EPA-660/2-74-020
APRIL 1974
Environmental Protection Technology Series
Evaluation of Polymeric Clarification
of Meat-Packing
and Domestic Wastewaters
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
4. 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 vievs
and policies of the Environmental Protection Agency, nor does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
-------
EPA-660/2-74-020
April 1974
Evaluation of Polymeric Clarification of
Meat-Packing and Domestic Wastewaters
K. D. Larson
D, A. Maulwurf
Project 12130 EKK
Program Element 1BB037
Project Officer
C. C. Oster
Minnesota-Wisconsin District Officer
U.S. Environmental Protection Agency
7401 Lyndal Avenue South
Minneapolis, Minnesota 55423
Prepared for
OFFICE OR RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
T6f «le by the 8apertat«nd«nt of Doenmtnti, U.8. Qovcrauwnt Printing Office, Washington, D.0.30102 - Price $2.JO
-------
ABSTRACT
Tue U.S. Environmental Protection Agency, the City of Souta St. Paul,
and the Sewage Disposal Commission of Soutix St. Paul (since October
1970 tne Metropolitan Sewer Board) undertook a project to demonstrate
tiie effectiveness of using organic and inorganic flocculating agents
in the treatment of combined packinghouse and domestic wastes. Tae
project was directed at not only a single system of flocculating
agents, but at dual or multi-systems of organic and inorganic
flocculating agents of wastewater.
A dual system of chemicals found effective in the treatment of tats
corauined waste was ferric cnloride and a combination of one of two
organic poly electrolytes, eitaer Dow A-23 or Halco 675. These tx/o
systems were effective in forming a floe large and dense enough
to settle out under the dynamic condition of an overloaded primary
sedimentation tame. Treatment with either of tuese dual systems
effectively reduced suspended solids in the effluent of toe primary
sedimentation tank over tuat achieved without the use of tuese dual
systems. This was demonstrated by running: (1) a parallel system of
identical test and control primary tanks and, (2) a full-scale plant
investigation for botu test and control periods.
Tue primary flocculation of tiie collodial and fine particles by
the ferric chloride produced a small volumnous floe. Secondary
flocculation of tiie organic polyelectrolyte increased tiie floe
size and density, producing a settleaule precipitate. Tue ferric
cnloride alone was unsatisfactory, producing a non-settleable
floe under prevailing uydraulic conditions.
ii
-------
Additional laboratory data were collected from various other sampling
points throughout the plant during the test and control period
when full-scale plant tests were run. Data were gathered on
the effluent from the secondary settling tank following the trickling
filters and on the effluent from the anaerobic stabilization pond which
is the last stage of treatment. The overall results showed a leveling
off of difference in removal achieved by chemical treatment between
effluents from the trickling filters and the anaerobic pond.
The cost of chemically treating the combined 438 I/sec (10 mgd) of
wastewater would be approximately $12 per million liters ($45 per
million gallons). The economic breakdown of applying the dual system
of inorganic and organic chemicals would be: 72% for chemicals,
15% for construction, 12% for labor, and 27, for utilities.
This report was submitted by the City of South St. Paul, Minnesota,
Metropolitan Sewer Board, in fulfillment of Project No. 12130 EKK,
which was partially sponsored by the U.S. Environmental Protection
Agency. Work was completed as of July 1970.
iii
-------
CONTENTS
SECTION PAGE
I. Conclusions 1
II. Recommendations 3
III. Introduction 5
IV. Design and Construction 11
V. Experimental Puase 19
VI. Plant-Scale Evaluation of Patented Process 37
VII. Laboratory Evaluations of lion-Patented Systems 70
VIII. Plant Evaluations of Non-Patented Systems 83
IX. Additional Discussion of Non-Patented Systems 96
X. Results 127
XI. Publication and Patents 149
XII. Appendices 150
iv
-------
FIGURES
NO. Page
1 Flow Diagram - Current 8
2 Ciiemical Mixing Caamoers and Splitter Boxes '12
3 Flow Diagram - Before Construction Project 13
4 Scnematic of Plant Site Including Anaerobic Pond 14
5 Detailed Flow Diagram - Current 15
6 Polyelectrolytes Initially Screen in tue Lab 32
7 Halco 675 and Anuydrous FeCl 102
8 llalco 675 and Waste FeCl3 103
9 Dow A-23 and Amiydrous Fed
104
10 Dow A-23 and Waste FeCl3 105
11 FeCl3 Only at 50 mg/1 113
12 FeCl Only at 100 mg/1 114
13 Full Plant Scale Cuemical Tests 134
-------
TABLES
No.. £*£
LABORATORY WORK. BLAISDELL AND KLASS PROCESS
1. Preliminary Work, Laboratory Screenings 21
2. Blaisdell and Klass Method, Laboratory Data 25
3. Polyelectrolytes Giving Best Results in Lab Tests, 27
Blaisdell and Klass Process
4. Effects of Variation of Polyelectrolytes and 28
Bentonite In Laboratory Tests
5. Polyelectrolytes Given More Extensive Lab Testing
30
6. Effectiveness of Sodium and Calcium Bentonite In 33
BOD Removal
7. Effectiveness of Different Forms of Sodium Bentonite 34
In Lab Tests
8. Effect of Concentration of Bentonite Suspension 34
9. Weighting Agent Laboratory Tests 36
PLANT SCALE TESTS. BLAISDELL AND KLAAS PROCESS
10. Chemical Dosage Anionic 120 mg/1, Cationic 3 mg/1 39
Bentonite Prepared At 8%
11. Chemical Dosage Anionic 120 mg/1, Cationic 3 mg/1, 40
Bentonite Prepared At 4.0% and 2.7% solids
12. Chemical Dosage Anionic 50 mg/1, Cationic 50 mg/1 42
13. Chemical Dosage Anionic 30 mg/1, Cationic 30 mg/1 43
14. Chemical Dosage Anionic 15 mg/1, Cationic 15 mg/1 44
15. Chemical Dosage Anionic 45 mg/1, Cationic 15 mg/1 46
vi
-------
No. IS8S.
16. 200 Mesh Bentonite, Starch and Ferric Chloride In 47
Separate Tanks. Chemical Dosage Anionic 30 mg/1,
Cationic 30 mg/1
17. Starch Separated From Ferric Chloride and Iron Ore, 47
Chemical Dosage Bentonite 30 mg/1, Ferric Chloride
50 mg/1, Iron Ore 10 mg/1, Starch 5 mg/1
18. Flash Mixers Off, Starch Separated From Iron Ore And 49
Ferric Chloride Chemical Dosage Same As Table 17
19. Detention Time 1-1/2 Hours, Starch Separated From 49
Ferric Chloride and Iron Ore, Chemical Dosage Bentonite
30 mg/1, Ferric Chloride 50 mg/1, Iron Ore 10 mg/1,
Starch 5 mg/1
20. Polyethylene Imine mixed separately from the ferric 51
chloride and iron ore. Chemical Dosage Bentonite
30 mg/1, Ferric Chloride 50 mg/1, Iron Ore 10 mg/1,
Polyethylene imine 5 mg/1
21. Iron ore mixed with starch; ferric chloride mixed 51
separately, Chemical Dosage Bentonite 30 mg/1,
Ferric Chloride 50 mg/1, Iron Ore 10 mg/1, Starch
5 mg/1
22. P.E.I, and iron ore mixed; ferric chloride mixed 52
separately, mechanical flocculators on. Chemical
Dosage Bentonite 30 mg/1, Ferric Chloride 50 mg/1,
Iron Ore 10 mg/1, PEI 5 mg/1
23. Flash Mixers off and flocculators off. Chemical 54
Dosage Bentonite 30 mg/1, Ferric Chloride 50 mg/1,
Iron Ore 10 mg/1, Starch 5 mg/1
24. Mechanical flocculators off. Chemical Dosage 55
Bentonite 30 mg/1, Ferric Chloride 50 mg/1, Iron
Ore 10 mg/1, Starch 5 mg/1
25. Mechanical flocculators off. Chemical Dosage 57
Bentonite 30 mg/1, Ferric Chloride 30 mg/1, Iron
Ore 10 mg/1, Starch 5 mg/1
26. Mechanical flocculators off. Chemical Dosage 58
Bentonite none, Ferric Chloride 15 mg/1, Iron Ore
10 mg/1, Starch 5 mg/1
vii
-------
No. Page
27. Mechanical flocculators off. Chemical Dosage 58
Bentonite 8 mg/1, Ferric Chloride 15 mg/1, Iron
Ore 10 mg/1, Starch 5 mg/1
28. Mechanical flocculators off. Chemical Dosage 60
Bentonite 4 mg/1, Ferric Chloride 9 mg/1, Iron Ore
6 mg/1, Starch 3 mg/1
29. Mechanical flocculators off. Flash mixers off. 61
Chemical Dosage Bentontie 8 mg/1, Ferric Chloride
15 mg/1, Iron Ore 10 mg/1, Starch 5 mg/1
30. Starch into parshall flume following parallel 62
addition of bentonite and ferric chloride. Chemical
Dosage Bentonite 4 mg/1, Ferric Chloride 37.5 mg/1,
Iron Ore none, Starch 3 mg/1
31. Order of addition same as Table 31, Chemical Dosage 64
Bentonite 4 mg/1, Ferric Chloride 9 mg/1, Iron Ore
none, Starch 3 mg/1
32. Ferric chloride and iron ore mixed. Starch into 65
parchall flume following parallel addition of ben-
tonite and ferric chloride. Chemical Dosage Bentonite
none, Ferric Chloride 9 mg/1, Iron Ore 6 mg/1, Starch 3
mg/1
33. Order of addition the same as Table 32. Chemical 66
Dosage Bentonite 4 mg/1, Ferric Chloride 9 mg/1, Iron
Ore 6 mg/1, Starch 3 mg/1
34. Bentonite added to two small flash mixers. Ferric 67
Chloride and Iron ore premixed and added to the
large flash mixers. Starch added to the parshall
flume. Chemical Dosage Bentonite 4 mg/1, Ferric
Chloride 9 mg/1, Iron ore 6 mg/1, Starch 3 mg/1
LABORATORY TESTS ON OTHER CHEMICAL SYSTEMS
35. Single Flocculant Evaluation 72
36. Single Flocculant Summary 73
37. Dual Flocculant Evaluation 74
38. Consistancy Tests 76
viii
-------
39. Order of Addition Tests - Dow A-23 & Ferric 80
Chloride
40. Order of Addition Tests - 675 and Ferric Chloride 81
PLANT SCALE TESTS ON OTHER CHEMICAL SYSTEMS
41. Control Day - No Chemicals Added 84
42. Dow A-21 86
43. Ferric Chloride and Nalco 675 90
44. Ferric Chloride and Dow A-23 92
45. Statistical Analysis of One Week Run 94
46. Summary of Test Data 99
47. Summary of Test Data 107
48. Ferric Chloride Only 112
49. Jar Tests on Separate Industrial Wastes 116
50. Jar Tests on Separate Domestic Wastes 117
51. Wastewater Separation Tests - Summary of Test 119
Data
52. Data for Wastewater Separation Tests 123
FULL PLANT SCALE TREATMENT
53. Average Daily Chemical Dosage 130
54. Flow and Chlorine Data, Control Period 135
55. Data of Full Plant Scale Chemical Treatment, 136
Composite Sample
56. Flow and Chlorine Data, Test Period 141
57. Data of Full Plant Scale Chemical Treatment, 142
Grab Samples
58. Flow and Chlorine Data Summary
ix
-------
N°i. Page
59. Settleable Solids Pond Effluent 144
60. Economics 147
-------
ACKNOWLEDGMENTS
Sincere appreciation is given to personnel from the U.S. Environmental
Protection Agency for their assistance in finding a workable system
of polyelectrolytes and setting up a workable system for plant-scale
tests. Personnel giving direct assistance were Mr. Jack L. Wituerow
and Mr. Robert Crowe of the Robert S. Kerr Environmental Research Laboratory
at Ada, Oklaaoma and Dr. Sid Hanna from tae National Environmental Research
Center at Cincinnati, Oaio.
Ttie Project Manager was H. G. Keeler. TUB Project Officer at tue
beginning of the project was 11. J. Snyder. C. C. Oster was made Project
Officer in November 1968. Project Director was K. D. Larson. Consulting
Engineer was D. S. Blaisdell.
Laboratory analyses were done by the staff of the Soutu_St. Paul
Treatment Plant Laboratory. The following personnel worked in the
laboratory for all or part of toe project: D. A. Maulwurf, R. A.
Zeroth, M. L. Maulwurf, J. F. Voss, J. K. Pike, M. C. Swansoti, L. A,
Pulford and R. A. Tauer.
Tae plant scale chemical treatment facilties were operated for all or
part of the project by C. R. De Wolf, 11. V. Reed, J. F. Voss, K. D.
Caristophersen, K. A. Caristopiiersen and II. C. Swanson.
Acting as a biological consultant for tae project were Dr. K. N. Knutson,
Professor of Biology at tae St. Cloud, Minnesota State College. He did
a considerable amount of work studying tae biota of the biological treatment
process of tne plant.
xi
-------
SECTION I
CONCLUSIONS
Tue use of chemical flocculation has been demonstrated in the treatment
of combined 90 percent meat packing and 10 percent domestic wastes. Various
chemical treatment schemes were demonstrated in an overloaded primary
portion of tae trickling filter plant at South St. Paul, Minnesota.
Of tae various treatment scaemes demonstrated, including an unsuccessful
multicuemical system using a patented process, a dual system of ferric
cnloride followed by a aiga molecular weight anionic organic polyelectrolyte
proved to be the most effective. Tnis combination of organic and inorganic
polyelectrolytes in the primary portion of the plant reduced tue suspended
solids and BOD in combined packinghouse and domestic wastes by 56 mg/1 and 58
mg/1 respectively over tae control period of one month, full scale operation
without chemical addition. Suspended solids removal in this process was
255 mg/1 whicli resulted in a primary effluent of 101 mg/1. This is a 72
percent reduction as compared to 56 percent for the control period.
Correspondingly, tae BOD removal for this dual chemical system was 314
mg/1 and the primary effluent aad a concentration of 269 mg/1. Tae efficiency
of BOD removal was also greater, 54 percent compared to 29 percent for
the control period.
Biological investigation of tae biota of the trickling filters
indicated that tue chemicals did not nave a detrimental effect on taat
treatment process. Tha suspended solids and BOD removal efficiency
of the trickling filter remained unchanged from the control period to
the test period. This resulted in a lower concentration of suspended
solids and BOD in tne effluent from the trickling filters. Test period
results for suspended solids and BOD concentrations were 52 mg/1 and 6.6
mg/1 respectively wuile tae control period snowed 68 mg/1 and 1Q7 mg/1
for the corresponding concentrations.
-------
These tests demonstrated' taat a significant increase in reduction
of suspended solids and BOD can be obtained by the use of a dual chemical
system in the primary portions of tae plant. Also demonstrated in
the investigations was tiie value of correlative jar testing
before field testing of flocculants. Laboratory jar testing was used
to screen tae various chemical treatment schemes to select tiie best
for full plant operation. This approach proved to be valuable in
tiie saving of time and money.
Based on a 438 I/sec (10 mgd) flow, the total cost of chemical treatment
was $12 per million liters ($45 per million gallons). These costs were in
addition to the normal operating costs of trie existing treatment facilities,
In a breakdown of the costs for full-scale utilization of the dual-chemical
system of anhydrous ferric culoride and DOW A-23 or Nalco 675, 71 percent
was required for caemicals, 15 percent for construction, 12 percent for
labor, and 2 percent for utilities.
There were two additional benefits from the chemical treatment process.
One was the reduction in culoriiie demand from 13.9 mg/1 to 5.5 mg/1. The
other was that the overall pnospaorus reduction for the plant was doubled.
This left a 4.7 mg/1 concentration in tae secondary effluent for a 52
percent removal.
-------
SECTION II
RECOMMENDATIONS
Tuis study was made on a plant testing combined domestic and packinghouse
wastes with primary treatment and secondary treatment consisting of trickling
filters and an anaerobic stabilization pond. Tests were conducted on tne
effectiveness of various chemical systems in the primary portion of the
plant and on tae overall effect of tne successful process through the
biological portions of the plant.
If the process found to be successful at this plant is considered for
other plants, additional tests saould be conducted before the process
is put into operation.
Further tests saould be conducted to determine if chemical addition at
a different point in the plant flow would be more effective. In the tests
at South St. Paul, chemicals were added after the grit chambers and
before primary sedimentation. Another possible point of addition could
be after the trickling filters and before final sedimentation.
It would also be desirable to run tests to determine what effect these
chemical systems have in plants having biological treatment other than
trickling filters. The overall effect of this dual system of chemicals
in an activated sludge plant might be different from that obtained in a
trickling filter plant. A biological examination of the effects of these
chemicals on activated sludge microorganisms would also be desirable.
Tests should also be conducted to determine if other plants experience a
chlorine demand reduction wita tae use of chemical treatment. It is possible
tuis effect was a result of the use of caemicals in conjunction with the
-------
treatment by the anaerobic stabilization pond at South St. Paul. If a
reduction did occur at otuer plants using tuese systems of cnemical
treatment, a savings in cost of chlorine would result which would partially
offset the cost of the caemicals.
Additional testing should be conducted to determine if the phosphate
removal could be improved by introducing the chemicals at some other
step in the treatment process other tuan in the primary clarifiers
and also if the addition of lime along with the flocculant ciiemicals
used would increase tne phospuate removal.
-------
SECTION III
INTRODUCTION
During 1940, South St. Paul, Minnesota, completed a 438 I/sec (10 mgd)
nigh-rate trickling filter plant to treat combined domestic and meat
packing waste. When the meat packers were operating the strength of
the waste increased from 445,000 population equivalents per day in 1940
to 917,000 population equivalents per day in 1964. These figures were
reduced to approximately 10 percent on weekends. During this period
the efficiency of the plant dropped from 72 percent to 58 percent removal
of biochemical oxygen demand (BOD)•
In 1958, it became apparent taat additional treatment would be required
by both tne adjacent Minneapolis-St. Paul Sanitary District Plant
and South St. Paul Plant. The primary concern at that time was the
low levels of dissolved oxygen in the Mississippi River during the
late summer months.
Pilot plant studies conducted at South. St. Paul indicated that
the plant effluent BOD could be reduced by more than 50 percent by
passing tne effluent througa an anaerobic pond of sufficient capacity
to afford at least a five-day detention period. This addition was
initiated in February, 1962.
During the three-year period, April 1, 1959 - Marcn 31, 1962, the
South St. Paul effluent nad an average BOD of 464 mg/1 during
killing days. The State of Minnesota alllowed an effluent quality
of 565 mg/1 BOD which was anticipated as matching that of the Minneapolis
St. Paul Sanitary District effluent waen their new (then under bid)
high-rate activated sludge plant was completed. The anaerobic pond
reduced the BOD of tne normal plant effluent by 66 percent, but this
-------
still left an effluent of approximately 150 mg/1, 2 1/2 tines more than
the 65 mg/1 allowed. Consequently, tae City of South St. Paul engaged
an engineering firm to study tne loads and capacity of the plant and
to recommend a program to meet the maximum BOD allowed of 65 mg/1, and a
method of sludge disposal other than lagooning.
Shortly after tne engineering report was presented to the City an
Upper Mississippi River Pollution Abatement Conference was held jointly
by Wisconsin, Minnesota, and tae U.S. Environmental Protection Agency
(EPA) . On completion of trie conference the plant effluent limit of
65 mg/1 BOD was lowered to 50 mg/1 and subsequently lowered to
35 mg/1 by tae State of Minnesota.
In February 1965, the engineering firm developed several new plans encompassing
second stage filtration and incineration costing approximately $10,000,000.
The packing industries contracted an engineering firm to review this
report waicn. recommended treating hot packing plant wastes by the anaerobic
contact process. The cost for tne entire package again approached $10,000,000,
a figure the industries found unacceptable.
The industries become interested in the possible efficiency and economy
of cnemical treatment of meat packing wastes, A review of tne literature
and existing patents indicated that the use of chemicals might reduce
construction costs by $2,000,000. The City of South St. Paul made application
to the EPA for a demonstration grant to determine the effectiveness
of a chemical flocculation process on combined domestic and packing plant
wastes, or on pure packing plant wastes. Hew grit chambers were necessary
along with a new intercepting sewer from tae industries and a new pumping
station to separate tne meat packing wastes. The increased costs of new
grit chambers, chemical feed building, flashmixers, feed equipment, chemicals
and personnel to undertake the demonstration were estimated at $045,000.
