plus drvina
is$158.57/day.                                                   K     y  9

The total treatment costs per day of processing for a cannery will be 0 & M costs
plus amortization of capital costs.   Amortization for this analysis was based on
120 operating days per year,  interest of 7% and expected equipment life of
10 years.  The total costs/day for an 8-peeler and a 4-peeler plant for each
treatment category follows in Table 37.  Taxes are not included.

                  TableSS:  WASTE TREATMENT ECONOMICS
                              TOTAL COST/DAY
                     (Amortized Capital Costs Plus O & M)
                                            8-Peeler        4-Peeler
          Treatment                          Plant           Plant
      	Cost/Day	Cost/Day

      Basic  Treatment                        208.22         154.40

      Basic  & Chemical                      280.83         216.46

      Basic, Chemical & Solids
        Drying                              330.05         248.23
An 8-peeler plant will process, on the average, 25,000 kg (55,000 pounds) of raw
shrimp during 8 hours of processing.  Approximately 3,628 kg (8,000 pounds) of
dried waste meal  can be expected  from dried shells, meat and sludge.  The
additional total costs/day for drying as compared to basic plus chemical is $49.22
per day.   Therefore, this analysis shows that if the waste meal is worth as little as
$0.0135Ag  ($0.006I/pound), the drying operation could  be economically feasible
for the 8 peeler cannery.

For a 4 peelerplant, average processing for 8 hours is 12,500 kg (27,500 pounds)
of raw shrimp.  Approximately 1,814 kg (4,000 pounds) of dried waste meal would
be produced.  The additional drying costs/day are $31.77/day  Therefore,  ,f dried
waste meal is worth as little as $0.0175Ag  ($0.0079 per pound), the drymg opera-
tion could be economically feasible for the 4-peeler cannery.
                                      107

-------
                                 SECTIONX

                               BIBLIOGRAPHY
1.    "Domestic Sewage and Seafood Processing, Wastewater Survey, Westwego,
      Louisiana, " U. S.  Department of Interior, Federal Water Pollution Control
      Administration, Robert S. Kerr Water Research Center, April,  1968.

2.    Preservation and Processing of Fish and Shellfish,  Quarterly Progress Report.
      Technological  Laboratory, Bureau of Commercial Fisheries, Fish and
      Wildlife Service, U.S.D.I., Ketchikan, Alaska, 1968.

3.    Pollution Abatement and By-Product Recovery in Shellfish and  Fisheries
      Processing, CRESA, EPA water Pollution Control Research Series,  12130
      FJQ, June, 1971.

4.    Soderquist, M.R.,  Canned and Preserved Fish and Seafoods Processing
      Industry Development Document for Effluent Limitations Guidelines and
      Standards of Performance, EPA.  Contract No. 68-01-1526, July, 1973.

5.    Peterson, P. L., Treatment of Shellfish  Processing Water by Screening and
      Air Flotation,  (unpublished), National Marine Fisheries Services,  Kodiak,
      Alaska, 1972.

6.    Dehn, W. T., and Holz, T.W.,  Investigation of Screening Equipment for
      Salmon Cannery Wastewater (unpublished) for National Canners Association,
      Seattle, Washington, December, 1972.

7.    Claggett,  F.G., and Wong, J., Salmon Canning Wastewater Clarification,
      Part II.  Fisheries Research Board of Canada, Vancouver Laboratory, British
      Columbia, February,  1969.

8.    Sawyer, C.N., Chemistry for Sanitary Engineers.  McGraw-Hill, 1960.

9.    Morrison,  R.T., and Boyd, R.N.,  Organic Chemistry,  2nd Edition,
      Allyn and Bacon, Inc., Boston,  1966.

10.   Weber, W.J., Jr., Physiochemical Processes for Water Quality Control,
      Wiley Interscience, 1972.
                                     108

-------
                                 /
11.    Eckenfelder, W. W., Jr., Industrial Water Pollution Control. McGraw-
      Hill, 1966.                             ~~	

12.    Claggett,  F.G., and Wong, J., Salmon Canning Wastewater Clarification,
      Part I.  Fisheries Research Board of Canada, Vancouver Laboratory, British
      Columbia, January, 1968.

13.    Mendenhall,  V., Utilization and Disposal of Crab and Shrimp Wastes,
      University of Alaska, Marine Advisory Bulletin No. 2.  NTIS, COM-
      71-01092, March,  1971.