EPA agreed to participate at 55 percent of the cost, not to exceed $450,000.
Tae City of South. St. Paul and the packing industries agreed to fund the
remainder of the project cost.
-------
Even without the anaerobic stabilization pond, the South St. Paul
Sewage Treatment Plant is not a conventional trickling filter plant
XFigure 1.).
The sewage entered the plant, passed through a 2.54 cm (1-inch) bar
screen, and was pumped into three grit tanks. At design flow of
438 I/sec (10 mgd) the detention time in these tanks was 30 minutes.
During the killing hours at the meat packing plants the peak flow
was 876 I/sec (20 mgd), thereby reducing the detention time by 50
percent. From the grit tanks the sewage flowed through mechanical
flocculators. The detention time at 438 I/sec (10 mgd) in the flocculation
tanks and primary clarifiers is 45 minutes and 1 1/2 hours, respectively.
Between the mechanical flocculators and rectangular primary clarifiers
is a baffle wall. The primary effluent goes to six trickling filters
and then to the final clarifiers which have a detention time of 11/2
nours at 438 I/sec (10 mgd) . lite secondary effluent is pumped to
the anaerobic pond with 5 days detention and then discharged to the
Mississippi River. The sludges from the primary and secondary clarifiers
are pumped to the sludge lagoons. The grit and paunch manure are stored
in a diked earth basin adjacent to the plant.
When the Federal demonstration grant was initially applied for and
received, the system of chemical treatment to be tested was that
described in Patents No. 2,434,321 and 3,142,638 held by Mr. Blaisdell
and Mrs. Klaas. The demonstration project testing the effectiveness
of polyelectrolytes was begun in June, 1967.
This system calls for the addition of a cationic polyelectrolyte,
an anionic polyelectrolyte, some finely divided flocculant aid such
as sodium bentonite clay and an inorganic flocculant aid such as ferric
chloride, ferrous sulfate or alum and a weighting agent which is
incorporated in the floe as it is formed to increase the density of the
-------
m
i
NEW
INDUSTRIAL
INTERCEPTOR
INTPffCEPTOR SEWER
TRANSPORTER
BUILDING
IKS | 1
CHEMICAL
*1 BUILDING
CHEMICAL
MIXERS
NEW INDUSTRIAL
PUMP STATION
MECHANICAL
FLOCCULATORS
r-
i
U_J
GRIT CHAMBERS |
._ i
' NEW i
|TRANSPORTER
(BUILDING i
- FLOW DIAGRAM - SEWAGE TREATMENT PLANT
SOUTH ST. PAUL, MINNESOTA
-------
floe and cause it to settle more rapidly. As described by tne patent
holders, the anionic polyelectrolyte is added with the bentonite and
the cationic polyelectrolyte is added with the ferric chloride and
iron ore. These two additions may be in series to one stream of wastewater
which are consequently mixed together.
Theoretically, the ferric chloride and cationic polyelectrolyte
would form a floe with the wastewater. The bentonite with a small
quantity of anionic polelectroljrte mixing in it would also form a
floe with the wastewater. These two streams of wastewater would be
combined and the two oppositely-charged floes would form a larger floe
which would immesh the iron ore, increasing the density of the floe,
causing it to settle. The patent also provided for variation in order
of addition of this multiple system of chemicals.
Tests were run for several months using this Blaisdell-Klass process.
The process was ineffective on the wastewater at the South St. Paul
plant, even though it had reportedly been effective at other
plants.
Several other systems were tiien. tried under the instruction and
supervision of personnel from the Robert S. Kerr Environmental Research
Laboratory at Ada, Oklahoma. Those systems tried were a single system
of chemicals using only an anionic polyelectrolyte, and a dual system
using ferric chloride followed by an anionic polyelectroyle.
The ferric chloride provided the trivalent ferric ion which was
adsorbed by colloidal particles, neutralizing the surface charge
and allowing them to coalesce into larger particles. Tills would
result in a floe too fine to settle out but which could be furtiier
coalesced into a larger heavier floe by the addition of an organic
polyelectrolyte.
-------
Two different polyelectrolytes were found effective in laboratory
tests and these were subsequently tested on a plant scale. A series
was also run using ferric chloride only. This was done to see how
great an effect the polyelectrolyte was having, and also to see how
much reduction could be obtained from the ferric chloride.
Tests were run on industrial packing house wastes, domestic waste,
and a combination of these two. The domestic waste contained stockyards
waste and packing plant wastes from two small slaughtering houses and
therefore, cannot be considered typical domestic wastes. These tests
using the dual system were run on a flow of wastewater which had heen
split between two primary sedimentation tanks. The flow to one tank
was treated with chemicals wui.le the second tank was used as a control.
Excess flow which was not treated was bypassed to a third sedimentation
tank. Finally, a test was run treating the whole plant flow with- that
system which was found to be most effective. This was done to determine
what the overall effect on the plant effluent would be after the treated
wastewater was passed through th.e trickling filters and anaerobic stabilization
pond. Since at the time this test was conducted the stabilization pond
had a seven day detention period, due to discontinuing of slaughtering at
Swift and Co. (otie of the two larger packing plants), it was decided
that a minimum test period of one month would be required to determine
what overall effect the addition of chemicals would have on the plant effluent.
This one month test was preceded by one month during which the plant was
operated in tne same manner that it would be during the test period except
flocculating chemicals were not added with the same parameters being examined
in the laboratory.
10
-------
SECTION IV
DESIGN AND CONSTRUCTION
The chemical mixing chambers were designed for addition of chemicals
either in series or in parallel. An arrangement of splitter boxes
was constructed ahead of the mixing chambers to allow the wastewater
being tested to be split, half being passed through mixing chambers
and a parshall flume meter into a settling tank, while the other
half was passed through a parsnall fume meter into a second settling
tank with no chemicals added. That portion of the wastewater to
which chemicals were added was first split between two small mixing
chambers and then recombined in one large mixing chamber before
it passed turough the parshall flume meter.
A new interceptor sewer was contructed as part of the plant expansion
which allowed meat packing plant industrial wastewater to enter
the plant separately from domestic wastewater. New grit chambers
and outside piping were constructed to allow this separation to
be maintained to the chemical mixing chambers. These grit chambers
are not conventional grit removal units, they more properly could be
described as preprimary and grit removal units since the detention
times varied from 15 to 60 minutes, depending on the rate of flow
and number of units in use. This allowed for testing of chemicals
on either industrial or domestic wastewater or a mixture of the two.
Two 122 cm (48") lines carried the wastewater from the grit chambers
to the chemical mixing chambers. At the influent to the chemical chambers
a series of gates were constructed to allow selection of flow from
either line and bypassing of flow from the other line to the third
settling tank. The gates allowed for equalization of flow between
the test and control tank.
Figure 2 shows the chemical mixing system. The plant flow diagram
before completion of the construction project is shown in Figures 3
and 4 and the flow diagram after completion of the new facilities is
shown in Figure 5.
IX
-------
DIVISION SYSTEM AND
FLASH MIXERS
WASTEWATER NOT TESTED
t-
cn
Ld
I. SMALL VARIABLE
SPEED MIXER
2. SMALL CONSTANT
SPEED MIXER
3. LARGE FLASH MIXER
4. SPLITTER DEVICES
5. PARSHALL FLUME
X
INDUSTRIAL LINE
X
DOMESTIC LINE
FI6URE-2
12
-------
Filters
"*
Waste Wash
Water And
Sewage Storag
a,
o
o
*-
CO
Q)
•*-
D
a ^
JC.
U)
O
Lagoon
Intermediate
Clarif iers
Floccu lotion
r
1
1
J
K~
LI —
i
i
i
l_.
Well 1
, _^
Pump
House
And
Control
Bldg.
c
<^
f&/_
\
CM
=:
1
^>
^
ro
=
5
a>
5
/
V
1
! 4
I1
I
1
1
1
.
1
-4-
| ,
,*
L
=
J
r~—
1
1
• n-
To Mississipj)i Rive
Final
Clarif iers
I
r
Sewage
Backwash
Lines Used For
Both Sewage
And Backwash
Sludge
FLOW DIAGRAM
Sewage Treatment
Plant
South St. Paul, Minn.
FIGURE- 3
13
-------
:-
J
-------
M / S 3 /3 -5 I
•• 'I
-------
The two small mixing chambers were equipped with wooden baffles which
forced the wastewater flow to enter at the bottom, flow through the
mixing chambers, and leave at the top. Each mixing chamber was
provided with a flash mixer unit. One of the small mixing chambers
was equipped with a constant speed mixer driven by a 1.12 Kw (1 1/2 hp)
motor. This was to be used for the addition of bentonite and the anionic
polyelectrolyte. The second small mixing chamber was equipped with a
variable speed mixer driven by a 2.24 Kw (3 hp) motor. It was initially
planned to add the catioiiic polyelectrolyte, ferric chloride, and iron
ore to this mixer. The variable speed provided for adjustment to the
speed necessary to keep tae iron ore in suspension. After the flow
of wastewater passed over the baffle and out of the two small mixing
chambers it passed into the large mixing chamber. This chamber was
equipped with a float and a mechanical liniarizer which indicated
flow as a percent of maximum flow. From the parshall flume the flow
passed into the test sedimentation tank. The half of the flow used
for a control passed from the splitter boxes through a flume leading
directly to a Parshall flume meter identical to the one following
the mixing chambers. From the meter it passed into the control sedimentation
tank.
In addition, a building was constructed to house the tanks used for
preparing chemical solutions, the proportioning pumps used to feed
the chemicals and provide storage space for chemicals. Four inixiiig
tanks of 1890 liter (500 gallon) capacity were installed, two for the cationic
combination of chemicals and two for the anionic combination. This
allowed continuation of chemical feed while a new tank of chemicals
was prepared. Each tank was equipped with a chemical mixer. The
tanks to be used for the cationic chemicals were also to have the
iron ore mixed in and were mixed with chemical mixers having 2.24 Kw
16
-------
(3 hp) motors. The tanks to be used for mixing the anionic chemicals
were equipped with chemical mixers having .75 kw (1 hp) motors.
A plastic impellor on the tanks mixing the cationic chemicals was
necessary because of the corrosiveness of ferric chloride. The
impellor and shaft was further covered with fiberglass to protect
the shaft and set screws from the strong ferric chloride solution. A
chemical proportioning puiap was provided for both the anionic and cationic
chemicals. These pumps were driven by a variable speed DC motor controlled
by a rheostat.
A small laboratory for on-the-spot tests was incorporated into the new
building. This was in addition to the existing, larger, well-equipped
laboratory in the main pumping station. Some additional facilties
were also built during th.e expansion which were not directly related
to the demonstration but aided in conducting experiments. These
included:
1. A new industrial pumping station equipped with automatic
bar screen and pumps of sufficient capacity to pump the wastewater
coining to the plant through th_e new industrial Interceptor.
2. A new ejector building to handle the paunch manure coming
into the plant and removed in the grit chambers (which are really more
of a preprimary tank than a grit chamber). Paunch manure is a
fibrous material which cannot be pumped and it is therefore necessary
to transport it to the sludge lagoons for disposal by an air ejection
system.
3. New chlorination equipment was Installed in the old ejector
building and the old grit cuambers were converted to chlorine contact
tanks.
17
-------
4. New distributor arms were installed on the trickling filters.
The original arms were quite old and of the two-arm type. These were
replaced by four-arm distributors to provide more even distribution
of water on the filters.
5. New pumps were installed in the main pumping station to provide
greater capacity to the filters and the pond, and to provide reciculation
of plant effluent onto th.e filters.
18
-------
SECTION V
EXPERIMENTAL PHASE
Shortly after the project got underway, laboratory evaluation of the
chemical treatment system in the Blaisdell and Klaas patents was
begun. Early in the project a cost limitation of about $2.60 per
million liter ($10 per million gallon) was set for chemical costs. Screenings
were made of a number of different polyelectrolytes used in the Blaisdell and
Klaas method. A number of combinations of chemicals which produced a fair
increase in settleable solids were uneconomical. Different types of
bentonite and iron ore were also tested to determine which was more
effective when used in the Blaisdell and Klaas system.
Laboratory evaluations were conducted with a six-gang floe
stirrer. One liter of waste was used with each, stirrer. The desired
chemicals or combination of chemicals were added to the beaker of
wastewater while it was stirred at 145 rpm. The stirring rate was
increased to 100 rpm for a 15 second period to insure complete mixing
and then reduced to 15 rpm for five minutes. Th.e waste was placed in
an Inihoff cone and allowed to settle for thirty minutes after whick
a sample was drawn off the top.
BOD and suspended solids were run on this sample.
Initially, a number of tests were run using visual inspection only to
determine the most effective combinations of chemicals. In these,
the polyelectrolytes recommended by the patent holders were used, but
variations were made in the concentration of the chemicals and the
order of addition. The polyelectrolytes recommended by the patent
holders were National Starch Floe Aid 1038 as a cationic polyelectrolyte
and Monsanto1s Stymer-S as an anionic polyelectrolyte. Flocculation
19
-------
aids were sodium bentonite and anhydrous ferric chloride. The weighting
agent was hematite iron ore tailings. Also, some inorganic chemicals
other than ferric chloride were tried. The results of these initial
tests are shown in Table 1.
At no time was a large heavy floe produced from any combination
of chemicals. However, some did show a fine floe and a marked increase
in settleable solids when allowed to settle in in an Imhoff cone. The
patent holders had suggested a combination of chemicals as follows:
The bentonite was to be added at a dosage of 3, to 6 mg/1 and the
anionic polyelectrolyte at 1/2000 the dosage of bentonite. The other
chemicals would be added together in a 1:2:3 ratio of cationic polyelectrolyte
to iron ore to ferric chloride, the total dosage of these cationic
chemicals was to be 3 to 6 mg/1. This combination of chemicals did not
prove particularly effective. It was necessary to greatly increase the
dosage to get an observable increase in settleable solids in the
Imhoff cones which resulted in too great a cost of chemicals. It was
found that by increasing the dosage of bentonite to 120 mg/1 and keeping
the rest of the chemicals at the lower concentrations, an observable
increase was made in settleable solids while a cost of $3.23 per million
liters ($12.20 per million gallons! was also obtained as indicated in
Table 2.
It must be concluded that the chemical dosages recommended by the
patent holders did not flocculate packing plant wastes in laboratory
tests. It is questionable if the mechanism by which BOD reduction occurred
at higher dosages of bentonite was the result of f,locculation or adsorption.
It was later discovered, however, that this higher degree of bentonite
was not effective in plant-scale tests. This was possibly due to the
higher concentration (8% suspension) at which the bentonite had to be
prepared when used in plant-scale quantities due to mechanical
20
-------
Table 1
Preliminary Work*
Laboratory Screenings
Settleable
Settleable
Anionic Cationic Solids
mg/1 mg/1 Blank ml/1**
3
12
1.5
60
60
30
120
120
60@1/40 Sty-S"
60O1/40 Sty-S
6
12
18
12
30
45
30
60
60
60
30
6
24
3
120
30
15
30
60
30
60
6
12
18
12*12 mg/1 lime
30
45
30*30 mg/1 lime
60
13.5
14.8
15.5
13.5
13.0
13.0
13.0
13.0
21.0
21.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
60 using PSI in- 5.0
stead of starch
60*100 mg/1 lime 5.0
30*25 mg/1 lime
5.0
Solids
Sample ml/1
13.0
14.2
15.0
40.2
16.0
15.0
20.0
23.0
28.0
30.0
5.0
5.0
6.5
6.5
5.5
7.5
7.5
10.0
36.0
43.0
7.0
$/MS
Cost
—
~_
__
—
--
—
—
--
—
...
--
—
-_
--
— -
_ _
—
$48 . 50
--
21
-------
Table 1 (continued)
Settleable Settleable
Anionic Cationic Solids Solids $/MG
tng/1 Blank ml/1** Sample ml/1 Cost
30 30*50 mg/1 lime 5.0 8.0 $28.12
30 30*75 mg/1 lime 5.0 10.5 35.75
30 30 7.2 9.5
30 30 PEI instead 7.2 15.0 57.60
of starch
30 30 FBI instead 7.2 18.0 65.30
of starch
+50 lime
30 30 PEI instead 7.2 26.0 166.60
of starch
+50 alum
30 30 FBI instead 7.2 12.0 133.20
of starch
+50 alum
30
30
30
60
90
120
150
90
90
30
60
5 PEI (alone) 7.2
5 PEI 7.2
+50 mg/1 lime
5 PEI (alone) 7.2
30 7.2
30 7.2
30 7.2
30 7.2
30 PSI-no starch 7.2
+50 mg/1 alum
30 PEI instead 7.2
of starch
30 19.0
30 FBI instead 19.0
9.5
10.0
16.0
13.0
16.5
30.0
46.0
47.0
38.0
25.0
29.0
—
120.70
26.70
29.20
31.70
34.20
205.80
62.60
—
__
of starch
+25 mg/1 alum
60 15 PEI instead 19.0 31.0 87.05
of starch
* 25 alum
-------
Table 1 (continued)
Anionic
mg/1
60
90
60
90
30
60
90
90
90
60
30
0
120
60
150
210
120
120
150
180
Settleable
Cationic
Solids
mg/1 Blank ml/1**
15 PEI-no starch
+50 mg/1 alum
15 PEI-no starch
*50 mg/1 alum
10 PEI-no starch
+25 mg/1 alum
10 PEI instead
of starch
+25 mg/1 alum
15 PEI instead
of starch
+25 mg/1 alum
15 PEI instead
of starch
*25 mg/1 alum
15 PEI instead
of starch
*25 mg/1 alum
15
15
15+15 mg/1 lime
15*30 mg/1 lime
15*15 mg/1 lime
15
15 +45 mg/1 lime
0
0
0
6
6
6
19.0
19.0
19.0
19.0
19.0
19.0
19.0
19.0
4.5
4.5
4.5
4.5
4.5
4.5
16,8
16.8
16.8
19.0
19.0
19.0
Settleable
Solids
Sample ml/1
36.0
43.0
30.0
36.0
25.0
31.0
38.0
27.5
6.1
5.8
5.3
6,2
10.2
6.2
35.0
57.0
28.0
29.0
43.0
52.0
$/MG
Cost
146.55
149.05
80.30
82.80
« M
70.35
72.85
--
18.35
18.16
17.97
17.78
20.85
22.78
12.50
17.50
10.00
14.34
16.84
19.34
23
-------
Table 1 (continued)
An ionic
mg/1
180
180
210
150
120
90
150
120
30
Cationic
mg/1
3
0
3
3
3
3
1.5
1.5
30 +50 mg/1
Settleable
Solids
Blank ml/1**
19.0
19.0
19.0
19.0
19.*
19.0
19.0
19.0
alum 8.5
Settleable
Solids
Sample ml/1
50.0
50.0
63.0
37.5
24.0
25.5
40,0
31.0
26.0
$/M3
Cost
17.17
15.00
19.67
14.67
12.17
9.67
13.59
11.09
—
Lime - Ca(OH)2 A.R.
P.E.I. - Polyethylene Imine
Starch - National Starch Floe Aid 1038
Alum - Al(S0> A.R.
* All anionic was bentonite with 1/2000 Stymer-S, and all cationic
was at the ratio of 1:2:3 National Starch (Hot Water), Hematite, FeClj.
** By blank is meant sample stirred and settled the same time as the
test sample but without addition of chemicals.
" Stymer-S
24
-------
Table 2
Blaisdell and Klaas Method
(Average of 5 trials each)
Sample No.
1
2
3
4
5
6
7
8
BOD
An ionic
Blank**
Control***
60
120
120
150
120
150
S9=!325
Cationic
--
—
3
3
0
3
3*10 alum
0
S,=l258
Cost/MG
—
—
7.17
12.17
10.00
14.67
16.13
12.50
to 4=2. 59
Set*
Solids
—
14.5
15,0
26.1
28.0
37.4
30.8
31.8
Po ,=0-05
Sus*
Solids
1138
508
510
394
--
336
—
—
Percent
Reduction
__~
55.2
55.1
65.3
--
70.4
--
--
BOD
1930
1326
1178
967
—
812
—
—
Percent
Reduction
30.9
39.1
49.3
58.0
* Settleable - Suspended Solids
** Blank - Sample As Taken
***Control - Stirred & Settled
-------
limitations of the mixing equipment. In this higher concentration
the particles of bentonite were not as well dispersed in the water.
Also, the floe formed in the laboratory was a fine floe and was settled
under static conditions in the sedimentation tanks. The dosage of
chemicals arrived at for further polymer screenings and initial
plant-sclae tests was 120 mg/1 of bentonite, 120/2000 mg/1 of
Stymer-S, 1.5 mg/1 of ferric chloride, 1.0 mg/1 of iron ore, and
.5 mg/1 of National Starch Floe Aid 1038. CA summary of the screening
results are shown in Table 2). A BOD was run on some samples of
wastewater treated with the above concentrations of chemicals and a
BOD reduction of 350 mg/1 was obtained.
Later when more extensive laboratory testing was done, it was
found the same results in BOD reduction could he obtained if polyelectrolytes
were left out and the sample of wastewater was treated just with,
bentonite, ferric chloride and iron ore. This was concluded from
an average of ten trails on which- suspended solids and BOD*s
were run. See Table 3. Increasing the. concentration of the
polyelectrolytes did not reduce BOD more than when bentonite was
used with no other additives. At 180 mg/1 of bentonite, however,
a greater BOD reduction was observed. See Table 4. A number of
laboratory tests indicated this BOD reduction to be in the order
of 250 mg/1. Technical personnel of the American Colloid Company
indicated that the actual mechanism by which this reduction occuppied
was one of adsorption of dissolved material on the bentonite with.
subsequent flocculation of the bentonite by ferric chloride or other
ions.
A number of polyelectrolytes were screened. A total of 50
polyelectrolytes from 16 manufacturers were checked. No great
difference in effectiveness was found in any of these polyelectrolytes
at the concentration tested. However, some did appear to be
slightly more effective than others. Of these, 12 were selected
26
-------
Table 3
Polyelectrolytes Giving Best Results In Lab Tests
Blaisdell And Klaas Process
Anionic Polyelectrolytes - Average Of Ten Trials
Polyelec troly te
Unsettled Sample
Blank Sample
Blaisdell & Klaas
Bentonite & Ferric
Chloride with Iron Ore
NS
Rohm & Haas A10
Nalco's 675
Goodyear's 708
Unsettled Sample
Blank Sample
Blaisdell & Klaas
Gen Mills Genfloc 263
Gen Mills Gendrive 162 14.20
Nalco's 0-2073
Garratt-Callahan's #78
Cost
$/MG
--
--
12.20
11.25
13.14
--
13.21
Cationic
--
--
12.20
--
14.20
__
Sus. Solids
ng/1 °>
796
309
232
240
238
260
265
Polyelectrolytes
943
352
295
253
226
264
292
t Red.
-
61.2
70.8
69.4
70.1
67.3
66.7
- Average
-
62.3
68.6
73.2
76.0
72.0
69.0
Total Solids
ng/1
2837
2318
2173
2158
2189
2180
2194
Of Ten Trials
2939
2307
2156
2134
2109
2133
2145
% Red.
18.2
23.4
24.0
23.1
22.8
23.1
32.6
-
21.4
26.7
27.4
28.2
27.4
27.0
BOD
mg/1
1440
940
682
698
728
735
705
1562
1079
809
772
758
787
768
% Red.
-
34.6
52.7
51.5
49.4
49.0
51.0
-
30.9
48.3
50.6
51.6
49.7
50.9
-------
Table 4
Effects of Variations of Polyelectrolytes
and Bentonite in Laboratory Tests
Sample BOD % Reduction
Untreated 1923
Control 1286 33.0
Blaisdell & Klaas2 962 50.0
Hercules 220 at .5 mg/13 913 52.5
Hercules 220 at 1.0 mg/13 914 52.4
National Starch 1063 at 0.5 mg/13 912 52.6
Bentonite only at 180 mg/1* 745 61.1
1. Stirred with no chemicals
2. 120 mg/1 bentonite, 120/2000 mg/1 Stymer-S, 0.5 mg/1 starch 1038,
1.0 mg/1 iron ore, 1.5 mg/1 ferric chloride
3* Same as 2 except for polyelectrolyte at indicated dosage
4. Bentonite at 180 mg/1, no other additives
28
-------
for more extensive testing. The results of these tests are indicated
in Table 5 and-Figure 6. Those polyelectrolytes which it was felt
might be tested on a plant-scale are indicated with an asterisk.
The results indicated are an average of 8 to 10 trails for each
polyelectrolyte. A control sample and a sample using the polyelectrolytes
recommended by the patent holders were carried along with each- trial.
A number of tests were also run on bentonite to determine whether
calcium or sodium bentonite would be most effective and to
determine the best way to add it. The patent holders indicated that
sodium bentonite was the most effective. A sample of calcium hentonite
(Panther Creek bentonite) was obtained and tests were made using 60 and
120 mg/1 of calcium bentonite and 60 and 120 mg/1 of sodium bentonite.
The results are indicated in Table 6. Tlie calcium bentonite showed
very little increased efficiency in BOD reduction while the sodium
bentonite showed the same effect obtained in previous tests.
A form of sodium bentonite with, the brand name of KWK bentonite
was also tested. This bentonite has larger particles of about 40
mesh and makes preparation of suspensions less difficult. The
bentonite used formerly was a powder of 200 mesh- and when placed in
water has a tendency to "ball up" and not disperse and produces
dust when handled in the large quantities needed on a plant scale.
The results of these tests indicate no difference in BOD reduction
between KWK bentonite and 200 mesh bentonite. The results are shown
in Table 7.
Tests were also run to determine if bentonite prepared in a more
concentrated suspension as necessary on a plant-scale woud change
its effectiveness. Laboratory scale tests had been run using a .3%
suspension. This was increased to 6.0% This increase in concentration
of the feed solution did greatly reduce its efficiency as indicated in
Table 8.
29
-------
Table 5
Polyelectrolytes Given More Extensive Lab Testing
(Average Of 8 To 10 Trials)
Polyelectrolyte Dosage - 0.5 mg/1
Inc. Treatment
Sample
Control ftl
Blaisdell & Klaas ffl
*Rohm & Haas A-10
*Nalco 675
*Goodyear's 708
Control #2
Blaisdell & Klaas #2
*A-1004 National
Starch 3286
Control #3
Blaisdell & Klaas #3
"General Mills 263
*General Mills 162
*Nalco's D-2073
*Garratt-Callahan #78
Control #4
Blaisdell & Klaas #4
*Hercules 220
BOD
mg/1
940
682
728
735
705
967
712
721
1079
809
772
758
787
768
1286
962
913
Percent Reduction
From Untreated
34,6
52.7
49.4
49.0
51.0
38.5
54.8
54.2
30.9
48.3
50.6
51.6
49.7
50.9
33.0
50.0
52.5
Percentage Point;
Above Control
—
18.1
14.8
14.4
16.4
—
16.3
15.7
—
—
19.7
20.7
18.8
20.0
._
17.0
19.5
30
-------
Table 5 (Continued)
Inc. Treatment
Sample
Control #5
Blaisdell & Klaas if 5
Nalco's 607
Nalco's 610
Control #6
Blaisdell & Klaas #6
*Tylite #9
Control Average
Blaisdell & Klaas Ave.
BOD
mg/1
935
675
693
680
927
670
662
979
751
Percent Reduction
From Untreated
42.6
58.6
57.5
58.2
41.2
57.5
58.0
36.8
53.6
Percentage points
Above Control
—
16.0
14.9
15.6
—
16.3
16.8
16.8
*Potential samples for plant scale test
31
-------
%BOD REDUCTION BY SAMPLE - % BOD REDUCTION CONTROL
-J
CD °
| *
uj !io
^ m°
cr .5
0 ^uj
o> b >
r^°
... ^m
_J w
o°*-
U LU-7
O
O
Q.
-J O^
UJ ^O
LU
O
cr
UJ
Q.
o
Q.
c;
w .
ui 2 m
o^ ^z
, £io=>
l- 2 "*
oc< Po
SWHH
S35o
S2S
1
1
1
1
1
1
1
1
1
i
I
1
1
I
1
1.
|
C
1
Q
o:1
.
< UI
< 1
P 1
o 1
1
TYL1T
NALCC
NALCC
HERCL
GARRA
NALCC
GENEf
GENEf
NATIOI
GOOD>
I NALCO
| ROHM/
BLAIS
E 9
610
S 607
LES 220
TT-CALLA
D- 2073
AL MILLS
AL MILLS
JAL STARC
EAR 708
675
>ND-HAAS
>ELL ft KL
HAN 78
162
263
H 3286
A- 10
*AS
in O in o in <
CM '
-------
Table 6
Effectiveness Of Sodium & Calcium Bentonite
In BOD Removal
Sodium Bentonite
Calcium Bentonite
LO
u»
Trial
1
2
3
4
Average
Reduction
Untreated
BOD mg/1
—
1475
1609
1540
1541
*• —
Blank*
BOD mg/1
1015
960
1275
1095
1086
29.4%
60 mg/1
BOD mg/1
870
935
1070
1095
992
35.7%
120 tng/1
BOD mg/1
790
760
960
845
839
46.6%
60 mg/1
BOD mg/1
900
990
1260
1045
1049
31.9%
120 mg/1
BOD mg/1
920
985
1095
1075
1019
33.9%
Sodium Bentonite vs. Calcium Bentonite at 120 mg/1
t= 8.78
* Stirred and settled, no chemicals added
-------
Table 7
Effectiveness Of Different Forms Of
Sodium Bentonite In Lab Tests
Bentonite 200
In Water
Bentonite KWK
In Water
Bentonite KWK
Added Dry
BOD mg/1
2002
843
896
1062
% Reduction
57.9
55.5
46.8
Sample
Untreated
0.3% Bentonite
Distilled Water
0.3% Bentonite
City Water
6.0% Bentonite
City Water
Table 8
Effect Of Concentration
Of Bentonite Suspension
BOD mg/1
1867
931
945
1116
%_Reduction
49.9
49.4
40.1
34
-------
A series of laboratory tests were run on several samples of iron ore
to determine what effect the weighting agent would have and which type
of iron ore would be most effective. When only 1 mg/1 of iron ore was
added to 1 liter of tap water, no perceptable amount could be seen, tt
was necessary to increase the dosage to above 10 mg/1 before the ore
was observable.
Tests were run on two grades of hematite (non-magnetic) and two grades
of magnetite. The magnetite was of a fine mesh, resulting from the taconite
process of refining iron ore. Both magnetite samples were run in magnetized
form and demagnetized form.
Samples were run using the chemical dosages (see page20^ used to
screen polyelectrolytes with the only variable being the iron ore.
The dynamic settling technique was used. Samples were drawn off
the surface of the one liter beaker while it was being stirred at
a low rpm in the jar test equipment. Thirty mg/1 of iron ore were
added to give a more pronounced effect.
Results of the test are shown in Table 9. Hematite No. 1 was slightly
coarser and of more uniform size than hematite No. 2. In both, 80%
would be retained by a 150 mesh screen. Both samples of magnetite
were more finely divided than the hematite. Magnetite No. 1 44% would
pass a 325 mesh screen and Magnetite No. 2 90% would pass a 325 mesh
screen.
Results of this single run were somewhat inconclusive. The difference-
in percent reduction observed was not large enough to be considered
meaningful.
35
-------
Table 9
Weighting Agent Laboratory Tests
Suspended
Sample Solids mg/1 % Reduction
Raw sewage used for testing purposes 750
Control - treated with flocculant 306 59.1
but no iron ore
All samples treated with flocculant 262 65.1
plus iron ore
Hematite
No. 1 only 275 63.3
No. 2 only 215 71.3
Average of No.'s 1 and 2 2U5 67.3
Magnetite
No. 1, magnetized and demagnetized 255 66.0
No. 2, magnetized and demagnetized 296 60.5
Samples 1 and 2, magnetized only 267 6U.4
Samples 1 and 2, demagnetized only 271 63.9
All magnetized samples 270 64.0
36
-------
SECTION VI
PLANT-SCALE EVALUATION OF PATENTED PROCESS
In October of 1968, pilot plant scale tests were started on the Blaisdell
and Klaas process, but operating difficulties resulting from the
corrosiveness of the ferric chloride solution delayed the actual
testing until November 1968.
These tests were run using concentrations of chemicals proven successful
in laboratory tests. The polyelectrolytes used were National Starch
Floe Aid 1038 on the cationic side and Monsanto's Stymer-S on the anionic
side. Testing was run on two shifts starting at 8:00 a.m. in the morning
and ending at midnight. These tests were run for a two week period
during which seven days of data were collected. The chemical dosages
used were 120 mg/1 of bentonite and 0.06 mg/1 of Stymer-S on the anionic
side. The chemical dosage of the cationic side was 0.5 mg/1 of 1038 starch,
1.0 mg/1 of iron ore and 1.5 mg/1 of ferric chloride. This was the
1:2:3 ratio of chemicals recommended hy the patent holders. The hentonite
suspension was prepared at 8% solids. The starch 1038 was mixed first
in hot water in small drums and then mixed with ferric chloride and iron
ore in the 500 gallon container. The concentration of the resulting
mixture of cationic chemicals was 2%. The cationic chemicals and
anionic chemicals were added separately to two parallel streams of wastewater
in the two small flash mixing chambers and these were then combined in
the large flash mixing chamber.
Samples were taken at the influent to the mixing chambers before
any chemicals were added and at the effluent to the test and control
tank. These samples were taken hourly and an 18 hour composite was
made on the basis of flow. Hourly settleable solids were run on grab
samples but all other tests were run on the composite sample. The
37
-------
primary parameter being tested was BOD. Suspended solids analysis
was also run on the composite sample. The results of the BOD and
suspended solids test are given in Table 10. Results of a "Student" t-
test on the data indicates significant reduction was not accomplished
in either BOD or suspended solids during the seven days of testing.
Since previous laboratory tests had indicated the effectiveness of
bentonite was dependent on a dilute concentration of tbje material
being added, lower concentrations of bentonite were tried for several
days. The screen testing period was limited to two days adding
bentonite at a concentration of 2.7%. The results of these screen
tests are shown in Table 11. Even at this reduced concentration of
the bentonite no meaningful reduction in BOD was observed by the
addition of chemicals. This method was successful in the laboratory,
but it had no effect when used in a plant scale process.
ADDITIONAL PATENTED MODIFICATION EVALUATIONS
After this process failed, one of the patent holders suggested a
number of variations in the patented process to effect a ROD reduction
in the wastewater by means of chemical additions. Over a siJMnonth.
period, from December 1968 to June 1969, 23 variations were tried.
No. 1
Starting the end of December 1968, and going into January 1969 a fiye-^day
series of tests were run increasing the dosage of cationic chemicals
and decreasing the dosage of anionic chemicals. The cationic and
anionic chemicals were both, added at 50 mg/1 of the total mixture on
38
-------
Table 10
Plant Scale Tests
Chemical Dosage Anionic 120 tng/1 Cationic 3 mg/1
Bentonite Prepared At 8%
Reduction Of BOD mg/1
Date
Nov., 1968
4
5
6
8
12
13
14
Average
Influent
1050
1230
1620
1015
1425
1450
1485
1325
Effluent
Test Tank
930
800
1005
820
865
930
920
896
Percent
Reduction
11.3
35.0
38.0
19.2
39.3
35.9
38.0
32.4
t= 1.55 p= .17
Reduction Of Suspended Solids
Date
Nov. , 1968
4
5
6
12
13
14
Average
Influent
450
900
930
610
570
810
711
t= 1.87
Effluent
Test Tank
340
370
400
220
270
210
301
p= .12
Percent
Reduction
—
~
—
—
—
—
57.7
Effluent
Control Tank
935
800
1070
800
870
970
940
912
mg/1
Effluent
Control Tank
290
290
150
290
150
160
221
Percent
Reduction
10.9
35.0
33.9
21.2
38.9
33.1
36.7
31.2
Percent
Reduction
—
—
—
—
--
—
69.0
39
-------
Table 11
Plant Scale Tests
Chemical Dosage Anionic 120 mg/1, Cationic 3 tng/1
Bentonite Prepared at 4% Solids
Reduction of BOD mg/1
Effluent %, Effluent %
Date Influent Test Tank Reduction Control Tank Reduction
Nov. 1968
25 1100 815 .. 800
26 1500 950 — 915
AVERAGE 1300 880 32.3 855 34.2
Bentonite Prepared at 2.7% Solids
Reduction of BOD mg/1
Effluent 7o Effluent %
Date Influent Test Tank Reduction Control Tank Reduction
Nov. 1968
28 1780 1020 — 1085
29 1445 885 ~ 910
AVERAGE 1610 950 41.0 1000 38.0
40
-------
each side. Bentonite (KWK) was added at 50 mg/1 with Stymer-S
added at .025 mg/1. Starch was added at 8.3 mg/1, iron ore at
16.7 mg/1, and ferric chloride at 25 mg/1. The starch used in this and
subsequent.tests was National Starch Floe Aid 1063. This starch
did not require as high a temperature in its preparation as the Floe
Aid 1038. The ferric chloride and iron ore were mixed in one tank
and the starch in the second cationic tank. Separate pumps were
obtained to pump the chemicals from the tanks. To avoid possible
degradation of the starch by the acidic ferric chloride solution. The
ferric chloride tank was refilled while chemicals were being fed and
the starch pump was shut down daily for 20 minutes to allow refilling.
The results are shown in Table 12. Significant effects could not
be observed by the addition of chemicals.
No. 2
A four-day test was run with chemical dosage at 30 mg/1 each for
cationic and anionic chemicals. The specific dosages were: 30 rag/1
bentonite (KWK), .015 mg/1 Stymer-S, 5 mg/1 Floe Aid 1063, 10 mg/1
iron ore, and 15 mg/1 ferric chloride. The results are shown in
Table 13. Again, floe was not visible nor was any significant reduction
in BOD and suspended solids achieved.
No. 3
Two days of testing were conducted adding 15 mg/1 each of cationic and
anionic chemicals. The specific dosages of chemicals were in the same
ratio as in the previous three tests. Again, no meaningful reduction
was achieved. The results are given in Table 14.
41
-------
Table 12
Plant Scale Tests
No. 1
Chemical Dosage 50 mg/1 Cationic 50 rag/1 Anionic
Reduction of BOD mg/1
Date
Dec.
Jan.
Jan.
Jan.
*Jan.
Influent
27,
2,
3,
7,
15,
1968
1969
1969
1969
1969
Average
1440
1360
1595
1710
1950
1611
t=
Test Tank
%
Eff. Reduction
1250
1020
1090
1143
1470
1195
.153
13.
25.
31.
32.
24.
25.
1
0
6
9
5
7
P=
Control Tank
Eff.
1240
990
1310
1398
1120
1212
.88
7
10
Reduction
13
27
17
18
42
24
.8
.0
.9
.3
.2
.7
Chemical Dosage 50 mg/1 Cationic 50 mg/1 Anionic
Reduction of Suspended Solids mg/1
Date
Dec.
Jan.
Jan.
Jan.
*Jan.
27,
2,
3>
7,
15,
1968
1969
1969
1969
1969
Average
Influent
740
670
650
1010
1040
822
t=
Test Tank
Eff.
590
300
290
460
;soo
428
.466
7.
Reduction
20.
55.
55.
54.
48.
47.
1
2
4
5
1
8
P=
Control Tank
Eff.
510
240
390
610
210
392
.77
%
Reduction
31
64
40
39
79
52
.0
.1
.0
.6
.8
.3
* A new ferric chloride pump was installed.
42
-------
Table 13
Plant Scale Tests No. 2
Chemical Dosage 30 rag/1 Cationic 30 mg/1 Anionic
Reduction of BOD mg/1
Test Tank 7, Control Tank
Influent Eff. Reduction Eff. Reduction
Jan. 16,
Jan. 17,
Jan. 21,
Jan. 22,
Average
Date
Jan. 16,
Jan. 17,
Jan. 21,
Jan. 22,
Average
1969 1330
1969 1345
1969
1969 1265
1313
t=
Reduction
Influent
1969 740
1969 630
1969 570
1969 470
602
930
925
--
855
903
2.65
of Suspended
Test Tank
Eff.
470
400
180
160
302
30.0
31.4
--
32.5
31.3
Solids
970
1015
--
880
955
p= .13
mg/1
7. Control Tank
Reduction Eff.