14.   Novak, A.F., Summary of the First Meeting of Sea Grant Food Science
      and Technology, May 24,  1970.  Sea Grant Marine Advisory Program,
      Oregon State University, Corvallis.

15.   Rousseau,  J. E.,  Jr., Shrimp Waste Meal; Effect of Storage Variables on
      Pigment Content. Commercial Fisheries Review 22 (4).

16.   Burkholder,  P.R., et al, Nutritive Value of Shrimp Flour, Nature
      211:860-861.

17.   Schmidt, P.J., and Baker,  E.G.,  Indirect Pigmentation of Salmon and
      Trout Flesh with  Conthaxanthin, Journal  of the Fisheries Research Board of
      Canada,   26: 357-360.

18.   Standard Method For the Examination of Water and Wastewater,  American
      Public  Health Association,  13th Edition, 1971.
                                     109

-------
                         SECTION XI




                         APPENDICES




                                                 Page




A.     Bench Scale Test Procedures                   111




B.     Laboratory Analytical Methods                115




C.     Cost Study Data                             120




D.     Conner Members - A.S.C.A.                  127
                             110

-------
        APPENDIX A



BENCH SCALE TEST PROCEDURES
             111

-------
                              APPENDIX A

                     BENCH SCALE TEST PROCEDURES
JAR TEST

The screening of coagulants and flocculants for possible flotation use was accom-
plished by the conventional jar test described as follows:

    1.  To obtain the approximate coagulant dosage, one liter of wastewarer at
        pH 7.3 was slowly stirred and a 1% alum solution added until floe forma-
        tion was observed (120 mg/l).
                               ^  \
    2.  To obtain optimal pH, five beakers containing one liter of shrimp can-
        nery wastewater were employed with respective adjustment of pH to 3.0,
        5.0, 6.0, 7.5, and 9.5.  The alum dosage obtained from step one was
        then introduced during a three minute rapid  mix  (at 100 rpm) period.
        This was followed by slow  mixing at 20 rpm for 20 minutes to enhance flo-
        cculation. After 20 minutes of sedimentation, the supernatents were ana-
        lyzed for TOC and the optimum pH obtained (5.0).  4 mg/l of Anionic
        polymer was added to each beaker towards the end of the rapid mix  per-
        iod.

    3.  To determine the optimal coagulant and flocculant dosages, wastewater
        samples were adjusted to pH 5.0.  The jars were then rapid mixed with
        dosages of alum of 0, 30,  80,  120, and 240 mg/l.  4 mg/l Anionic  835A
        polymer was introduced at the end of the rapid mix  period.  Mixing con-
        ditions and analysis was  identical to that described  in step two.

        Next using an alum concentration of 80 mg/l at  pH 5.0,  the Anionic
        835A dosage was varied  at 0, 1, 2, 4, and 8 mg/l and optimal dosage
        determined,

    4.  The procedure employed to evaluate other coagulants and flocculants was
        similar to that described above.  The pH was maintained at 5. 0 except for
        the lime  study where the lime dosage itself determined the pH.

-------
AIR FLOTATION BOMB

The effect of varying operating conditions on ihe dissolved air flotation process
for the shrimp wastewater was evaluated by employing a flotation cell as shown in
Figure 13 .   The laboratory procedure as suggested by Eckenfelder and Ford is out-
lined as follows:

     1.   A two liter sample of wastewater (if direct pressurization was used) was
         introduced into the pressure chamber.  Prior to addition, the pH of the
         sample was adjusted to pH 5.0 and alum when necessary added.

     2.   Compressed air was applied  to the pressure chamber to attain the desired
         operating pressure (30 to 60 psi).

     3.   The air-liquid mixture in  the pressure bomb was shaken for one minute
         and allowed to stand for three minutes to  attain saturation.

     4.   The wastewater was next transferred to a graduate glass cylinder where
         polymer was added. For the pressurized recycle investigation, a mea-
         sured volume of effluent (computed from the desired recycle ratio) was
         pressurized and released to the appropriate volume of raw wastewater in
         the graduate cylinder.