36.5
36.5
68.5
66.0
49.8
390
450
200
210
312
27.1
24.5
--
30.5
27.1
%
Reduction
47.4
28.5
64.9
55.3
48.1
t= .324 p= .77
43
-------
Table 14
Plant Scale Tests No. 3
Chemical Dosage 15 mg/1 Cationic 15 mg/1 Anionic
Reduction of BOD mgA
Test Tank % Control Tank %
Date Influent Eff. Reduction Eff. Reduction
Average
1545
1255
1410
1135
930
1032
26.5
26.0
26.6
1005
905
955
35.0
27.9
32.3
Reduction of Suspended Solids mg/1
Test Tank 70 Control Tank %
Date Influent Eff. Reduction Eff. Reduction
Jan. 23, 1969
Jan. 24, 1969
Average
760
560
660
400
140
270
47.4
75.0
59.1
380
170
275
50.0
69.6
58.3
44
-------
No. 4
Two days of tests were run adding concentrations of 15 mg/1 cationic
chemicals and 45 mg/1 of anionic chemicals. The specific doses were:
45 mg/1 bentonite (KWK), .022 mg/1 Stymer-S, 2.5 mg/1 Floe Aid 1063;
5.0 mg/1 iron ore; and 7.5 mg/1 ferric chloride. The results of this
trial are shown in Table 15.
A substantial reduction was made in suspended solids on the first
day, but not on the second. A dense floe was not produced and the
BOD reduction was slight. Further tests at this concentration were
not made.
No. 5
In February 1969, a different grade of hentonite was tried. A 200
mesh bentonite was used instead of KWK 40 mesh bentonite. The doses were;
20 mg/1 bentonite, .015 mg/1 Stymer-S, 5 mg/1 Floe Aid 1063 and 15 mg/1
ferric chloride. BOD reduction was not obtained. The results are
given in Table 16.
No. 6
Some short tests adding high dosages of chemicals to the wastew_ater
indicated a fine floe could be formed at dosages of 50 to 100 mg/1
of ferric chloride. Four days of testing were done with the dosage
of 50 mg/1 ferric chloride. The dosages of the other chemicals added
were: 30 mg/1, Bentonite (KWK); .015 mg/1, Stymer-S; 5 mg/1, Floe
Aid 1063, and 10 mg/1 iron ore. The results are shown in Tahle
17. Again, significant BOD reduction was not achieved.
45
-------
Table 15
Plant Scale Tests No. 4
Chemical Dosage 15 mg/1 Cationic 45 n*g/l Anionic
Reduction of BOP mg/1
Test Tank % Control Tank %
Bate Influent Eff. Reduction Bf £. Reduction
Jan. 31, 1969
Average
1260
995
1128
885
800
842
29.8
19.5
24.6
890
835
862
29.5
16.0
23.4
Reducjtiqn of Suspended Solids rag/I
Test Tank % Control Tank
Date Influent Eff. Reduction Eff. Reduction
Jan. 27, 1969
Jan. 31, 1969
Average
450
420
435
170
210
180
62.2
50.0
58.5
380
230
305
16,5
45,3
29.8
46
-------
Table 16
Plant Scale Tests
No. 5
200 mesh bentonite
Starch and ferric chloride in separate tanks
Chemical Dosage 30 mg/1 Cationic 30 mg/1 Anionic
Reduction Of BOD mg/1
Date
Feb. 5, 1969
Feb. 6, 1969
Average
Percent Reduction
Influent
1090
1370
1230
—
Table 17
Test Tank
Effluent
865
920
892
27.5
Control Tank
Effluent
900
895
898
27.0
Plant Scale Tests
No. 6
Starch separate from ferric chloride and iron ore
Chemical Dosage: Bentonite at 30 mg/1, Ferric Chloride
Iron ore at 10 mg/1, Starch at 5 mg/1, Stymer-S at .015
Date
Feb. 4, 1969
Feb. 7, 1969
Feb. 12, 1969
Feb. 13, 1969
Average
Percent Reduction
Reduction Of BOD
Influent
1155
1435
1325
1470
1346
— —
mg/1
Test Tank
Effluent
915
1035
1123
895
992
26.4
at 50 mg/1,
mg/1
Control Tank
Effluent
930
1010
1042
900
970
27.9
t= .972 p= .40
47
-------
No. 7
Two days of testing were run with the dosages of chemicals used in
test No. 6, but with the flash mixers shut off. It was thought
the flash mixers might be shearing the floe and it was not reforming.
The results are shown in Table 18. Slightly greater BOD reduction
was achieved in the test tank as compared to the control tank, but the
difference was not great enough to justify the use of chemicals.
No. 8
A series of six days of testing were run with the flow of wastewater
reduced to give 1 1/2 hour detention in the primary sedimentation tank.
The chemical dosages were the same as in tests No. 6 and No. 7. The
results are shown in Table 19. A significantly greater BOD reduction
was achieved in the test tank than in the control tank. This could be
experienced since this high dosage of ferric chloride formed a fine
floe which would settle out given a long detention time hut would not
settle out at the normal rate of flow through the tank. Since the
object was to form a heavy floe which would Incorporate the weighting
agent and then settle out in a short detention tijae, further variations
in the patented process were continued.
The patent holders recommended a change in the cationic polyelectrolyte
to achieve a heavy floe. Polyethylene imlne OfEI) was used instead
of the modified starch which had been used. It was the opinion
of the patent holders that the ?EZ while considerably more expensive
was much more effective in the formation of a floe.
No. 9
Two days of tests were run using the same dosages of chemicals used
in tests No. 6, 7, and 8, hut substituting the £EH for Floe Aid 1Q63.
No substanial floe formation was observed other than that formed by the
48
-------
Table 18
Plant Scale Tests
No. 7
Flash mixers off, starch separate from ore and ferric chloride
Chemical Dosage same as Table 17
Reduction Of BOD mg/1
Date
Feb. 14, 1969
Feb. 17, 1969
Average
Percent Reduction
Plant Scale Tests
No. 8
Detention time 1% hours, starch separated from ferric chloride and
iron ore.
Chemical Dosage: Iron ore at 10 mg/1, Bentonite at 30 mg/1,
Ferric Chloride at 50 mg/1, Starch at 5 mg/1, Stymer-S at .015 mg/1
Reduction Of BOD mg/1
Influent
1585
1540
1562
—
Table 19
Test Tank
Effluent
1000
800
900
42.5
Control Tank
Effluent
1050
850
1000
36.0
Date
Feb. 18,
Feb. 19,
Feb. 21,
Feb. 24,
Feb. 25,
Feb. 26,
Average
1969
1969
1969
1969
1969
1969
Influent
1125
1255
1690
1020
1235
1140
1244
Test Tank
Effluent
595
635
675
688
655
633
647
Control Tank
BE fluent
712
755
810
726
830
700
755
f\ *
t= 5.425
49
-------
50 mg/1 of ferric chloride. On the first day the detention period
in the test and control settling tanks was increased to 1 1/2 hours.
A BOD reduction due to chemicals was observed which was about the
same as that observed when 50 mg/1 of ferric chloride had been added
in combination with the Floe Aid 1063 and other chemicals being
used. The results are shown in Table 20.
No. 10
The patent holders suggested a more effective means of incorporating
the iron ore into the floe would be to mix it with the polyelectrolyte
rather than with the ferric chloride as had been done in previous
tests. A four day test was run mixing the ferric chloride in one
tank and the Floe Aid 1063 in the other tank. The dosage of
chemicals were: 30 mg/1 Bentonite,0.015 mg/1 Stymer-S,5 mg/1 Floe
Aid 1063, 10 mg/1 iron ore, and 50 mg/1 ferric chloride. The
results of these tests did not indicate any significant BOD removed
by the chemicals nor was any heavy floe formed. Only the fine floe,
resulting from the 50 mg/1 of ferric chloride was produced. The
results are indicated In Table 21.
No. 11
Two days of testing were done following the same procedure, hut
using PEI instead of Floe Aid 1063. The iron ore was mixed in
with the PEI and the ferri.c chloride was mixed separately. The
same concentration of chemicals was used. Again, no appreciable
floe was formed and no substantial reduction in BOD or suspended
solids was obtained. The data is presented in Table 22.
50
-------
Table 20
Plant Scale Tests
No. 9
Polyethylene imine mixed separately from the ferric chloride and
iron ore.
Chemical Dosage: Bentonite at 30 mg/1, Iron ore at 10 mg/1,
Ferric Chloride at 50 mg/1, Polyethylene imine at 5 mg/1, Stymer-S at
.015 mg/1
Reduction Of BOD mg/1
Date
Feb. 20, 1969
Percent Reduction
February 27, 1969
Percent Reduction
Normal Flow
Influent
1120
Reduced Flow
1270
Table 21
Test Tank
Effluent
770
31.4
710
44.2
Control Tank
Effluent
840
25.0
825
35.1
Plant Scale Tests
No. 10
Iron ore mixed with starch; ferric chloride mixed separately
Chemical Dosage: Bentonite at 30 mg/1, Iron ore at 10 mg/1,
Ferric Chloride at 50 mg/1, Starch at 5 mg/1, Stymer-S at .015 mg/1
Reduction Of BOD mg/1
Date
March 3, 1969
March 4, 1969
March 5, 1969
March 6, 1969
Average
Percent Reduction
Influent
1260
1540
1830
1450
1520
Test Tank
Effluent
830
910
845
1010
899
40.9
Control Tank
Effluent
790
920
920
955
896
41.0
t= .085
p >.90
51
-------
Table 22
Plant Scale Tests
No. 11
P.S.I, and iron ore mixed, Ferric chloride mixed separately,
mechanical flccculators on
Chemical Dosage: Bentonite at 30 mg/1, Iron ore at 10 tng/1,
Ferric Chloride at 50 mg/1, P.E.I. at 5 mg/1, Stymer-S at 0.015 mg/1
Reduction Of BOD mg/1
Date
March 17, 1969
March 19, 1969
Average
Percent Reduction
Date
March 17, 1969
March 19, 1969
Average
Percent Reduction
Influent
1370
1950
1660
.on
Reduction Of Suspended
Influent
530
850
690
on
Test Tank
Effluent
685
1170
927
44.2
Solids mg/1
Test Tank
Effluent
140
320
230
65.7
Control Tank
Effluent
845
1210
1027
38.1
Control Tank
Effluent
150
290
220
68.1
52
-------
No. 12
It was suggested by the patent holders that perhaps packinghouse grease
was being emulsified in the flash mixers, thereby creating a large surface
upon which the chemicals were absorbed, reducing their effectiveness.
To test the idea, a single day's run was made with the flash mixers
and mechanical flocculators shut off. The iron ore was mixed with the
starch and the ferric chloride was mixed separately. The dosages
were the same as had been used in previous tests. The results are given
in Table 23. For this one day's run, there appeared to be some reduction
in BOD and suspended solids.
No. 13
The mechanical flocculators were possibly shearing the floe and
reducing its size. Five days of tests were conducted with the
mechanical flocculators shut off, but with the flash mixers on.
The specific dosage of chemicals used were the same as had been used
in previous tests: 30 rag/1, bentonite; O.Q15 mg/1, Stymer-S;
5 mg/1 Floe Aid 1063; 10 mg/1 Iron ore and 50 ,mg/l. ferric chloride.
The iron ore was again mixed with the atarch. and the ferric chloride
mixed separately. The mechanical flocculators were not operated.
The results are given in Table 24. Again, some reduction appeared
in BOD and suspended solids, however substantial floe formation
was not visible. The t-test for significance of difference betwen
two means was run on data of both the suspended solids and BOD
tests. The results indicated no significant difference in suspended
solids between the test and control tank. The BOD did show a difference
between the test and control tank which was significant. Howeyer, the
efficiency of the BOD reduction in this tes.t was not as great as was
hoped for.
53
-------
Table 23
Plant Scale Tests
No. 12
Flash mixers and flocculators off.
Chemical Dosage: Bentonite at 30 mg/1, Iron ore at 10 mg/1,
Ferric Chloride at 50 mg/1, Starch at 5 mg/1, Stymer-S at 0.015 mg/1
Reduction Of BOD mg/1
Date
March 21, 1969
Percent Reduction
Influent
1272
Test Tank
Effluent
675
47.0
Control Tank
Effluent
830
34.9
Date
March 21, 1969
Percent Reduction
Reduction Of Suspended Solids mg/1
Influent
Test Tank
Effluent
820
90
89.0
Control Tank
Effluent
150
81.7
54
-------
Table 24
Plant Scale Tests
No. 13
Mechanical flocculators off.
Chemical Dosage: Bentonite at 30 mg/1, Iron ore at 10
Ferric Chloride at 50 tag/1, Starch at 5 mg/1, Stymer-S at 0.015 mg/1
Reduction Of BOD mg/1
Date
March 24, 1969
March 25, 1969
March 26, 1969
March 27, 1969
March 28, 1969
Average
Percent Reduction
Influent
1040
1350
1243
1577
1235
1289
t= 3.93
Reduction Of Suspended
Date
March 24, 1969
March 25, 1969
March 26, 1969
March 27, 1969
March 28, 1969
Average
Percent Reduction
Influent
730
570
710
380
500
578
^ —
Test Tank
Effluent
650
810
698
857
762
755
41.4
Solids mg/1
Test Tank
Effluent
260
150
110
130
170
164
71.6
Control Tank
Effluent
735
1005
1022
1262
907
986
23.2
p= .03
Control Tank
Effluent
210
200
720*
220
280
227
60.7
*Excluded from average
t= 1.47
p= .24
55
-------
No. 14
Several days of tests were conducted at a reduced dosage of 30 mg/1
ferric chloride. All other chemical dosages and procedures were
kept the same. The mechanical flocculators were left off. The results
of these tests are shown in Table 25. Again, large floe was not obtained
and the BOD reduction obtained by use of chemicals was not significant
as indicated by the results of the "Student" t-test run on data in Table
25.
No. 15
A one day test was run without any anionic chemicals and adding the
cationic chemicals in the following dosages: 5 mg/1, Floe Aid 1Q63,
10 mg/1, iron ore, and 15 mg/1 ferric chloride. The mechanical flocculators
were left off. Large floe was not formed. BOD reduction was achieved
as indicated in Table 26, but not enough, to consider the systems
effective.
No. 16
A Five-day test was run using the same cationic dosages but adding
8 mg/1 of bentonite along with it. Large floe was not formed and the
results did not indicate a significant reduction in BOD. The results
are shown in Table 27.
No. 17
The chemical feed was further reduced. The dosages added w,ere:
4 mg/1, Bentonite, .002 mg/1, Stymer*-S, 3 mg/1, Floe Aid 1063,
6 mg/1 iron ore, and 9 mg/1 ferric chlorde. The mechanical flocculators
were left off. Again, floe was not formed. A slight increase in BOD reduction
56
-------
Table 25
Plant Scale Tests
No. 14
Mechanical flocculators off
Chemical Dosage: Bentonite at 30 mg/1, Iron ore at 10 mg/1,
Ferric Chloride at 30 mg/1, Starch at 5 mg/1, Stymer-S at 0.015 mg/1
Reduction Of BOD mg/1
Date
March 31, 1969
April 1, 1969
April 2, 1969
Average
Percent Reduction
Date
March 31, 1969
April 1, 1969
April 2, 1969
Average
Percent Reduction
Influent
1247
1260
1562
1356
.on
t= 1.087
Reduction Of Suspended
Influent
550
550
940
680
.on
Test Tank
Effluent
682
993
760
812
39.3
p= .4
Solids tng/1
Tes t Tank
Effluent
80
150
240
157
76.9
Control Tank
affluent
845
945
845
878
35.2
0
Control Tank
Effluent
380
260
300
313
54.0
t= 2.148
p= .17
57
-------
Table 26
Plant Scale Tests
No. 15
Flocculators off
Chemical Dosage: Ferric chloride at 15 mg/1, Starch at 5 mg/1,
Iron ore at 10 mg/1, Bentonite - none, Stymer-S - none
Reduction Of BOD mg/1
Date
April 3, 1969
Percent Reduction
Influent
1080
Test Tank
Effluent
610
Reduction Of Suspended Solids mg/1
April 3, 1969 510 180
Table 27
Control Tank
Effluent
807
25.3
270
Plant Scale Tests
No. 16
Chemical Dosage: Ferric chloride at 15 mg/1, Iron ore at 10 mg/1
Starch at 5 mg/1, Bentonite at 8 mg/1, Stymer-S at 8/2000 tng/1
Reduction Of BOD mg/1
Date
April 4, 1969*
April 7, 1969
April 8, 1969
April 9, 1969
April 10, 1969
April 11, 1969
Average
Percent Reduction
t= 1.906
*Good Friday - low flow, excluded from average
Influent
748
1155
1275
1375
1060
940
1157
Test Tank
Effluent
446
780
717
717
765
607
717
37.9
Control Tank
Effluent
521
975
704
745
803
702
786
32.0
p= 0.14
58
-------
was achieved as indicated in Table 28. A statistical t-test did not
indicate this difference to be significant at the 0.05 level. The
actual improvement was 110 mg/1.
It appeared that slight BOD reduction was being achieved, but was
not considered enough to justify the use of chemicals. The fact
that some BOD reduction was now being achieved while none had been
achieved in earlier tests was attributed to mechanical flocculators,
which were not now being operated, shearing the floe.
No. 18
A three day series of tests were run with the flash mixers and
the mechanical flocculators shut off. The chemical dosages were;
8 mg/1, bentonite, 5 mg/1, Floe Aid 1063, 10 mg/1, iron ore, and
15 mg/1 ferric chloride. Again floe was not formed. The results
indicated no difference in BOD or suspended solids reduction (Table
29).
No. 19
Anionic and cationic chemical addition was changed from parallel to
series. A concentration of 37.5 mg/1 ferric chloride was added to the
variable speed flash mixer along with .022 mg/1 Stymer-S. National
Starch Floe Aid 1063 was added at a point just beyond the large flash
mixer. (Figure 2). Turbulence in the flume leading to the settling
tank plus the mechanical flocculators was sufficient to provide mixing
of the starch with the wastewater. The results are shown in Table 30.
Reduction of BOD and suspended solids were obtained, but not enough- to
justify continuing the test. A large or dense floe was not formed.
59
-------
Table 28
Plant Scale Tests
No. 17
Flocculators off
Chemical Dosage: Ferric chloride at 9 mg/1, Iron ore at 6 mg/1,
Starch at 3 mg/1, Bentonite at 4 mg/1, Stymer-S at 4/2000 mg/1
Reduction Of BOD mg/1
Date
April 28, 1969
April 29, 1969
April 30, 1969
May 1, 1969
May 2, 1969
Average
Percent Reduction
t= 1.921 p=
Influent
1240
1805
1315
1447
1215
1404
Test Tank
Effluent
735
800
922
770
802
845
39.9
Control Tank
Effluent
1130
940
998
960
750
955
32.1
60
-------
Table 29
Plant Scale Tests
No. 18
Flash mixers off, flocculators off
Chemical Dosage: Ferric chloride at 15 rag/1, Iron ore at 10 mg/1,
Starch at 5 rag/1 , Bentonite at 8 mg/1, Stytner-S at 8/2000 mg/1*
Reduction Of BOD mg/1
Date
May 5, 1969
May 6, 1969
May 7, 1969
Average
Percent Reduction
Influent
1250
1305
767
1107
Test Tank
Effluent
775
950
457
727
34.3
Control Tank
Effluent
890
830
478
733
33.7
t= .218
p= .85
Reduction Of Suspended Solids mg/1
Date
May 5, 1969
May 6, 1969
May 7, 1969
Average
Percent Reduction
Influent
Test Tank
Effluent
700 240
670 230
470 90
613 187
69.5
t= .138 p->.90
* 8 parts of Stymer-S mixed with 2000 parts of Bentonite
Control Tank
Effluent
270
180
100
187
69.5
61
-------
Table 30
Plant Scale Tests
No. 19
Starch into parshall flume following parallel addition of bentonite
and ferric chloride
Chemical Dosage: Ferric chloride at 37.5 mg/1, Iron ore - none,
Starch at 3 mg/1, Bentonite at 4 mg/1, Stymer-S at 4/2000 mg/1
Reduction Of BOD mg/1
Test Tank Control Tank
Date Influent Effluent Effluent
May 13, 1969 1430 580 820 (1 grab sanple)
May 14, 1969 1040 917 982 (3 samples comp)
Average 1223 748 901
Percent Reduction — 39.0 26.4
Reduction Of Suspended Solids mg/1
Date
May 13, 1969
May 14, 1969
Average
Percent Reduction
Influent
900
260*
580
__
Test Tank
Effluent
160
300
230
60.2
Control Tai
Effluent
300
310
305
47.4
* Probably atypical results
62
-------
No. 20
Two days of tests were run using the same procedure but at reduced rate
of chemicals. The dosages were: 9mg/l, ferric chloride, no iron ore,
3 mg/1 starch, 4 ing/1 bentonite and .002 mg/1 Stymer-S
No. 21
Two days were run adding iron ore mixed with 6 mg/1 ferric chloride.
but with no bentonite and Stymer-S.
No. 22
A final two days of tests were run with all the chemicals added. The
results are shown in Tables 31-33. No effective floe formation was
achieved and the BOD reduction was very small.
No. 23
The patent holder suggested the order of addition of chemicals bus
changed. Bentonite and .002 mg/1 Stymer-S were added to both, small
flash mixers in a split stream. Ferric chloride and iron ore mixed
together were added to the large flash, mixer and Floe Ai.d 1063 was
added to the flume just beyond the large flash mixer. These tests were
run for four days. The results are indicated in Table 34. Floe production
was not effective and significant reduction of suspended solids or BOD was
not achieved. Very little success was obtained from any of the variations
tried. At no time was a dense floe formed during any of these tests.
By June 1969 six months of testing of the Blaisdell and Klaas patented
procedure for wastewater flocculations and sedimentation had been run
without success. Of all procedures tried at the South St. Paul treatment
plant none achieved a sufficient reduction in BOD and suspended solids.
63
-------
Table 31
Plant Scale Tests
No. 20
Starch into parshall flume following parallel addition of bentonite
and ferric chloride
Chemical Dosage: Ferric chloride at 9 mg/1, Iron ore - none,
Starch at 3 mg/1, Bentonite at 4 mg/1, Stymer-S at 4/2000 mg/1
Reduction Of BOD mg/1
Date
May 15, 1969
May 13, 1969
Average
Percent Reduction
Influent
835
1345
1090
«•••
Test Tank
Effluent
650
535
592
45.8
Control Tai
Effluent
712
585
648
U0.6
Reduction Of Suspended Solids mg/1
Date
May 15, 1969
May 16, 1969
Average
Influent
430
730
580
Test Tank
Effluent
110
150
130
Control Ta
Effluent
250
180
215
Percent Reduction
77.6
62.9
64
-------
Table 32
Plant Scale Tests
No. 21
FeCl3 and iron ore mixed; starch into the parshall flume following
parallel addition of bentonite and ferric chloride
Reduction Of BOD mg/1
Date
May 19, 1969
May 20, 1969
Average
Percent Reduction
Influent
930
1115
1022
_••
Test Tank
affluent
625
640
632
38.2
Control Tai
Effluent
650
798
724
29.2
Reduction Of Suspended Solids mg/1
Date
May 19, 1969
May 20, 1969
Average
Percent Reduction
Influent
500
570
535
^ —
Tes t Tank
Effluent
210
430
320
40.2
Control Ta
Effluent
200
280
240
45.2
65
-------
Table 33
Plant Scale Tests
No. 22
Iron ore and ferric chloride mixed together. Starch added to the
parshall flume following parallel addition of bentonite and
ferric chloride
Chemical Dosage: Ferric chloride at 9 tng/l» Iron ore at 6 tng/1,
Starch at 3 mg/1, Bentonite at 4 mg/1, Stymer-S at 4/2000 mg/1
Reduction Of BOD mg/1
Date
May 21, 1969
May 22, 1969
Average
Percent Reduction
Date
May 21, 1969
May 22, 1969
Average
Percent Reduction
Influent
1010
1050
1030
,on
Reduction Of Suspended
Influent
430
550
490
on
Test Tank
Effluent
685
680
682
33.8
Solids mg/1
Test Tank
Effluent
140
190
165
66.3
Control Tank
Effluent
610
760
685
33.5
Control Tank
Effluent
190
220
205
58.2
66
-------
Table 34
Plant Scale Tests
No. 23
Bentonite added to the two small flash mixers. Ferric chloride and
iron ore premixed and added to the large flash mixer. Starch added
to the parshall flume.
Chemical Dosage: Ferric chloride at 9 mg/1, Iron ore at 6 mg/1,
Starch at 3 mg/1, Bentonite at 4 mg/1, Stymer-S V2000 mg/1
Reduction Of BOD mg/1
Date
May 26, 1969
May 27, 1969
May 28, 1969
May 29, 1969
Average
Percent Reduction
Influent
1322
1517
1040
1045
1181
_-
Test Tank
Effluent
915
872
689
805
820
30.7
Control Tai
Effluent
922
867
700
785
818
30.6
t= .250
p= .85
Reduction Of Suspended Solids mg/1
Date
May 26, 1969
May 27, 1969
May 28, 1969
May 29, 1969
Average
Percent Reduction
Influent
690
570
570
490
580
• M
Test Tank
Effluent
160
280
230
180
212
63.5
Control Tai
Effluent
220
280
260
170
232
60.0
t= 1.27
p= .39
67
-------
There was no success in producing a floe beyond what could be expected
from the higher doses of ferric chloride added during some of the
tests. During this six month period one of the patent holders, a
practicing engineer, had acted as a consultant and visited the
plant on an average of once a week. In addition, field representatives
from the companies whose products were being used in the tests had
visited the plant providing advice on methods they had found successful
at plants treating wastewater other than packinghouse wastes. Success
was not had in flocculating the combined mixture of domestic, stockyards,
and packing house wastes. .
A total of 11 of the 23 tests were run for a long enough time to provide
sufficient data for statistical analysis of difference in the means of
the test and control effluents by the "Student'1 t-test. The othej:
12 tests were terminated with three or less days run because a dense
floe was not being formed or because the reduction of BOD and suspended
solids was not considered enough to justify the use of chemical treatment.
Only 2 of the 11 showed a significant difference in BOD reduction between
the test and control tanks. These were tests No, 8 and No. 13. The, results
of test No. 8 were obtained by increasing the dosage of ferric chloride
to produce a fine floe and increasing the detention time in the test and
control tank to 1 1/2 hours. At the normal wastewater flow through the three
primary tanks of 1 1/2 hours detention time could not be achieved and the BOD
reduction was not obtained without it as Indicated in test No. 6. Analysis
for suspended solids was not run on test No. 8.
The second test in which a significant BOD reduction was obtained was
test No. 13. The reduction obtained in suspended solids data was not
68
-------
significant. This test was run with tne mechanical flocculators shut
off. The chemical dosage was the same as had been used in test No. 8.
The cost of chemicals was $11.69 per million liter ($44.20 per million
gallons). If costs for labor, utilities, capital construction and
maintenance were included, the total cost of chemical treatment would
be about $15.00 per million liter ($58.00 per million gallons). This
cost was too high to be practical considering the amount of BOD reduction
obtained. A number of additional tests were made with the mechanical
flocculators off and reduced chemical dosage, but a significant difference
in BOD and suspended solids was not obtained.
69
-------
SECTION VII
LABORATORY EVALUATION OF NON-PATENTED SYSTEMS
Advice was sought of experienced persons in the use of organic and inorganic
flocculating agents in wastewater, and who did not have a proprietary interest
in the success of the patented process being treated or in any chemical
flocculating agent or combination thereof.
Personnel from EPA laboratories were asked to provide assistance in an
effort to determine if successful treatment could be achieved by th_e use
of chemical additions. EPA personnel came to the South St. Paul plant site
to review previous grant reports, make on-site inspection of plant facilities,
and considerable laboratory work was performed in screening of polyelectrolytes
or systems of polyelectrolytes to determine what would be effective
in flocculation of the combined South. St. Paul wastewater. In
reviewing the results of over 20 different test conditions of the patented
process which had been tested at the plant, no large difference was
found between test condltons and control results for most tests. The
differences shown were in the same range as the differences between the
test and control tank without chemical additives.
A malfunction of the flow meter in the test channel was discovered and
corrected.
Several variations of the patented process were tested and found to
be inactive with the combined wastewater at the plant. As a result
of this, it was recommended that evaluation of the patented chemical
process be discontinued until the patent holders had found and demonstrated
active mixtures in the laboratory.
70
-------
A number of single and dual systems of chemical flocculating
agents were tested. In addition, lab tests were run for flocculation
consistency for those chemical systems which produced a satisfactory
floe.
A single system is the most practical from the standpoint of
handling and feeding chemicals. The chemicals tested singly are
listed in Table 35. In summary of this work, extremely high dosages
of the cationics, Calgon ST-260 and Hercules 220 were effective
in flocculation. The cost of treating wastewater at these dosages
would have been economically prohibitive.
The anionic material found to be effective in single systems were
Dow A-21 and Nalco 610. The dosage of these chemicals was in a
range which was economically feasible. Costs of the single systems
tested are given in Table 36.
In addition to organic flocculants tested, ferric chloride was
tested and found active at an optimum dosage of 100 mg/1. However,
even at this dosage the flocculation was not rapid and the ferric
hydroxide floes were voluminous in nature which would indicate
poor settling under dynamic conditions.
Several cation-anion dual flocculant systems were attempted. A
list of these are shown in Table 37. The purpose of this was to see
if two chemicals could be used to flocculate the waste at a reduced
cost. The only dual systems found which produced a floe large enough
for sedimentation in the overloaded primary sedimentation tanks was
ferric chloride with Nalco 675, ferric chlorie with Nalco 610, and
ferric chloride with Dow A-23.
71
-------
Chemical
Table 35
Single Flocculant Evaluation
Optimum Or
Maximum Dosage
(mg/1)
1. Calgon ST-260
2. Calgon 269
3. Nalco 607
4. Nalco 675
5. Nalco 610
6. Nalco 603
7. Hercules 220
8. Dow A-21
9. Dow A-22
10. Dow A-23
11. Dow C-31
12. Dow C-32
13. Dow SA 118.1A
14. Dow SA 11881.F
15. Dow SA 1569
16. Dow SA 1767
17. Dow 1621.2
18. Dow N-ll
19. Dow N-12
20. Dow N-17
21. Feds (anhydrous)
20
2
20
2
20
40
20
4
2
2
50
50
50
50
50
40
40
2
2
2
100
Results
Medium, voluminous floe formed
Poor flocculation
Poor flocculation
Poor flocculation
Medium flow with some clarity
No activity
Large, dense floe formed
Large, dense floe formed
No activity
No activity
No activity
No activity
No activity
No activity
Ho activity
No activity
No activity
No activity
No activity
No activity
Medium voluminous floe with
excellent clarity
72
-------
Table 35
(Continued)
Optimum Or
Maximum Dosage
Chemical (mg/1) Results
22. General Mills* 162 35 No activity
23. General Mills 263 35 No activity
24. Polyethylene inline 50 Fine voluminous floe formed
* General Mills Inc., Kankakee, Illinois
Table 36
Single Flocculant Summary
Price Optimum Dosage Treatment Cost
Chemical ($/lb.) (mg/1) ($/MG)
2. Calgon ST-260
3. Hercules 220
4. Dow A-21
5. Nalco 610
* Later determined to be 20 and 410
0.055
1.75
1.75
1.00
2.45
100
20
20
4
2*
M4MMMMM
46
290
290
33
41*
73
-------
Table 37
Dual Flocculant Evaluation
Chemical #1
1. Fed3 (Anhydrous)
2. FeCl3 (Anhydrous)
3. FeCl3 (Anhydrous)
ft. Calgon 260
5. Hercules 220
6. Calgon 260
7. Hercules 220
8. FeCl3 (Anhydrous)
9. FeClj (Anhydrous)
10. FeCl3 (Anhydrous)
11. FeCl3 (Anhydrous)
12. FeClj (Anhydrous)
13. Dow A-21
Chemical #2
Dow A-21
Nalco 610*
Nalco 675
Dow A-21
Dow A-21
Nalco 610
Nalco 610
Iron Ore
Calgon 260
Reten 220
A-22
A-23
Iron Ore
Results
No synergism
No synergism*
Synergism suspected
No synergism
No synergism
No synergism
No synergism
No synergism
No synergism
No synergism
No synergism
Synergism suspected
No synergism
* Synergism was suspected until an error of concentration
was discovered.
74
-------
Tests were then made of the consistency of these systems with
changes of waste during the day and night. Along with these tests
several suspended solids determinations were made in order to obtain
some estimates of suspended solids removals and to be able to compare
the systems under further investigation. Suspended solids were
taken from jar tests by withdrawing the liquid from 2 inches below
the surface at a paddle speed of 10 rpm.
Table 38 illustrates the consistency of these tests with the time
of day. Nalco 610's effectiveness and 1 and 2 mg/1 in the single and
dual system was due to a mistaking of concentration by the manufacturer.
When the concentration was found to b.e too high-by a factor of 10, testing
of Nalco 610 was discontinued because of high- costs. Three systems
which remained technically and economically feasible were:
1. Dow A-21 alone,
2. A dual system of Nalco 675 and ferric chloride, and
3. A dual system of Dow A—23 and ferric chloride.
The suspended solids results in Table 38 confirm the previous
visual observations. The results also confirm the voluminous
nature of the ferric hydroxide floe and its resistance to sedimentation
under dynamic conditions. When the ferric chloride treated waste was
allowed to settle under static conditions, it showed the test
overall results.
The results of the laboratory evaluations of the chemicals for
treatment of combined waste are:
1. Dow A-21 was the most active, economical, and consistent
of single chemical systems tested at dosages between 2 and 5
mg/1.
75
-------
ON
Table 38
Laboratory Results On Consistency Tests
Suspended Solids mg/1
Tests
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Day
••••Ma*
6/19/69
6/19/69
6/19/69
6/19/69
6/19/69
6/19/69
6/19/69
6/19/69
6/1 9/69
6/20/69
6/20/69
6/20/69
Time
6:00 a.m.
6:00 a.m.
6:00 a.m.
9:00 a.m.
9:00 a.m.
9:00 a.m.
9:00 a.m.
9:00 a.m.
9:00 a.m.
9:30 a.m.
9:30 a.m.
9:30 a.m.
Chemical
Dow A-21
FeCl3*/Nalco 610
FeCl3*/Nalco 610
Dow A-21
Dow A-21
Dow A-21
FeCl3*/Nalco 610
FeCl3*/Nalco 610
FeCl3*
Dow A-21
Nalco 610
FeClo*/Naloo 610
Dosage
(mg/n
2.0
10/10**
20/10**
3.0
4.0
5/0
40/10**
50/10**
50
4.0
2.5
50/1
Degritted
Raw
192
192
192
1111
1111
1111
1111
1111
1111
804
804
804
Control
142
164
164
769
769
769
824
824
824
704
692
728
Treated
53
47
49
560
311
204
144
172
1051
268
824
884
* Anhydrous
** Thought to be 1 mg/1 at time of investigation
-------
Table 38 (Continued)
Suspended Solids mg/1
Tests
13.
14.
15.
16.
17.
18.
19.
20.
21.
Day
^••(•^••B
6/20/69
6/20/69
6/20/69
6/20/69
6/20/69
6/20/69
6/20/69
6/20/69
6/20/69
Time
10:40 a.m.
10:40 a.m.
10:40 a.m.
10:40 a.m.
12:40 p.m.
12:40 p.m.
12:40 p.m.
12:40 p.m.
12:40 p.m.
Chemical
Dow A-21
FeCl3*/Dow A-21
FeCl3*/Dow A-21
FeCl3*
Dow A-21
Dow A-21
FeCl3*/Nalco 610
FeCl3*/Nalco 610
FeCl */Nalco 610
Dosage
(mg/1)
4.0
40/3
40/4
100
4.0
5.0
40/1
40/5
40/10
Degritted
Raw
1040
1040
1040
1040
1200
1200
1200
1200
1200
Control
956
976
976
976
1136
1136
1064
1064
1064
Treated
292
1028
580
1276
312
252
1220
696
140
* Anhydrous
** Thought to be 1 mg/1 at time of investigation
-------
2. There was a definite indication of synergism and Increased
performance over a single system in both dual systems, i.e.,
ferric chloride plus Nalco 675 and ferric chloride plus Dow A-23.
3. These systems consistently flocculated during waste changes
with the time of day.
Some additional laboratory tests were run on order of addition of
ferric chloride and the poly electrolyte. The two polyelectrolytes
(Dow A-23 and Nalco 675) which had indicated activity in the previous
tests were used. Tests were made on the six place gang stirrer.
Samples were run as follows.
1. A control sample which was stirred, but to which nothing
was added,
2. A sample to which ferric chloride was added first followed
after a five-minute interval by the polyelectrolyte.
3. A sample to which ferric chloride was added followed at a
ten-minute interval hy a polyelectrolyte,
4. A sample to which the polyelectrolyte was added first followed
at a five-minute interval hy ferric chloride.
5. A sample to which the polyelectrolyte was added first followed
at a ten-minute interval by ferric chloride.
6. A sample to which the ferric chloride and polyelectrolyte
were added at the same time.
78
-------
Samples were also taken about 5 centimeters (2 inches) below the surface
of the wastewater while it was being stirred at 10 rpm. Suspended solids
determinations were made for each sample and also for the sample of
the degritted waste. In all cases, 50 mg/1 ferric chloride and 1.0
mg/1 polyelectrolyte were used.
Some general observations were made for both of the polyelectrolytes.
In those samples to which ferric chloride was added first a fine floe
formed after a short time. When the polyelectrolyte was added a dense
heavy floe slowly formed. This floe did not disappear on further
stirring and began to settle as soon as stirring was slowed. There
was not observable difference between samples having five and ten
minute delays before addition of polyelectrolytes. In those samples
to which the polyelectrolyte was added first there was no observable
effect due to the additon of the polyelectrolyte. When ferric chloride
was added last the floe built more slowly than where it was added first.
Also, the floe did not become nearly as large or dense. In some trials
a denser floe was formed in the sample haying a ten minute waiting
period before addition of ferric chJ.ori.de, but this was not consistently
true. In the sample to which ferric chloride and the polyelectrolyte
were added at the same time a fairly large floe formed almost at once.
When the chemicals were added at the same time the floe was long and
filamentous. It continued to build up for a short time after the
addition of chemicals and then started to break up. In some trials
it broke up to the point where it was not raucli_ heavier than a ferric
chloride floe. Other times a heavier floe would remain, but not as
heavy as in the case where ferric chloride was added first followed by
the polyelectroyte. The results of the tests run are shown in Tables
39 and 40. These tests indicate the best order of addition of the
dual system was ferric chloride first followed after a 5 to 10 minute
period by the polyelectrolyte.
79
-------
00
o
Table 39
Order Of Addition Tests
A-23 - FeCl3
Suspended Solids tng/1
Trial
1
2
3
4
5
6
7
8
9
10
Ave.
Raw
Degritted
1268
1124
892
844
664
660
1108
1136
1976
1428
1110
Control
808
600
740
604
556
508
760
720
1348
996
764
FeCl3
Then A-23
At 5 Min.
68
116
120
120
128
148
132
132
540
328
183
Then A-23
At 10 Min.
88
112
116
128
196
116
296
164
244
252
171
A-23
Then FeCl3
At 5 Min.
200
212
220
212
216
320
420
456
400
468
312
A-23
Then Fed 3
At 10 Min.
168
196
296
196
228
236
256
272
540
356
274
A-23
and Fed 3
Same Time
284
176
256
400
292
436
632
544
672
532
422
-------
00
Table 40
Order Of Addition Tests
675 -
Suspended Solids mg/1
Trial
1
2
3
4
5
6
7
8
9
10
Degritted
792
544
940
824
572
372
1104
1976
1164
1276
Control
448
302
564
568
548
216
468
808
1200
596
Fed 3
Then 675
At 5 Min.
80
68
60
56
96
52
72
128
108
120
Then 675
At 10 Min.
60
108
88
88
192
60
52
72
92
104
675
Then FeCl3
At 5 Min.
148
172
176
308
324
176
184
388
316
292
675
Then Fed 3
At 10 Min.
152
196
164
228
204
124
204
196
216
108
675
and Fed 3
Same Time
172
256
188
264
264
156
256
468
376
492
Ave. 956 581 84 92 248
159 289
-------
Some laboratory tests were run using a waste ferric chloride material
obtained from the Buckbee Hears Manufacturing Company of St. Paul,
Minnesota. The results of these tests indicated that the waste ferric chloride
material was slightly more active than the bulk (anhydrous) ferric
chloride. On further investigation a considerable price difference
was found between the waste ferric chloride and the bulk ferric
chloride with the bulk ferric chloride being several times more
costly than waste ferric chloride. The anhydrous ferric chloride was
an industrial grade purchased from a local chemical company in 61 kilogram
(135 pound) drums.
Ferric chloride waste solution was the byproduct of the Buckbee
Mears Manufacturing Company. The ferric chloride is used during th£
manufacture of steel sheets to make television screens, electric shaver
heads, steel sieves, and other items where small uniform high--density
holes in steel plate are needed.
The ferric chloride is regenerated from time to time by the. addition
of hydrochloric acid and chlorine. Eventually some of It is wasted.
Specific gravity tests indicated a 45% concentration of ferric chloride
An analysis was performed by a consulting laboratory which verified
the ferric chloride content to be approximately 45% with small quantities
of other metals present as indicated in the analysis.