     5.   After flotation separation in the graduate  was completed (approximately
         20 minutes), the clarified effluent was siphoned and analyzed for  total
         organic carbon.
SLUDGE DRAINAGE

The effect of pilot plant chemicals on the floated sludge was evaluated by a sett-
ling test in a 1,000 mi graduated cylinder.   The procedure was as follows:

     1.   A 1,000 ml graduated cylinder was filled  with the sludge skimmings from
         the DAF pilot plant.

     2.   An interface of floating sludge and water  formed.  Periodic readings of
         total time elapsed and interface height were taken.

     3.   A plot was made of interface height versus time.  The slope of this line
         at any point was the drainage rate.  The plots are shown in Figures  14
         through 21.
                                        113

-------
BUCHNER FUNNEL

The dewatering characteristics of the DAF process sludge by vacuum filtration was
investigated by the Buchner funnel test method.  The method used follows:

    1.  A conventional Buchner funnel, vacuum flask and vacuum pump
        setup was used.

    2.  A filter screen and filter cloth was placed in the bottom of the
        Buchner funnel.

    3.  The filter cloth was wet with distilled water and a  0.51 m (20-inch)
        Hg vacuum was pulled on the vacuum flask.

    4.  200 ml of sludge well mixed with a polymer was placed in the
        Buchner funnel.

    5.  The volume of sludge drawn through the filter was recorded every
        15 seconds until  the vacuum broke.

    6.  The filtered volume was plotted against time and a slope computed.
                                      114

-------
          APPENDIX B


LABORATORY ANALYTICAL METHODS
              /

   AND SAMPLING PROCEDURES
               115

-------
                               APPENDIX B

                   LABORATORY ANALYTICAL METHODS

                      AND SAMPLING PROCEDURES
COMPOSITE SAMPLING

Flow proportional samples were collected periodically during the pilot plant runs
and while sampling for the wastewater surveys.  The samples were collected every
15 minutes during the pilot plant work because of the short run times.  Samples
were collected and flow meters read hourly during the wastewater survey at Plant
A.  All samples were plant day composites which  varied from 2 hours to 12 hours.
A shorter interval would have been desirable because of short plant runs, however,
because of the large number of sample locations,  one sampling run took approxi-
mately 45 minutes.  The samples were mixed with previously collected samples for
a given sample station immediately after collection.  The  mixed samples were refri-
gerated at approximately 5° C.  Analyses usually  began immediately after the com-
posite was collected and was complete approximately 24 hours later.
SAMPLE PREPARATION

All samples were collected as they were leaving the plant process or discharge be-
ing sampled.  No screening of settleable solids were done by the investigators.
Some of the wastewaters were screened by the plant, however.  This was explained
in Section VI, Wastewater Surveys.

There was no  sample preparation for the settleable solids test.  Settleable solids
were read from the raw water as collected.  The wastewater used for the settleable
solids test was not re-used for the remaining tests. The hourly sample quantity at
each station was adjusted daily after an estimated plant run time was received so
that approximately 1.5 liters were collected at each station.  This was sufficient
volume to perform all tests.

The comppsite sample remaining after the settleable solids test was blended in a food

                                     116

-------
blender at high speed for approximately 30 seconds. This was necessary because of
the large amount of shrimp flesh and shell fragments in most of the samples and the
extreme difficulty in pipetting small representative volumes.  After blending,  a
separate sample for Kjeldahl  nitrogen was acidified accordinq to "Standard
Methods" ( 18  ).
CHEMICAL OXYGEN DEMAND

The COD sample was withdrawn from a blended, well mixed composite sample and
diluted in a volumetric flask because the COD for most of the waste streams was
strong.  The COD was then analyzed by the dichromate reflux method of "Standard
Methods" ( 18) for a 10 ml sample.  Because of the high chloride content of the
wastewater control,tests were performed to optimize the dosage of mercuric sulfate.
A dose of 0.2 grams per sample was sufficient. A reflux time of 2 hours was used
for all determinations.
 BIOCHEMICAL OXYGEN DEMAND

 The "Standard Methods" ( 18) procedure was used for BOD-5 determinations.  A
 Delta Scientific Model 85 oxygen meter with a BOD bottle fitted agitated probe
 was used for D.O. determinations.  All samples were neutralized to pH 7.0 before
 dilution in the BOD bottles.  The dilution water was not seeded because of the
 extreme biodegradabilify of the wastewater samples.  The D. O. was measured in
 each bottle prior to incubation and after incubation.
 PH

 A Photovolt Model 85 A pH meter was used for all pH measurements.  The meter
 was standardized daily with standard buffer.
TURBIDITY AND ORTHOPHOSPHATE

A Hach Model DR filter photometer was used for these measurements.  Samples
were diluted in order to make mid-range readings.  For orthophosphate determi-
nations the sample was filtered prior to dilution and reagent addition.
OIL AND GREASE

The ASTM standard  test (D1340-60) for Oily Matter in Industrial Waste Water

                                       117

-------
was used with revisions for oil and grease determinations.  Two extractions were
made with hexane on an acidified sample.  The oil-solvent layer was placed in a
tared Erlenmeyer flask.  The solvent was then condensed at a very low heat, the
flask was dried at 103° C, cooled in a dessicator and weighed.  The residual was
computed in mg/l  and reported as oil and grease.
ALKALINITY

The alkalinity was determined by titration with 0. 02 N h^ SO ..  The titration end
point for total alkalinity was determined l?y constructing a titration curve for each
sample location and titrating to the end point by using a pH meter.
SOLIDS

Total solids were determined by first evaporating to dryness 50 ml of sample in a
tared evaporation dish at  103  C.   The dish was then cooled  in a dessicator and
weighed with an analytical  balance.  The residue was  reported as total solids in
mg/l.  The residue was then ignited in a muffle furnace at 600  C for  15 minutes,
allowed to cool  in a dessicator and again weighed.  The remaining residual was
reported as total fixed solids in mg/l.  The difference  in the  two yalues was report-
ed as total volatile solids.

Suspended solids were determined by first filtering a sample through a  tared asbes-
tos mat filter.  The tared crucible containing the filter and filter residue was dryed
at 103° C.  The dish was then cooled in  a dessicator and weighed.  The residue
was reported as total suspended solids in  mg/l.   The  same procedure as above was
used for determining fixed suspended solids and volatile suspended solids.

Dissolved solids were determined by subtracting suspended  solids values from total
solids values.
SETTLEABLE SOLIDS

Settleable solids were determined by filling a 1,000 ml Imhoff cone with an un-
blended , unscreened sample.  The sample was allowed to settle for 45 minutes and
then the cone was gently turned several times to loosen solids sticking to the sides
of Hie cone.   The sample was then allowed to settle for an additional 15 minutes.
The settleable solids in ml/1 were then read from the graduated cone.
                                      118

-------
PROTEIN
The "Standard Methods"  (18) procedure was used for organic Kjeldahl nitrogen,

The protein concentration was obtained by multiplying the organic Kjeldahl
 .,	j.	j.?	I	L nc
nitrogen concentration by 6.25.
                                         119

-------
   APPENDIX C




COST STUDY DATA
      120

-------
                                APPENDIX C

                             COST STUDY DATA


GENERAL

Cost study data includes estimated costs (January, 1974) of the following:

          1.    Basic wastewater treatment operations,

          2.    Auxiliary operations to the basic operations above,

          3.    Plant piping and process changes,

          4.    Solid waste disposal operations,

          5.    Operation and maintenance costs.

The first four groups constitute a complete treatment and disposal system.  A
schematic of the system is shown in Figure 31.   Estimated treatment costs of each
of the four groups above are made for two basic shrimp canning plant sizes;
an 8 peeler plant with a total process wastewater  discharge of 600 gpm (0.0378
m  /sec.), and a 4 peeler plant with a total process wastewater discharge of 300
gpm (0.0189 mVsec.). For plants of different peeling capacities than listed above,one
can interpolate between the two costs given to obtain a reasonable estimate.

Basic wastewater treatment operations include the following:

          1.    Screening by tangential screens.

          2.    Dissolved air flotation (DAF) cell complete with flocculation tank,
                pressurization  pump  and tank, recycle piping, sludge scraper
               mechanism, sludge hopper and sludge removal pump.

          3.   Sludge dewaterlng by centrifugation.
                                        121

-------
          4.    Wet hauling of screenings and dewatered sludge to a landfill.