82
-------
SECTION VIII
PLANT EVALUATION OF NOH-PATEHTED SYSTEMS
Personnel from the Robert S. Kerr Environmental Research Laboratory
returned to the plant in July 1969 to set up plant-scale tests
to determine if sufficient suspended solids and BOD removal could be
obtained in the primary clarification portions of tue plant to warrant
more extensive study of the removals. If so, tuis would be
accomplisned in the primary portion of the plant and eventually more
extensive studies of overall efficiency could be accomplished tarough-
tiie entire plant.
The three flocculating systems which, were evaluated briefly on the
primary portion of the plant were:
1. Dow A-21 at 2 to 5 mg/1,
2. A dual system of 20 to 60 rag/1 of ferric chloride plus
.75 to 1.4 mg/1 of Halco 675, and
3. A dual system of 30 to 50 mg/1 of ferric chloride plus
.75 to 1.3 mg/1 of A-23.
CONTROL CHECK
Tue first day of the week was used to make polymer solutions, calibrate
pumps, and prepare otaer details necessary for chemical feed. This day
was therefore used for an additional control period to illustrate the
equality of the two primary tanks. Tue results of this control day are
illustrated in Table 41. Tue results of this control day indicated
no significant difference between the test and control tanks as Shown
by the results of tae t-test in Table 46.
83
-------
Table 41
Control Day - No Chemicals Added
July 14-15, 1969
Time
8 a.m.
9 a.m.
10 a.m.
11 a.m.
12 Noon
1 p.m.
2 p.m.
3 p.m.
5 p.m.
6 p.m.
7 p.m.
8 p .m.
10 p.m.
12 Mid.
2 a.m.
4 a.m.
6 a.m.
24 hour composite
Influent
604
1228
1236
884
1120
776
736
1364
728
608
740
536
288
336
252
176
244
744
Suspended Solids mg/1
Test
64
140
256
228
272
300
308
272
436
300
300
272
272
612
112
80
48
232
Control
68
196
232
260
276
296
-
272
320
404
280
264
156
408
88
76
76
224
84
-------
SINGLE SYSTEM
The single system of Dow A-21 was tested for 48 hours on July 15 and 16.
The results of tiiese tests (Table 42) indicated increase in removal of
suspended solids in test tank versus the control tank. The composite
sample confirmed tais showing, 208 mg/1 of suspended solids in the
control tank compared to 136 mg/1 in tiie test tank.
Tne effluent from tne test tank treated with Dow A-21 appeared extremely
turbid and visually appeared approximately the same as the effluent from the
control tank.
The major visual difference between the control and test tanks was the
massive floes formed in the test tank, BOD results were also obtained
on a few selected samples and tae data indicated very small differences
in BOD removals.
THEORY OF EFFECTIVENESS OF THE DUAL SYSTEM
Colloidal particles are kept in suspension in the wastewater because of
their small size. These particles are covered with a surface coarge
wuich causes them to repel each, otoer thereby preventing the forming of
larger particles which would settle out of suspension. When multi-
valent ferric ion is added in solution, it neutralizes tae surface charge
on the colloidal particles and allows tuem to merge. Mixing of taese small
particles then causes tae larger particles to build, as the natural
cohesive forces come into play, and results in coagulation of a floe.
The couesive forces holding the floe together are relatively weak and
shearing of the floe occurs before it can grow very large. Ta_e addition
of ferric chloride alone results in a fine floe which will not settle out
in the dynamic situation of a sedimentation tank but instead passes over tae
weir. The suspended solids test measures the amount of solids retained on
a filter when a sample of wastewater is passed through.. When a wastewater
85
-------
Table 42
Purifloc A-21 (Dow Chemical Company)
July 15-16, 1969
Time
8 a.m.
9 a.m.
10 a.m.
11 a.m.
12 Noon
1 p.m.
2 p.m.
3 p.m.
4 p.m.
5 p.m.
6 p»m.
7 p.m.
8 p.m.
10 p.m.
12 Mid.
2 a.m.
4 a.m.
6 a.m.
Flow Pu
(mgd) D
4.9
6.6
6.6
3.9
7.9
7.7
7.7
7.9
6.1
5.8
5.5
5.5
4.4
6.6
4.4
4.2
3.9
5.0
rifloc A-21
losage mg/1
2.25
2.60
2.85
3.30
4.30
5.00
4.20
4.00
5.00
4.00
2.40
4.00
3.20
3.00
2.30
2.00
2.00
2.00
Suspended Solids mg/1
Influent
548
832
1092
1180
1276
1172
1014
840
696
796
544
508
452
620
312
316
252
240
Test
96
180
172
212
212
184
196
260
232
248
224
232
172
128
76
76
72/80
54
Control
92
196
212
224
224
276
272
332
304
340
460
272
264
192
-
136
76
108
86
-------
Table 42 (Continued)
Purifloc A-21 (Dow Chemical Company)
July 16-17, 1969
Flow
Time (mgd)
8 a»tn« 6.1
9 a.m. 7.1
10 a.m. 7.2
11 a.m. 7.2
12 Noon 7.4
1 p.m. 7.4
2 p.m. 6.9
3 p.m. 7.4
6 p.m. 5.7
7p.m. 6.6
8 p.m. 6.1
10 p.m. 5.3
12 Midn. 4.2
2 a.m. 3.9
4 a.m. 3.7
6 a.m. 3.7
Mean
July 15-16, 1969
July 16-17, 1969
Composite
July 15-16, 1969
July 16-17, 1969
Purifloc A-2]
Dosage mg/1
3.00
2.10
3.70
4.00
3.80
4.00
3.20
4.00
4.00
3.00
3.60
3.00
4.50
4.50
5,00
5.00
L Suspended Solids mg/1
Influent
500
824
1108
1000
840
916
1004
1048
680
780
512
364
204
228
204
212
705
647
572
572
Test
100
132
180/184*
224
208
192/196*
208
200
212
196/200*
160
144
80
56
60/72*
40
168
150
156
136
Control
72
168
236/232*
236
240
248
256
284
248
224
264
104
120
116
88
96
234
176
244
208
* Duplicate analysis
87
-------
sample not treated with ferric culoride ±a passed through tne filter some of
the larger colloidal particles are entrained, but the smaller colloidal
particles pass through. If ferric chloride has been added to the wastewater
some smaller colloidal particles are coagulated and thus retained on the
filter in a suspended solids test. Since no additional solids have been
settled out in sedimentation tank, the amount of suspended solids are
actually increased because of the coagulated small colloidal particles.
Since there is no actual increase of organic solids, the BOD does not
substantially change.
When the organic poly electrolyte is added to the wastewater after the
addition of ferric chloride, the long molecules with their localized
charge hook into the fine ferric chloride floe causing bridging between
the fine particles. This results in the formation of a floe whichu is
large and dense enough to settle out and still strong enough,to resist
the suearing action which occurs in the wastewater. This is what makes
the dual systems tested effective. The addition of toe polyelectrolyte
alone is not effective since it does not have sufficient capacity to
neutralize all the surface charge on the realtively dense, fine
colloidal solids found in the stronger packinghouse waste. It is taerefore
necessary to precede the polyelectrolyte by a flocculating aid sucu
as ferric chloride in a dual system.
DUAL SYSTEM
On July 17 of the test week a dual system of ferric chloride and llalco
675 was attempted. It was felt that it would be desirable to add trie
polyelectrolyte at a point In the flow approximately 10 to IS minutes
after the addition of the ferric chloride. It was therefore necessary
to construct a distribution pipe to feed the polyelectrolyte approximately
1/3 to 1/2 the distance from the entrance into the flocculation tank. The
point of addition was determined by adding a dye tracer to the influent and
tracing the dry through the flocculation tank.
-------
The ferric chloride dosage was held at approximately 50 mg/1 from 8 All
to 8 PM and 30 mg/1 from 8 PM to SAM. The polyelectrolyte dosage was
held at 1.0 mg/1 from 8 AM to 8 PM and 0.75 mg/1 from 8 PM to 8 AM.
In comparing this dual system with tue single polymer system previously
used it appeared from the data that greater and more consistent removals
of solids and BOD were obtained witn tive dual system of ferric chloride
*
and Halco 675. The composite sample was reduced from 252 mg/1 of suspended
solids in the control tank effluent to 132 rag/1 in the test tank effluent.
In addition, the BOD in the composite effluent in the control tank was 830 mg/1
compared to 545 mg/1 in the test tank. The results of this day's run are
shown in Table 43. v
On July 18 the dual system of ferric chloride and Dow A-23 was run. The
dosage level of chemicals was maintained at the same rate as on the previous
day. The results of tnis day's test was shown in Table 44.
The composite samples indicated suspended solids reduction of from 2QQ mg/1
in the control tank to 96 mg/1 in the test tank with a corresponding BOD
reduction from 680 mg/1 in tue composite effluent of the control tank
to 96 mg/1 in the test tank with a corresponding BOD reduction from 680 mg/1
in the composite effluent of the control tank to 493 mg/1 in the
test tank.
A statistical evaluation of this one week run is given in Table 45. The
following conclusions were indicated:
1. The single polymer Dow A-21 achieved a significant difference in
removals of suspended solids in the primary sedimentation portion of the
plant. In comparision to tne dual systems attempted, the single polymer
system appeared inferior. Tine Dow Chemical Company announced that the
production of A-21 had been discontinued and there were no plans for
restarting production of this chemical.
89
-------
Table 43
FeCl3 And Nalcolyte 675 (Nalco Chemical Company)
July 17-18, 1969
Chemicals
VO
o
Time
8 a.m.
9 a.m.
10 a.m.
11 a.m.
12 Noon
1 p.m.
2 p.m.
3 p.m.
4 p.m.
5 p.m.
6 p .m.
7 p.m.
8 p.m.
Flow
(mgd)
•••MMMIIMIM
6.1
7.4
6.1
6.1
6.3
6.0
6.6
6.1
6.1
5.9
6.2
5.5
4.6
FeCl 3
mg/1
••Wft^W^
0
43
40
41
45
40
43
57
50
46
52
50
50
6"75
mg/1
0
.94
1.00
.95
.95
.90
.85
1.10
1.00
1.20
.74
1.40
1.20
Suspended Solids mg/1
Influent
536
992
1196
976
868
1044
1188
848
1116
784
620
1124
284
Test
72
132
120/108
148
176
140/104
152
284
228
256/228
172
176
120
Control
92
160
228/232
216
248
228
248
360
324
340
392
296
196
BOD mg/1
Influent Test
__
._
680
630
—
—
600
—
643
—
560
—
360
Control
--
--
800
715
—
--
770
--
740
—
805
—
470
-------
Time
9 p.m.
10 p.m.
12 Mid.
1 a.m.
6 a.m.
Mean
Table 43 (Continued)
Fed3 And Nalcolyte 675 (Nalco Chemical Company)
July 17-18, 1969
Flow
mgd
5.8
4.4
4.2
_
•**
Fed 3
mg/1
20
30
17
_~
-«
675
tagA
.75
.38
.57
-—
—
Suspended
Influent
324
280
312
224
204
718
Solids
Test
80
108
52
60
72
139
mg/1
Control
160
136
168/152
108
84
221
BOD mg/1
Influent Test Control
—
—
—
__ __ --
—
-------
Table 44
FeCl3 And Purifloc A-23 (Dow Chemical Company)
July 18-19, 1969
Chemicals
Time
8 a.m.
9 a.m.
10 a.m.
11 a.m.
12 Noon
1 p.m.
2 p.m.
3 p.m.
4 p.m.
5 p.m.
6 p .m.
7 p.m.
8 p.m.
9 p.m.
Flow
ragd
•••••MB
5.7
6.6
6.0
5.7
6.2
5.4
6.3
6.3
-
6.5
6.1
5.0
4.4
4.1
Fed 3
rag/1
IMHH^HHIHaB
44
35
29
57
49
42
50
59
-
45
47
27
30
0
A-23
mg/1
1.1
1.3
1.0
1.0
1.0
1.0
1.0
1.1
-
.8
.8
1.0
1.2
1.2
Suspended Solids
Influent
__
936
1008
852
892
572
1028
908
764
700
576
396
308
420
Test
__
116
160/140
140
116
172/152
108
92
220
256/252
200
120
72
68/64
mg/1
Control
__
268
208/200
220
224
264
—
332
396
284
504
268
248
172
BOD mg/1
Influent Test Control
-_ _- __
-_ -_ —
733 858
—
453 795
—
468 685
— — —
523 798
— — __
686 962
— — — — __
430 612
-
-------
VO
CO
Table 44 (Continued)
Fed3 And Purifloc A-23 (Dow Chemical Company)
July 18-19, 1969
Chemicals
Time
10 p.m.
11 p.m.
12 Mid.
1 a.m.
2 a.m.
3 a.m.
4 a.m.
5 a.m.
6 a.m.
7 a.m.
8 a.m.
Mean
2^ hour c
Flow
mgd
5.0
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
-
omposite
FeCl3
mg/1
36
26
33
13
20
20
20
20
16
27
20
—
A-23
mg/1
• •••H ••HI
.85
.85
.85
1.10
1.10
.8s
.68
.85
.85
1.10
—
—
__
Suspended
Influent
260
248
120
220
280
216
216
176
216
180
232
488
536
Solids
Test
104
72
52
48/48
36
44
40
40/48
44
36
32
99
96
Control
144
252
112
96
88
92
100
76
52
76
76
197
200
BOD mg/1
Influent Test Control
295 351
—
—
—
—
_-
—
—
__
—
—
—
1160 493 680
-------
vo
•IS
Table 45
Statistical Analysis Of One Week Run
Sus. Solids
Standard
Date
1969
July 14
July 15
July 16
July 17
July 18
fiomn.
Chemical
Dosage
No Treatment
2-5 mg/1 A-21
2-5 n«/l A-21
.75 - 1.3 rag/1 675
+20 - 57 rag/1 FeCl 3
.75 - 1.3 mg/1 A-23
+30 - 50 mg/1 FeCl3
Ave, Diff, Test
Less Control
18.25
-53.44
-37.12
-83.47
-102.83
-96.00
Deviation In
Ave. Diff.
73.32
61.49
35.95
47.72
75.82
20.65
t
0.996
-3.687
-4.131
-7.624
-6.644
-9.295
# Of Paired
Observations
17
17
17
16
16
4
*
**
***
***
***
***
* Indicates not significant
** Indicates p<.01
*** Indicates p<,001
-------
2. The dual system of ferric chloride and organic polyelectrolyte
tested significantly incresed removals of suspended solids and BOD in
primary sedimentation portions of tiie plant.
3. The two dual systems tested were not judged different in
removal efficiencies.
95
-------
SECTION IX
ADDITIONAL DISCUSSION OF NON-PATENTED SYSTEMS
ADDITIONAL PLANT TESTS
In the year following tuese initial tests a considerable amount of
additional plant scale testing was done.
Tiie year of testing was divided on December 1, 1969, because this was
when Swift & Company discontinued slaughtering operations at their
South St. Paul plant. This reduced tae packing plant wastes coming into the
plant by about half and considerably cuanged the characteristics of the
wastewater. Testing results, however, were similar although- not directly
comparable for the two parts of the year. In the part of this year before
Swift & Company discontinued slaughtering operations the following tests
were run:
1. A dual system: ferric cliloride Canhydrous) followed by Dow A-23
2. A dual system: ferric chloride waste material (Buckbee Hears,
See Appendix D) followed by Dow A-23
3. A dual system: ferric choride (anhydrous! follwed by Nalco 675
4. A dual system: ferric chloride waste material (Buckbee
followed by Nalco 675
5. Variations in the ferric chloride dosage in an attempt to
optimize the best dosage which, could be achieved.
6 . The use of a ferric cliloride (anhydrous J with_ Nalco Dr2339_.
This poly electrolyte had proved to he active when used w,itu_ ferric
chloride in laboratory tests.
96
-------
7. The addition of ferric chloride at a dosage of 50 rag/1.
S. The addition of ferric chloride at a dosage of 100 mg/1.
9. Tests on packinghouse waste only with, ferric chloride
(anhydrous) and Nalco 675.
Tests run after Swift and Company discontinued slaughtering operations were:
1. Tests on separated "domestic" waste. These wastes were predominantly
a mixture of wastewater from the residential area of the City of South
St. Paul and wastes from the stockyards. Small quantities of
packinghouse wastes were in this stream, but these were not substantial.
2. Some tests were run using ferric chloride and Nalco 675 on the
waste as received after the reduction in volume of packing industries.
These tests were run on the combined wastewater and on both the
separated industrial and domestic wastewater.
3. A one month test was run on the entire plant flow of combined
wastewater using ferric chloride and Nalco 675, Samples were taken
throughout the plant and an evaluation made on the effectiveness of
this dual system of chemical treatment on overall plant efficiency.
4. A control montu with no chemicals added.
Tests 1 through 4 were conducted in the following order:
1. Anhydrous ferric ciiloride followed by Dow A-^23
2. Buckbee Mears ferric chloride followed by Dow Ar-23
3, Pennsalt ferric clilori.de followed by Nalco 675, and
4. Buckbee Mears ferric chloride followed by Nalco 675.
97
-------
The ferric chloride was added to tiie wastewater at a point just after
it was split between the test and control tanks. The wastewater was
then passed through the mixing chamber and into the mechanical flocculators.
The polyelectrolyte was added to the waste approximately 10 to 15 minutes
after the addition of ferric chloride in the mechanical flocculators.
Measurements were taken of the water depth of the flow in the parshall
flume meters several times a week during these tests and all subsequent
tests to assure that both meters were working accurately and that the
stream of wastewater was being divided equally between the test and
control tanks.
Samples were taken at the influent to the test system and at the effluent
to the test and control tanks. Samples were taken hourly on a 24 hour
basis from which a composite sample was prepared. Grab samples were taken
every two hours during the day from 8 AM to 8 PM and every four hours at
night. Suspended solids were run on each, of the grab samples and ROD's
were run on four sets of the grab samples taken at 1Q AM, 12 Noon, 2 PM,
and 4 AM. The summary of the results of the. plant scale tests are given
in Table 46.
OBSERVATIONS FROM PLANT SCALE TESTS
All four tests showed a significant reduction in suspended solids and
BOD for the test tank versus the control tank. This was true of botiv the
grab samples and the composite samples. Since a much larger number of
grab samples were available for analysis, the results of the tests on
grab samples are more reliable. The results are given in Table 46.
Statistical analysis for significance of difference in suspended solids in
the effluents of the test and control tank indicate a significant difference
for all four tests. The results of this statistical analysis are also
indicated in Table 46.
98
-------
VO
Table 46
Summary Of Test Data
Suspended Solids mg/1 BOD mg/1
Chemicals Inf.
Grab Sample Average
A-23 FeCl3 621
Anhydrous
A-23 FeCl3 Waste 552
(As Liquid)
675 FeCl3 644
Anhydrous
675 FeCl3 Waste 803
(As FeCl3)
Composite Samples
A-23 FeCl3 631
Anhydrous
A-23 FeCl. Waste 525
(As Liquid)
675 FeCl3 635
Anhydrous
675 FeCl-j Waste 875
Test
113
117
120
106
106
131
130
118
Red.
81.8
78.8
81.4
86.8
83.2
75.0
79.5
86.5
Control
174
162
211
202
168
147
183
220
Red.
72.0
70.7
67.2
74.9
73.3
72.0
71.2
74.9
Test -
Control
61
45
91
96
65
16
53
102
Inf.
1322
1244
1338
1272
985
814
952
1178
Test
549
604
555
632
451
514
486
519
Red.
58.5
51.5
58.5
50.3
54.2
36.8
48.9
56.0
Control
663
707
691
772
536
584
618
584
Red.
49.8
43.3
48.3
39.4
45.5
28.2
35.2
50.3
(As FeCl3)
-------
Table 46 (Continued)
Comparison Of Test And Control
o
o
Test
A-23 And Anhydrous FeCl3
A-23 And Waste Fed3*
675 And Anhydrous FeCl3
675 And Waste FeCl3**
Sus. Solids
Average Diff.
Std. Deviation
No. Of Paired
A-23 and Anhydrous FeCl3
vs.
675 and Anhydrous FeCl3
A-23 and Anhydrous FeCl3
vs.
A-23 and Waste FeCl-j
675 and Anhydrous FeCl3
vs.
675 and Waste FeCl3
Test- Control
-61
-45
-91
-96
Average Diff.
31.4
47.6
65.1
51.8
Comparison Of Tests***
Test Mean Variance
1st
2nd
1st
2nd
1st
2nd
113.
120.
113.
121.
120.
108.
12 4033.50
74 3811.54
12 4033.50
23 3127.65
74 3811.54
32 2511.89
t Observations
-11.39
- 5.55
- 8.27
- 9.45
No. Of
Samples
34
35
34
31
35
25
34
34
35
26
t
-0.498
-0.544
0.820
.62
.59
.41
* Added as mg/1 of liquid
** Added as mg/1 of solid
*** Comparison of means of effluent concentrations were made rather than difference in concentration
because of more similtar populations
-------
Figures 7 through 10 indicate tne suspended solids concentration for tue
four tests using frequency distribution curves. These curves serve to
illustrate the significance of tiie data. In all four cases the suspended
solids in the effluent of the test tank were significantly less than
those in the effluent of the control tank.
When results of the test A-23 and anhydrous ferric chloride are compared
to the use of A-23 with the waste ferric chloride, the anhydrous ferric
chloride was slightly more efficient. However, the waste ferric chloride
was added as 50 mg/1 of solution which meant that only 22 mg/1 of ferric
chloride was being added compared to an average dosage of 40 mg/1 of
ferric chloride for the weeks run when anhydrous ferric chloride was used.
Statistical comparison of tne suspended solids in the effluents of the test
tank during the two tests indicate no significant difference.
During the test with Nalco 675 and anaydrous ferric chloride the ferric
chloride was added at a dosage of 50 rag/1 during the daytime. When the
saiae polyelectrolyte was run using Buckbee Hears waste ferric chloride, the
ferric chloride solution was added at 50 mg/1 of ferric chloride.
A comparison of these two tests based on data obtained from the grab
samples taken would indicate little difference between the suspended solids
reduction and BOD reduction. The two types of ferric chloride appear to
be equivalently effective in the flocculation process though the. waste
ferric chloride was somewhat less costly. Again, comparing the suspended
solids in just the effluent of the test tank by statistical means, no
significant difference appeared.
A comparison of the tests using Dow A-23 in the dual system and Nalco 675
in the dual system indicate tnat the Nalco 675 and Dow A-23 were equivalent.
The daytime dosage of ferric chloride in the dual system using Nalco 675
101
-------
1200-
1000-
800-
Influent
o
10
o
(U
o
z
IU
2}
o
c
3)
m
i
600-
400
200-
Test-— -z^ ^~~~& — - — TA
Effluent ^~~~^
90
i
80
i
70
i •
60
i
50
i i
40 30
i
20
i
10
95
PERCENT OF CONCENTRATIONS LESS THAN
FREQUENCY OF OCCURRENCE
THAT SHOWN
- NALCO 675 + ANHYDROUS FeCl3
-------
G>
c
3)
m
i
00
1200-
1000-
800-
to
o
_l
o
v>
600-
Q
Ul
Q
2
a 400
200-
Influent
Control
Effluent
Test
Effluent
i
90
80
I
70
I I I 1
60 50 40 30
20
10
PERCENT OF CONCENTRATIONS LESS THAN THAT SHOWN
FREQUENCY OF OCCURRENCE
- NALCO 675 * WASTE FeCI3
-------
1200-
1000-
- 800
c
•x
m
i
<0
V)
O
Ul
o
111
Q.
CA
600-
400-
200-
Influent
Test
Effluent
1 1 1 1 1 1 1 1
90 80 7O 60 50 40 30 20
PERCENT OF CONCENTRATIONS LESS THAN THAT
FREQUENCY OF OCCURRENCE
I '""
10
SHOWN
- DOW A23 * ANHYDROUS FeCI3
-------
m
i
1200-
1000-
800-
o 600
_i
o
tn
Q
LJ
g 400
a
in
200-
Inftuent
Test
Effluent
i
90
I
80
I
70
I
60
I
50
i
40
I
30
i
20
I
10
PERCENT OF
CONCENTRATIONS
FREQUENCE OF
LESS THAN THAT SHOWN
OCCURRENCE
DOW A23 + WASTE FeCI,
-------
showed an average of about 50 mg/1 while ferric chloride dosage with A-23 was
40 tng/1. Statistical analysis of suspended solids in the effluent of tne
test tank during the two tests indiactes no significant difference.
The conclusions drawn from tnese tests indicate little difference in the
effectivness in the two types of ferric cnloride in the dual systems
tested. Also, from a statistical evaluation of the data the two
polyelectrolytes (Dow A-23 and Nalco 675) appeared to be about equally
effective.
FURTHER EVALUATION OF WASTE FERRIC CHLORIDE
Further tests were run with Dow A-23 and Buckbee Hears waste ferric
chloride to determine just how effective higher concentrations of
ferric chloride would be with the Dow A~23. Also, tests using two
different concentrations of ferric chloride were run in an attempt
to determine if optimum dosage of ferric chloride could he obtained.
The dosages of ferric chloride used for the two tests were 50 mg/1 and
40 mg/1, respectively.
In the first test the actual average ferric chloride dosage used in this dual
system was 49.3 mg/1. The results of these tests are shown in Table.
47 along with the data collected in the four tests described previously.
The average suspended solids in the grab samples taken of the effluent
for the test and control tanks showed 235 mg/1 in the effluent of the
control tank and 129 mg/1 in the effluent of the test tank. This
is comparable to the data collected for the previously described tests
run using the dual system of Nalco 675 and ferric chloride at an actual
average dosage of 46.9 mg/1 for the test period. Statistical analysis
comparing the data of the effluents of the dual systems of ferric chloride
and Nalco 675 and ferric chloride and Dow A~23 verified there was no
significant difference in the effectiveness of these polyelectrolytes.
106
-------
Table 47
Summary Of Test Data
Ave. Sus. Solids mg/1 Test Less BOD mg/1
Chemicals Daytime Dosage InFT feat Corit. Control Inf. Test C~otitrol Diff.
A-23/Anhydrous PeCl3 37.1 621 113 174 61 1322 549 663 114
A-23/Waste FeCl3 19.1 552 117 162 45 1244 604 707 103
675/Anhydrous FeCl3 45.4 644 120 211 91 1338 555 691 136
675/Waste FeCl3 46.9 803 106 202 96 1272 632 772 140
A-23/Waste FeCl3 49.3 744 129 235 106 1606 669 869 200
A-23/Waste FeCl3 38.0 754 137 230 93 1481 650 852 202
D-2339/Anhydrous FeCl3 61.2 1180* 159 245 86 —
-------
Table 47 (Continued)
Comparison Of Test And Control
Sus. Solids Standard
Ave. Diff. Test Deviation In # Of Paired
Test
A-23 and Anhydrous FeCl3
A-23 and Waste FeCl3
675 and Anhydrous FeCl3
675 and Waste FeCl3
00 A-23 and Waste
A-23 and Waste Fed3
D-2339 and Anhydrous
Less Control
-61
-45
-91
-96
-106
-93
-86
Ave. Diff.
±31.4
147.7
165.1
±52.2
±85.7
±71.1
±72.8
t
-11.39
- 5.55
- 8.27
- 9.45
- 6.70
- 7.45
- 7.68
Observations
34
34
35
25
49
38
42
P
.01
.01
.01
.01
.01
.01
.01
-------
The second test run was an attempt to optimize the dosage of ferric
chloride and an attempt was made to hold the daytime dosage at 40 mg/1
of ferric chloride with the nighttime dosage at 20 mg/1. The actual
average daytime ferric chloride dosage was 38.0 mg/1 The average suspended
solids for the grab samples taken at the effluent of the control tank
was 230 mg/1 while the effluent of the test tank had a suspended solids
of 137 mg/1. The BOD of the control and test tank effluent was 869 mg/1
and 669 mg/1. respectively.
These results when compared to the ottuar tests of the dual system
of Dow A-23 and waste ferric chloride indicated that the dosage did not
tend to optimize but rather the effectiveness of flocculation of the
system increased with increased dosage of ferric chloride. This is
illustrated by the average difference in suspended solids obtained
from the control minus the test tank as compared with, the three dosages
of the waste ferric chloride used with Dow A-21. This is as follows;
Ferric Chloride Dosage Suspended Solids Controls-Test
19.1 mg/1 45 mg/1
38.0 mg/1 93 "B/1
49.3 mg/1 106ng/l
TESTING NEW DEVELOPMENT PRODUCT
Nalco Chemical Company had a chemical which was thought might be more
effective than Nalco 675 and slightly less costly. Nalco's field
representatives were given an opportunity to demonstrate the effectiveness
of the new product in the laboratory. The product was D.2339-, and
while it had a development number, it was in full production at the
time the test was made but had not yat been assigned a product number.
109
-------
The result of the laboratory tests was good. It appeared to be slightly
more active than the Nalco 675. On the basis of these results it
was decided to give this polyelectrolyte a one-week trial on a plant
scale.
The polyelectrolyte was added at an estimated dosage of 1.0 mg/1
from 8 AM to 10 PM and at a dosage of 0.75 mg/1 from 10 PM to 8 AM. It
was added to the flocculating chambers as had been done in previous tests.
The ferric chloride used was anhydrous ferric chloride. The dosage was
61.2 mg/1 from 8 AM to 10 PM. The night time dosage was about 35 mg/1.
The results of this test are saown in Table 47. The average suspended
solids run on grab samples was 245 mg/1 for the control tank and 159 mg/1
for the test tank. BOD's were not run on samples taken during this week.
The results of suspended solids, however, did indicate a substantial
difference between the test and control tanks. The results obtained in
suspended solids reductions did not show a large difference from those
obtained using Nalco 675 or Dow A-23. The price of the polyelectrolyte
D-2339 was 5 to 15% less than Nalco 675 depending upon tke quantity
purchased. Since the ferric chloride represented about two-rtuirds of the
chemical costs, the overall reduction of the chemical cost by the use
of D-2339 compared to Nalco 675 would be in the order of 3%. This
percentage would be further reduced when labor costs and equipment
costs were taken into consideration.
PLANT EVALUATION OF FERRIC CHLORIDE ALONE
In order to determine how much the polyelectrolytes being run in the dual
systems were contributing to the reduction of suspended solids, two
plant scale tests were run in which- only ferric chloride was added witlx
no polyelectrolytes. The estimated dosage of ferric chloride was
50 mg/1 in one test and 100 mg/1 in the other.
110
-------
In the first test the average actual dosage of ferric chloride fed
was 47.6 mg/1. The ferric cnloride formed a fine floe as was observable
in previous laboratory and plant tests. But, with no organic polyelectro-
lyte to further flocculate this fine floe it was carried through the
tank and over the weirs. The floe was readily observable in the test tank
effluent as it flowed over the weir and as settleable solids tests in
the imhoff cone test. It was also reflected in suspended solids tests
run on grab samples.
The suspended solids in the effluent of the test tank were actually
higher than those in the effluent of the control tank as indicated by
the average of suspended solids run on grab samples. This was also the
result in the average of the composite samples. The effluents of the
two tanks showed no significant difference in BOD reduction between the
two tanks. This was true for both the grab samples and the composite
samples.
In the second test the actual average dosage of ferric chloride was
98.4 mg/1. Again a fine floe could be observed passing over the weirs
of the test tank wuile none was observed passing over the weirs of
the control tank. The suspended solids in the effluent of tne test
tank were again higher than the suspended solids in the control tank.
The BOD in the effluent of ttie two tanks was not significantly different.
The results of the tests are saown in Table 48. The data is graphically
illustrated in Figures 11 and 12. These figures plot the frequency
distribution of the suspended solids analysis for each of the two tests
using 50 mg/1 and 100 mg/1 of ferric chloride, respectively. As indicated
for samples having concentrations of over 200 mg/1 of suspended solids,
the test tank effluent had a higher suspended solids concentration than
did the control tank. The results confirmed the explanation for the
effectiveness of the dual system of ferric chloride plus a polyelectrolyte.
Ill
-------
Table 48
FeCl3 Only
Grab Samples
Fed 3
Dosage
50 mg/1
100 ng/1
50 mg/1
100 mg/1
Fed 3
Dosage
50 mg/1
100 mg/1
50 mg/1
100 mg/1
Suspended Solids rag/1
% % Test Less
Inf. Test Red. Control Red. Control
863 348 59.6 262 69.6 -86
923 312 66.6 223 75.9 -89
No. Of
Samples t p
36 2.20 .03
36 1.66 .10
Composite Samples
Suspended Solids rag/1
% % Test Less
Inf. Test Red. Control Red. Control
891 351 60.5 2142 72. 8 -109
1014 484 52.4 236 76.7 -248
No. Of
Samples
5
4
Inf. Test
1589 905
2140 861
No. Of
Samples
14
5
Inf. Test
1389 725
2075 630
No. Of
Samples
5
4
BOD mg/1
%
Red. Control
43.0 916
59.6 921
t
0.27
0.80
BOD mg/1
%
Red. Control
47 . 7 744
69.5 805
% Test Less
Red. Control
42.3 11
57.0 60
P
.79
.47
% Test Less
Red. Control
46.4 19
61.0 175
-------
t400-
1200-
1000-
800-
2 600
_j
o
o
g 400
Q.
-------
1400
I2OO
1000 -
^ 800-
o>
E
VI
o
6OO -
o
ui
o
z
ui
Si 4OO
CD
C
31
m
I
ro
200-
Influent
Control
Effluent
I I I I I I I T I
9O 80 7O 6O 50 40 3O 20 10
PERCENT OF CONCENTRATIONS LESS THAN THAT
FREQUENCY OF OCCURRENCE
SHOWN
- FeCI3 ot I00mg/l
-------
PLANT EVALUATION ON SEPARATED PACKINGHOUSE AND DOMESTIC WASTES
Following the tests on ferric chloride some tests were run using the
dual system on separated packinghouse and domestic waste. Packinghouse
waste alone was tested for a number of days followed by tests on
domestic waste only. The domestic waste was not truly domestic waste
but included waste from the stockyards and waste from two small
packing plants. While it was stronger than normal domestic waste, it
more closely resembled a domestic waste than did the combined waste.
For both wastes preliminary jar tests were made to determine which of the
polyelectrolytes previously used would he most effective on the separated
waste. The results of these jar tests are shown in Tables 49 and 50.
For the industrial waste 50 mg/1 of ferric chloride was added followed
by 1.0 mg/1 of the polyelectrolyte. Dow A—23 was added to one beaker
and Nalco 675 was added to a different beaker. In addition, a control
sample of waste to which nothing had been added was stirred and s.ampled
in the same manner. The dynamic method of sedimentation was used with.
samples for suspended solids being drawn off just below the surface while
the flocculated wastewater was being stirred at 10 rpm. A sample of the
degritted waste being tested was also taken. The results of these jar
tests indicated no significant difference in effectiveness between Dow A-23
and Nalco 675.
It was decided it would be worthwhile testing either one of the poly—
electrolytes on a plant scale. However, there was not enough time
to give both a satisfactory test. Since a quantity of Dow A-23 sufficient
for the test on industrial waste had been purchased earlier, It was
decided to use this polyelectrolyte in testing the industrial waste.
The preliminary jar tests on domestic waste indicated that when using
the dual system of ferric chloride and A-23 the dosage of ferric chloride
115
-------
Table 49
Jar Tests On Industrial Wastes
FeCl3 50 mg/1 Polyelectrolyte 1.0 tng/1
Suspended Solids mg/1
Test
1
2
3
4
5
6
Ave.
A-23 vs.
Degritted Waste
596
772
1012
1128
1248
1000
959
675 t= .142
Control
388
572
620
496
680
592
558
p= .89
A-23
80
204
256
144
140
152
163
675
80
300
220
124
124
144
165
116
-------
Table 50
Jar Tests On Domestic Waste
Polyelectrolyte 1.0 tng/1
Degritted
Test Waste Control
Dow A-23
With 20 mg/1
Fed*
Nalco 675
With 20 rag/1
FeCl,
Nalco 675
With 50 rag/1
FeCU
1
2
3
4
5
Ave.
476
700
600
576
524
555
396
436
516
464
452
453
84
112
76
80
80
86
112
100
144
148
176
136
88
80
120
-
72
90
A-23 + 20 mg/1
Fed 3 vs.
Nalco 675 + 20 tng/1
FeCl3
A-23 + 20 mg/1
Fed 3 vs.
Nalco 675 + 50 mg/1
FeCl 3
t= 2.98
p .04
t= 0.036
p )0.90
117
-------
could be reduced to 20 mg/1 and still produce a good floe. To obtain a
floe of the same consistency with Nalco 675 it was necessary to use a
dosage of 50 mg/1 of ferric chloride. These results were verified by
suspended solids tests as indicated in Table 50. The dosage of poly-
electrolytes in both cases was 1.0 mg/1.
Eleven days of plant scale tests on the industrial waste were run first.
This was enough to produce sufficient data for reliable statistical
evaluation of the effectiveness of this dual system on the industrial waste.
The chemical dosage used was 50 mg/1 of ferric chloride and 1.0 mg/1
of Dow A-23 from 8 AM to 10 PM with 30 mg/1 of anhydrous ferric chloride
and 0.75 mg/1 of polyelectrolyte being added from 10 PM to 8 AM. Sampling
and laboratory testing was done as in previous tests. A summary of the
results of these tests are indicated in Table 50. The average suspended
solids for the grab samples in the effluent of the control tanks was
162 mg/1 and for the test tank tt was 128 mg/1. The BOD showed a
reduction from 951 mg/1 in the control tank to 826 mg/1 in the test.
For suspended solids the reduction was somewhat less than observed for
treatment of the combined waste. The conclusion drawn from this test was
that treatment of separated industrial waste was not more effective than
treatment of the combined waste. There would be no advantage to
treating industrial waste only since a greater reduction In terms of
total pounds of BOD and suspended solids could be obtained by treating
the much larger flow of combined waste.
Following the test on industrial wastewater a test was run on domestic
wastewater only. Initially a four day test was run using the higher dosage
of ferric chloride with A-23. The dosages used were 50 mg/1 of ferric
chloride and 1.0 mg/1 of Dow A-23 from 8 AM to 10 PM and 30 mg/1
of ferric chloride and 0.75 mg/1 of Dow A-23 from 10 PM to 8 AM. These
dosages produced a dense, heavy floe. The dosage was then reduced to
20 mg/1 of ferric chloride and 1.0 mg/1 of Dow A-23. This dosage was
fed 24 hours a day for a 7 day period. The summary of the results of these
tests are shown in Table 51.
118
-------
Table 51
Plant Scale Tests
Data For Wastewater Separation Tests
Summary Of Test Data
Chemica1 Dosage A-23 1.0 mg/1 And Anhydrous FeGl
Suspended Solids tng/1
Test
Average Grab
Combined
Industrial
Domestic
Domestic
FeCl 3
Dosage
Sample
37.0
49.7
49.7
20.5
Inf.
621
824
591
523
Test
113
128
78
98
Control
174
162
139
130
Test Less
Control
-61
-34
-61
-32
Inf.
1322
1571
1052
1019
BOD mg/1
Test
549
826
585
418
Control
663
951
690
528
Diff
114
125
105
110
Average Composite Sample
Combined
Industrial
Domestic
Domestic
37.0
49.3
49.7
20.5
631
896
572
605
106
140
88
92
171
172
142
138
-65
-32
-54
-46
985
1412
594
790
451
761
440
385
536
848
445
445
85
87
5
60
*Actual average daytime dosage indicated in mg/1
-------
During the first four days with the chemical dosage at the higher rate
the average difference in suspended solids for grab samples in the
effluent of the two tanks was about the same as it had been for the
tests on the combined wastewater.
This was true in spite of the fact that the floe which was produced at
the higher dosage appeared heavier and denser than that obtained with
the combined waste. The average suspended solids in the grab samples
of the effluent of the system was 130 mg/1 for the control tank and
78 mg/1 for the test tank. The BOD in the effluent of the control tank
was 690 mg/1 and in the effluent of the test tank 585 mg/1.
During the last seven days of the test on domestic wastewater the chemicals
were added at a lower dosage rate. The floe that formed appeared to he
about the same as that which had been formed during treatment of combined
waste. The average reduction of suspended solids in grab, samples did
not bear this out, however. The suspended solids In the effluent of
the control tank were 130 mg/1 while 98 iag/1 in the test tank, an
improvement of only 32 mg/1. The improvement in BOD reduction in
this test was about the same as for the first four days when chemicals were
added at the higher dosage.
The conclusions drawn from this data are that the same dosage of 50 mg/1
of ferric chloride and 1.0 mg/1 of A-23 during the day and a dosage
of 30 mg/1 of ferric chloride and 0.75 mg/1 at night were optimum for
all three systems. Since no great improvement in effectiveness of
the system nor any substantial reduction in chemical dosage could be
realized by treating one or the other of the separated sewages, the most
effective way to obtain the greatest overall reduction of suspended solids
was to treat the combined flow.