Auxiliary wastewater treatment operations will greatly increase the efficiency of
the DAF unit and may be required by enforcement agencies.  Hie auxiliary
operations are:

          1.    Chemical coagulant feed system.

          2.    Polymer feed system.

          3.    Acid feed system with automatic pH control at pH 5.0.

          4.    Caustic feed system to treated effluent to meet  pH discharge
                standards.

Plant piping and process changes cost estimates are given for the following:

          1.    Conversion from wet fluming of shrimp to dry conveying.

          2.    Separation of concentrated wastewaters and relatively clean
                wastewaters.

Solid waste disposal operations include the following:

          1.    Screenings dewatering.

          2.    Drying of dewatered screenings and sludge.

          3.    Air pollution control devices for the dryer.

          4.    Grinding of dried solids.

Operation and maintenance costs include the following:

          1.    Salary of an operator.

          2.    Maintenance costs including power.

          3.    Chemical costs.

The cost estimates given are for a finished installation, therefore, the  costs
include equipment and installation costs.  The cost estimates, for each of the
above groups follows:

                                      122

-------
BASIC TREATMENT SYSTEM
                                                       4-Peeler    8-Peeler
                                                        Plant       Plant
                  —                                  Cost       Cost

!•   Screens                                           $  7f5QQ   $ 15/000
2.   Pump to Screens                                      ^800      2,300
3.   Dissolved Air Flotation Unit                          45,000     6o'oOO
4.   Pump to DAF Unit                                    1,800      2^300
5.   DewateredSludge Pump                               4,300      7 500
6.   Dewatering Centrifuge & Pump                        30,000     35,000
7.   Screenings Conveyor                                 3,700      3,700
8.   Dewatered Sludge - Screenings Holding Tank            8,000     12,000
9.   Tank Truck                                         20,000     25,000
10.  Miscellaneous                                       5,000     10,000
          Total                                         $127,100   $172,800

AUXILIARY OPERATIONS TO DAF UNIT

1.   Coagulant Feed System                             $  3,800   $  4,200
2.   Polymer Feed System                                V4,500      5,000
3.   Acid Feed System with Automatic pH Control           14,000     16,000
4.   Caustic Feed System                                  9,000     10,000
          Total                                         $ 31,300   $ 35,200
 IN PLANT CHANGES

 1.    Conversion from Wet Fluming of Shrimp to Dry
        Conveying                                     $  12,000   $ 20,000
 2.    Separation of Waste Discharges to Clean and
        Polluted Wastewaters                               3*°°0      5'600
          Tofa|                                        $  15,000   $ 25,600
                                       123

-------
SOLID WASTE DISPOSAL
              Item
1 .    Screenings Dewatering with a Compactor
2.    Rotary Dryer for Dewatered Screenings and
        Sludge
3 .    Dryer Afterburner
4.    Grinder
5.    Building
6.    Miscellaneous - Contingencies & Equip. Foundations
          Total
OPERATION & MAINTENANCE

A.    Basic Treatment

      1 .   Operator - Treatment Plant
            8 hours @$4.00/h°ur
      2.   Driver - S olid Wastes
            8 hours@$3.00/hour
      3.   Land  fill fee
            $5.00/Load
      4.   Tank  Truck
            $0.25/mile
      5.   Maintenance and Power®
             Total

B.    Auxiliary  Operations

      1 .   Alum
            $0.05/pound
      2.   Polymer
            $1 .75/pound
      3.   Acid
            $40.00/ton
      4.   Caustic
            50%@$.50/Gal
      5 .   Power - 2 Hp = 1 .5 kw
6.  Maintenance
        Total

   a - A Day is defined as 8 hours of operation.

                                 124
4-Peeler   8-Peeler
  Plant       Plant
  Cost       Cost

$  3,000   $  37700

  53,000     68,000
  23,500     30,000
   1,800      2,250
   7,500     10,000
   2,000      2,400
$ 90,800   $116,450
 Cost/Day"   Cost/Day0


$  32.00   $  32.00

   24.00      24.00

   15.00      25.00

   15.00      15.00
    6.80       8.00
$  92.80   $ 104.00
$   6.00   $  12.00
    4.38
                                                                8.75
10.00
4.25
.10
2.00
20.00
8.50
.12
2.00
                                                      $   28.83
              51737

-------
OPERATION & MAINTENANCE (Continued)
                    Item

C.   Solid Waste Disposal

     1 .   Operator
            8hours@$3.00A>ur
     2.   Dryer Fuel
            $0.75/1 ,000 cf of Nat.  Gas
     3 .   Power
     4.   Packaging
     5.   Maintenance
             Total
 4-Peeler    8-Peeler
  Plant      Plant
 Cost/Day   Cost/Day
$  24.00   $  24.00

   16.50      33.00

    1.00       1.20
    2.50       5.00
    3.00       4.00
$  47.00   $  67.20
SUMMARY

Capital Costs are:

      Basic Treatment System

      Basic Treatment System plus Auxiliary
       Chemical Operations

      Basic Treatment System plus Auxiliary
       Chemical Operations plus solids drying
       system.  (Holding Tank and Tank Truck
       eliminated)

The operation and maintenance costs/day are:

      Basic Treatment System

      Basic Treatment System
       plus Auxiliary Chemical Operations

      Basic Treatment System plus Auxiliary
       Chemical Operations plus Solids Drying
       system. (Tank truck, driver and land fill
       fee eliminated)
                                                           Cost       Cost

                                                        $ 127,000 $  172,800


                                                          158,300    208,000
                                                          221,100   287,450

                                                        Cost/Day  Cost/Day

                                                        $  92.80  $  104.00
                                                          121J
             155.37
                                                          114.1
             158.57
                                       125

-------
 It must be emphasized that the above operation and maintenance cost estimates
are based on 8 hours of treatment plant operation per day.  If a plant processes
for only 2 hours, then the operation and maintenance costs will be approximately
25% of the costs shown above.

Total costs/clay include amortization of capital costs plus operation and maintenance
costs.  Amortization is based on 120 operating days per year, annual interest of
7% and an equipment life of 10 years.  Operation and maintenance costs are
based on 8 hours of processing.  Taxes are not considered.

Total Costs/day are:                                       4-peeler    8-peeler
                                                           Plant         Plant
                                                         Cost/Day   Cost/Day

      Basic Treatment System                               $  154.40   $208.22

      Basic Treatment System plus
         Auxiliary Chemical Operations                       216.46    280.83

      Basic Treatment System plus
         Auxiliary Chemical Operations
         plus Solids Drying System
         (Tank truck, driver and holding tank
         eliminated).                                         248.23    332.05

To obtain the treatment costs for a cannery of different size than the  two listed,
a reasonable estimate can be obtained by straight line interpolation.
                                       126

-------
             APPENDIX D

           MEMBER CAHNERS
AMERICAN SHRIMP CANNERS ASSOCIATION
                  127

-------
Sponsor of this Shrimp Canning Waste Treatment Study was the American Shrimp
Canners Association, P. O. Box 50774, New Orleans, Louisiana, 70150. Member
canners are listed below:
          Biloxi Canning Co., Inc.
          P.  O. Box 1168
          Biloxi, Mississippi  39533

          Authement Packing Co., Inc.
          P.  O. Box 380
          Dulac, Louisiana  70353

          Calvin J. Authement Packing Co., Inc.
          Grand Caillou Box 311
          Houma,  Louisiana 70360

          Buquet Canning Co.
          P.  O. Box 909
          Houma,  Louisiana 70360

          C. C. Canning Co., Inc.
          P.  O. Box 12
          Biloxi, Mississippi  39533

          Cutcher Canning  Co.,  Inc.
          P.  O. Box 8
          Westwego, Louisiana  70094

          DeJean Packing Co.
          P.  O. Box 509
          Biloxi, Mississippi  3o533

          Deepsouth Packing Co., Inc.
          P.  O. Box 13145
          New Orleans, Louisiana 70125

          Grand Caillou Packing Co., Inc.
          P.  O. Box 430
          Houma,  Louisiana 70360

          Gulf Coast Packing Co., Inc.
          Grand Caillou Route
          Houma,  Louisiana 70360
Mr. D. J. Venus
Mr. Huey Authement
Mr. Calvin J. Authement
Mr. A. J. Buquet
Mr. William Cruso, Jr.
Mr. A. J. Cuccia, Jr.
Mr. R. H. Sewell
Mr. Ray Skrmetta
Mr. Emile Lapeyre, Jr.
Mr. Richard B. Samanie
                                      128

-------
Indian Ridge Canning Co.,  Inc.
P. O. Box 550
Houma,  Louisiana  70360

Johnson Seafood Co.
P. O. Box 117
Biloxi, Mississippi  39533

Louisiana Packing Co.,  Inc.
P. O. Box 129
Chauvin, Louisiana 70360

Mavar Shrimp & Oyster Co., Ltd.
P.O. Drawer 208
Biloxi, Mississippi  39533

Reuther's Seafood Co., Inc.
P. O. Box 50773
New Orleans, Louisiana  70150

Robinson Canning Co., Inc.
P. O. Box 10
Westwego, Louisiana  70094

Sea  Coast Co.,  Inc.
P. O. Box 502
Biloxi, Mississippi  39533

Southern Shell Fish Co., Inc.
P. O. Box 97
Harvey,  Louisiana 70058

Southern Shell Fish Co., Inc.
P. O. Box 162
Biloxi, Mississippi 39533

Southland Canning & Pkg. Co.,  Inc.
P. O. Box 23220
New Orleans, Louisiana 70123

Trosclair Canning Co.
P. O. Box 67
Cameron, Louisiana  70631
Mr. Gordon Mi I let
Mrs. Louis Johnson
Mr. Lynn Chauvin
Mr. Vic Mavar
Mr. C. G. Reuther, Jr.
Mr. Alan J. Robinson
Mr. Nick Cerinich
Mr. Victor Blereau
Mr. Delacaiz
Mr.  Paul P.  Selley
Mr. Roland Trosclair, Jr.
                              129

-------
                 Weems Brothers Seafood Co.       Mr. Charles Weems
                 1124 East Bay View
                 Biloxi, Mississippi  39530

                 LeTur Foods Co.,  Inc.            Mr. James J. Tur
                 P. O. Box 1163
                 Slidel'l, Louisiana  70458
The consulting engineer for this project was Domingue, Szabo & Associates, Inc.,
P. O. Box 52115, Lafayette,  Louisiana  70501.
                                       130

-------
  SELECTED WATER
  RESOURCES ABSTRACTS
  INPUT TRANSACTION FORM
1. Report No.
                                                                        3.  Accession No,
                     w
  4.  Title
  Shrimp Canning Waste Treatment Study
  7.  Author(s)

  Mauldin. A. Frank, and Szabo. A.  J.
                      5.  ReyortDate
                      jf. - •+•*•
                      Si  Performing Organization
                         Report So.
                     10.  Project No.
  9.  Organization
  Domingue, Szabo & Associates, Inc.
  Consulting Engineers
  Lafayette, Louisiana
                     11.  Contract/Grant No.
                          S 800 904
                     13.  Type o./Report and
                         Period Covered
  12.  Sponsoring Organization
  IS.  Supplementary Notes

  Environmental Protection Agency report number, EPA-660/2-7U-061, June 1971*
  16.  Abstract
  Wastewater surveys were performed at several Gulf shrimp canneries over a period of three canning
  seasons. Water used for each process within the plant was metered and the wastewater was tested
  for biological, chemical and physical characteristics.  Pilot screening tests were made over two
  canning seasons.  Tangential, rotary and vibrating screens were evaluated.  A 272 cu m/day ( 50
  gpm) dissolved air flotation pilot plant with themical addition and pH control was tested at the
  study plant over two canning seasons. A pilot basket centrifuge was evaluated for sludge dewatering
  The study demonstrated that:  (1) the waste poundage discharged per pound of raw shrimp processed
  is similar in most Gulf shrimp canning plants; (2) screening removal of heads and shells can be per-
  formed efficiently and with few operational problems; and (3) air flotation showed promise as a
  wastewater treatment method.  When performing properly, treatment efficiencies were good, how-
  ever, the operation was sensitive and treatment efficiencies that can  be expected on a plant scale
  remain to be demonstrated.
  17a. Descriptors

  Dissolved air flotation, screening, wastewater characterization,  in plant changes, sludge
  dewatering, treatment costs, chemical optimization.

  17b. Identifiers

  Gulf shrimp processing and canning wastewater.
  17c. COWRR Field & Group   05A,  05B, 05 D
{8. Availability
19. ' Stf&rit? Ctass.
(.Report.)
2(7. Security Class.
• (P4ge) ,
21. Ho.ol:
Pages ...
22. Price
Send To:
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 2O24O
                                           Institution
WREIC IO2 (REV. JUNE 1971)

-------