When A-23 was used with 20 mg/1 of ferric chloride it was significantly
less effective than 675 with 50 mg/1 of ferric chloride. When A-23 was
120
-------
used with 50 mg/1 of ferric chloride the difference in effectiveness
compared with 675 and 50 mg/1 of ferric chloride was not significant.
This seemed to negate the conclusion that Dow A-23 was more effective
on domestic waste than was Nalco 675.
PLANT EVALUATIONS USING NALCO 675 AFTER DECEMBER 1, 1969
Further tests were run on Nalco 675 using combined, domestic and
industrial waste to see how these compared with tests run using Dow A-23.
It was necessary to run a new series of tests with Nalco 675 because
Swift's packing plant had discontinued slaughtering operations since
Nalco 675 was last tested.
An overall comparison of all data was made to determine if one or the
other of the polyelectrolytes was more effective.
A test was made using full plant flow. Chemicals were added to the full
plant flow in all three primary sedimentation tanks and samples taken
throughout the plant for laboratory analysis. Th.e chemical dosage
was that found to be most effective considering all previous tests.
Laboratory parameters analyzed were: suspended solids, BOD, total
phosphate and total kjeldahl nitrogen with, the primary parameter being
suspended solids. This test was run for a full month in order to
determine what effect the chemical treatment would have on the biological
treatment processes at the plant. Preceding this test was an equal
period of time during which chemicals were not added to the plant but
during which the same laboratory analyses were performed as a control.
A five day series of tests was started on combined waste using a dual
system of ferric chloride and Nalco 675. The chemical dosages were
121
-------
ferric chloride at 50 mg/1 and Nalco 675 at 1.0 mg/1 from 8 AM to 10 PM
and ferric chloride at 30 mg/1 and Nalco 675 at 0.75 mg/1 from 10 PM
to 8 AM. The results are shown in Table 52.
The average of suspended solids for the tank effluent was 164 mg/1
and the test tank effluent was 98 mg/1. The BOD of the effluent was
589 mg/1 for the control tank and 465 mg/1 for the test tank. Those
results were equivalent to results obtained when tests were run using
Dow A-23 and ferric chloride at equivalent dosages.
Following the tests on combined waste, tests were intiated on separated
industrial waste from Armour's packing plant. The results are also
shown in Table 52. The average suspended solids for grab samples taken
at the effluent of the control tank was 204 mg/1 compared to 130 mg/1
in the effluent of the test tank. Tests for BOD in the effluent of these
two tanks showed a reduction from 797 mg/1 in the control tank to
585 mg/1 in the test tank.
The third test run using the dual system of ferric chlrodie. and Nalco 675
was run on the domestic wastewater received at the plant. As mentioned
previously this was not entirely domestic wastewater hut also consisted
of waste from the stockyards and two smaller packing plants. The chemical
dosages were the same as had been used in the two previous tests. A grab
sample showed a suspended solids average for the control tank effluent of
149 mg/1 while the test tank effluent was 74 mg/1. The BOD reduction for
this test went from 414 mg/1 for the control tank to 298 mg/1 for the
test tank.
A summary of results of tests run on combined and separated industrial
and domestic systems are shown in Tables 50 and 52. The results indicate
that Dow A-23 and Nalco 675 are both effective in reducing suspended solids
and BOD. Statistical analysis of the data presented in Table 46 indicates
the observed difference is not significant and based on the data
available it must be concluded that two polyelectrolytes are equivalent.
122
-------
CO
CO
Table 52 A
Plant Scale Tests
Data For Wastewater Separation Tests
Grab Samples
Summary Of Data
Comparing Dow A-23 With Nalcolyte 675
Chemical Dosage Dow A-23 1.0 tng/1 And Anhydrous Fed3*
Diff. Standard
Waste
•MOTMMBMMM
Comb.
Ind.
Dora.
A-23
at 20
Dose
mg/1
37.0
49.3
49.7
And FeCl3
mg/1
Suspended Solids mg/1 Test
Inf.
621
824
591
675
vs' at
Test
113
123
78
Cont.
174
162
139
And FeCl3
50 mg/1
-Cont .
-61
-34
-61
t=3.46
Chemical Dosage Nalcolyte 675
Comb.
Ind.
Dom.
50.1
49.4
48.2
494
723
351
98
130
74
164
204
149
-66
-74
-75
Deviation Of
Ave. Diff.
131.4
±50.5
155.9
Data
Pairs
34
79
33
A-23 And FeCl3
at 50 mg/1
1.0 mg/1 And
150.0
147.5
145.5
Anhydrous
42
43
41
Inf.
1322
1571
1052
vs.
FeCl3
1001
1490
693
BOD mg/1
Test
549
826
585
Cont.
663
951
690
675 And FeCl3
at 50 mg/1
465
584
298
589
797
414
Diff.
114
125
109
t= 1.42
p= .19
124
213
116
Waste—Combined, Industrial and Domestic
*Actual daytime dosages indicated in mg/1
Continued on next page
-------
to
Table 52 B
Data For Wastewater Separation Tests (Continued)
Composite Samples
Summary Of Data
Dow A-23
Suspended Solids mg/1 Diff. Test
Waste
Combined
Industrial
Domestic
Combined
Industrial
Domestic
Inf.
631
896
572
488
842
358
Test
106
140
88
103
130
76
Control -Control
168
172
142
175
208
141
-68
-32
-54
Nalcolyte 675
-72
-78
-65
Standard
Deviation
In Ave.
i20.1
116.2
t31.9
i!8.7
129.4
i 4.25
Inf.
985
1412
694
843
1467
465
BOD mg/1
Test
451
761
440
446
568
248
Control
536
848
445
524
652
320
Diff.
-85
-87
- 5
-78
-84
-72
Continued On Next Page
-------
Waste
Comb ined
Industrial
Domestic
Table 52 G
Data For Wastewater Separation Tests
Statistical Analysis
Grab Samples
Summary Of Data Suspended Solids
Dow A-23
Fed 3
Dosage
37.0
49.3
49.7
-11.39
- 6.30,
- 9.65
{.01
Combined
Industrial
Domestic
Nalcolyte 675
50.0 - 9.27
49.4 - 7.24
48.2 -10.04
125
-------
Of the six tests using Dow A-23 and Nalco 675 on industrial, domestic, and
combined wastewater, the greatest removal of solids and BOD was obtained
on the combined wastewater.
The industrial wastewater when treated with ferric chloride and Nalco
675 resulted in a greater difference in concentration of suspended solids
in the test and control tanks. Because of the larger volume of the
combined waste definitely greater total quantities of suspended solids
and BOD would be removed from the combined waste by chemical treatment than
from either the industrial or domestic wastewater. To maximize BOD
removal to meet effluent requirements the best system in a full plant-
scale test would be the dual system of ferric chloride and Nalco 675
on the total combined wastewater coming into the plant.
126
-------
SECTION X
RESULTS
All previous tests had been run on the influent and effluent to the
sedimentation tanks. While these results indicated a significant
increase in the effectiveness of primary treatment they did not in any
way indicate how the chemical treatment process would effect the overall
efficiency of the plant. This efficiency could be increased if the load
on the biological treatment processes was reduced resulting in decreased
effluent concentration from these units. If the chemicals resulted
in lower efficiency of the biological treatment processes, however,
the overall results would be uncertain. This lower efficiency might
be the result of toxicity to microorganisms from the high- dosage of
ferric chloride or might simply result when a lesser organic load was
applied to the treatment processes.
Sampling and laboratory testing were started a month before the
actual chemical treatment tests. Parameters tested for in the laboratory
were: suspended solids, BOD, total phosphorus, and total Kjeldahl
nitrogen. Samples were taken hourly of tli& raw industrial and raw
domestic influent and every two hours for the effluents of the grit
chambers, primary sedimentation tanks, and secondary sedimentation
tanks. The plant effluent was sampled every four hours. It was not
necessary to sample the plant effluent more often than this since the
seven day detention in the anaerobic stabilization pond tends to level
off the flow and concentration resulting in a relatively uniform effluent,
A composite sample based on flow was made for each sampling point.
Laboratory tests were then run on these composite samples. Suspended
solids tests were run on the composite samples and also daily on five
grab samples from each of the sampling points.
127
-------
In addition to the previously mentioned parameters which were examined
to determine the effectiveness of the treatment process a biological
examination was made to determine if there was any detrimental effect on
the biota of the filters and the anaerobic stabilization pond. A
consultant was hired to make a biological assay of the species and
relative numbers of individuals of species on the surface rocks of the
filters and at several stations in the pond from near the bottom to near
the surface.
The necessary equipment for addition of chemicals was prepared. Two more
distribution pipes were prepared to feed polyeletrolytes to the two tanks
to which chemicals had not been previously fed. A larger feed pump was
installed for the polyelectrolyte to enable the pumping of three times
the volume of solution thereby allowing the solutions to he maintained
at the same concentrations. The strength, of the ferric chloride solution
was increased to three times the concentration previously used to allow
the use of the same chemical feed pump on full plant flow.
During the total plant scale chemical feeding test the entire flow of
wastewater was directed through the chemical mixing chambers. The parallel
flume which had carried the control wastewater in previous tests was
shut off. Also, the line which had diverted excess wastewater not being
tested to a third sedimentation tank was closed. After the entire flow
had been passed through the chemical mixing chambers it was split so
as to feed the same amount of wastewater to each, of the three sedimentation
tanks.
During the month preceding the chemical addition data was collected
for the control period of the tests. During this period the plant was
operated at maximun treatment capacity as It was in the following one
128
-------
month test period. All units were kept in continuous operation and any
preventive maintenance requiring the shutdown of a unit was postponed until
the completion of the tests. In addition, as much secondary sedimentation tank
effluent as possible was recirculated dailjr. A 4.5 million liter (1.2
million gallon) storage tank was filled during the peak flow and this
was recirculated through the plant at night.
After the one month control period chemical feed was started. An attempt
was made to hold the daytime dosage of ferric chloride at 50 rag/1 and
the dosage of polyelectrolyte at 1.0 mg/1. The chemical dosage at night
time and weekends was held at approximately 30 mg/1 of ferric chloride
and 0.75 mg/1 of polyeletrolyte. The actual average chemical dosages as
fed are given in Table 53. The overall average chemical dosage for the
test period was 51.3 mg/1 daytime ferric chloride feed and 1.Q7 rag/1
daytime polyelectrolyte feed. The average actual dosage for night time
and weekends was 28.6 mg/1 ferric chloride and 0.70 mg/1 polyelectrolyte.
The biota of the trickling filters and the pond was sampled biweekly
during both control period and after chemical addition commenced. The
addition of chemicals did not have a detrimental effect on the biota of
either the trickling filters or the pond. While no significant change
was observed in species of microorganisms, the species present appeared
to be in greater abundance after chemical treatment. Appendix A is
the report of the effect of chemical treatment on the biological treatment
processes.
A summary of the data collected during the control and test periods are
given in Tables 54 through 59. The suspended solids reduction for grab
samples in the primary sedimentation tank was 63.0% for the control period
and 78.5% for the test period. The average suspended solids removed in
the primary tank during the control period was 320 mg/1 compared to 369 mg/1
during the last period. This is an average of approximately 100 sets of
data obtained from the grab samples taken.
129
-------
Table 53
Average Daily Chemical Dosage
tng/1
July Ferric Chloride !*alco _ ^
J y --- — - - m* 8 am - IJTpm TO pm - 8 am*
1.00 0.67
0.76 0.72
1.30 0.62
0.83
0.74
1.00 0.72
1.10 0.58
1.08 0.55
0.85 0.79
0.82 0.86
0.68
0.75
1.20 0.63
1.39 0.62
1.05 0.55
1.11 0.70
0.90 0.69
0.70
0.69
1.11 0.76
1.07 0.93
0.83 0.76
1.00 0.80
0.92 0.71
0.64
0.58
0.89 0.71
0.98 0.72
Average 51.3 28.6 1.07 0.70
970
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
8 am - 10 pm
47
66
71
M *•
_ —
55
45
50
48
47
*.«
• —
45
54
53
47
62
_—
_—
51
53
43
62
44
__
__
41
43
10 pm - (
28
34
22
29
28
30
28
31
34
32
26
27
24
29
21
22
29
25
33
28
36
33
25
26
29
20
28
29
* 24 hour dosage on Saturdays & Sundays
130
-------
This data follow the same trend as data collected on partial flow tests
run on the primary sedimentation tank as descirbed previously. This
reduction in suspended solids did not carry throughout the plant, however.
A smaller loading on the filters and th.e anaerobic pond resulted in less
reduction in these units. For suspended solid run on grab samples, the
filters showed a reduction of 118 mg/1 for the control period and only
59 mg/1 for the test period. The pond showed a suspended solids reduction
on grab samples of 20 mg/1 for the control period and 9 mg/1 for the test
period.
The suspended solids run on the composite samples for which there were
approximately 20 sets of data for both the test and control period showed
about the same results. A summary of these results are given in Table 55.
During the control period 202 mg/1 of suspended solids were removed in
the primary settling tanks while 255 mg/1 were removed during the test
period. The filters again showed a greater reduction in suspended solids
during the control period when loadings were greater. A reduction of 89 mg/1
was obtained as the wastewater passed over the filters and through the
secondary settling tank during the control period while only 49 mg/1 of
suspended solids were removed by this secondary treatment process during the
test period. The lower suspended solids and BOD removal on the filters was
a result of the reduced load applied to the filters. The efficiency of the
filters was not significantly changed as indicated hy the results of the
"Student" t-test run on efficiencies and presented in Table 55. The
composite sample showed a suspended solids reduction of 48.4% during the
test period and 57.0% during the control period and a ROD reduction of 64.5%
during the test period and 70.0% during the control period. This difference
in efficiencies is not statistically significant. The BOD reduction
obtained in the primary tank was 53.8% in the test period and 29.2% in
the control period. This was significant at the .05 level as indicated
131
-------
in Table 55. The reduction obtained in the primary tanks did not
result in a change in efficiency on the filters and the reduction in
the primary tank resulted in a significant difference in the effluent
of the secondary treatment process for the test and control periods.
The suspended solids reduction in the pond for composite samples was
about the same for the control and test periods, being 18 mg/1
and 20 mg/1, respectively. For both, the composite sample and the
grab sample the pond effluent did have slightly higher concentrations
of suspended solids during the control period than during the test period.
This is probably because the control period was started before the
spring turnover in the pond was complete and a larger amount of
settleable solids were in the effluent than during the test period.
The settleable solids for the composite sample are given in Table 59.
The average of settleable solids for the control period was 0.63 mg/1
compared to 0.12 mg/1 for the test period. These solids are relatively
stable from digesting on the bottom of the pond having volatile suspended
solids of about 64%. They, therefore, do not show up as an increased
BOD. Due to these settleable solids the observed difference in suspended
solids in the pond effluent for the test and control period was probably not
the result of chemical addition.
The results of the BOD test on composite samples are shown in Table 55.
A significantly greater amount of BOD was removed in the primary sedimentation
tanks during the test period than during the control period. A reduction of
148 mg/1 of BOD was obtained during the control period, compared to a
BOD reduction of 314 mg/1 during ths test period. Again, the reduced
loadings on the filters and pond caused a lower removal in these two units
for this parameter. The efficiency of the filters remained unchanged,
however. The BOD reduction on the filters for the control period was
132
-------
250 mg/1 while a reduction of 203 mg/1 was obtained during the test
period. The BOD of the secondary settling tank effluent was substantially
lower during the test than during the control period being 66 mg/1
and 107 mg/1, respectively. These results are shown in Figure 13.
The phosphorus removal by the chemical treatment process was significant
as indicated in data for the composite sample in Table 55. Over twice as
much phosphorus was removed in the primary sedimentation tank during the test
period as during the control period. The phosphorus reduction during the
test period was 5.3 mg/1 compared to 2.5 mg/1 during the control period.
The amount of phosphorus removed on the filters was about the same for
the test period and control period. In the pond a slight increase in
phosphorus was observed for both the test and control period. This was
due to the fact that the sludge supernatant from the sludge
lagoons flows into the pond returning some of the phosphorus previously
removed. The amount of phosphorus returned to the pond during the test
period was only slightly greater than that returned during the control period.
This overall phosphorus reduction for the plant was 51.9% for the test period,
or about twice that for the control period which was 26.0%.
The results of total Kjeldahl nitrogen tests run during the test and
control period are given in Table 55. A small but significant difference
was observed in the reduction of total Kjeldahl nitrogen as a result of the
chemical treatment process. Again, an increase in nitrogen was observed
in the pond for both the test and control period because of carry over
in the sludge supernatant.
The flow data and data on chlorine requirements for the control and test
periods are given in Tables 54 and 56, respectively. The average
wastewater flow is slightly less than the influent flow as a result of
leaching at the bottom of the pond and evaporation from the 11 hectare (27
acre) surface area.
133
-------
BOD AT SAMPLING POINTS
CONTROL PERIOD AND TEST PERIOD
FULL PLANT SCALE CHEMICAL TESTS
600
5QO
400
300
20O
IOC
0
0
z
H
0
r~
m
O
o
H
m
CO
m
0
O
H
Z
TJ
C
Z
/— \
X
3J
rn
0
r~
o
o
H
'JU
o
r-
TJ
m
o
o
H
m
H
T3
m
o
w
-D 0 H > g
= m - >
> s S m 2
3J o r~ o ^
-< > ~ m v.
m
2 0
H 0
> Z
H
tj ^
O
0
/^
INTROL
-i
m
CO
H
o o
TI
r 2 o
H
m
•o
m
30
o
™ z -^
m H 0
CO 0 _, g
~ m z
-o tn -) -i
m H 3) m
"D O O5
_ -0 fl H
o m '
O 33 -g TJ
" o mm
O _ — .
-> o °
o o
nrn
TJ
c
OJ
PLANT
INFLUENT
GRIT CHAMBER
EFFLUENT
PRIMARY
EFFLUENF
SECONDARY
EFFLUENT
PLANT
EFFLUENT
-------
Table 54
Flow And Chlorine Data
Control Period
Date
1970
May 20
May 21
May 22
May 25
May 26
May 27
May 28
June 1
June 2
June 3
June 4
June 5
June 8
June 9
June 10
June 11
June 12
June 15
June 19
June 22
June 23
June 24
June 25
June 29
Flow
Influent
M •»
—
10.10
10.22
9.47
10.78
8.61
9.28
8.82
8.51
8.00
9.15
9.08
9.25
9.66
8*60
11.88
12.13
10.76
10.00
•* •
-_
__
11.22
MGD
Effluent
8.05
9.12
12.76
8.69
9.37
11.22
12.63
9.08
8.88
8.41
7.49
8.20
8.48
9.40
9.93
8,08
9.79
9.46
8.70
7.93
9.10
9.46
9.75
9.03
Chlorine Data
MBIN 10.31
0 Of Samples
9.29
Chlorine Demand tng/1
16.8
18.6
17.6
15.9
10.0
10.0
12.2
17.0
17.5
18.5
14.1
16.7
14.5
15.5
15.1
13.1
15.4
18.2
13.6
13.4
12.9
10.9
10.5
10.1
13.9
24
Pounds Used
1130
1416
1820
1132
780
932
1280
1290
1240
1296
878
1144
1028
1216
1222
1060
1252
1308
984
884
980
860
840
762
1113
24
* Based on actual measurement
135
-------
Table 55 A
Data Of Full Plant Scale Chemical Treatment
Composite Sample
Test Period
Control Period
Sample
Point
Raw
Grit Eff.
Prim. Eff.
Secondary
Pond Bff.
Sus. Solids
mg/1
552
356
101
Sff. 52
32
Reduction
196
255
49
20
Percent
Reduction
35.5
71.7
48.4
38.5
Sus. Solids
mg/1
621
359
157
68
50
Reduction
262
202
89
18
Percent
Reduction
42.2
55.6
57.0
26.5
Percent removal by the entire plant
Number of samples 20
94.2
92.0
24
Test Period
Control Period
Sample
Point
Raw
Grit EEf.
Prim. Bff.
Secondary Eff.
Pond Eff.
BOD
mg/1
•^•^••MBW
616
583
269
66
22
Reduction
33
314
203
44
Percent
Reduction
5.3
53.8
64.5
66.6
BOD
mg/1
634
505
357
107
28
Reduction
39
148
250
79
Percent
Reduction
4.6
29.2
70.0
73.8
Percent removal by the entire plant
Number of samples 20
96.4
94.5
22
-------
Table 55 B
Statistical Evaluation Of Change In
Unit Efficiency With Chemical Treatment
Unit And
Parameter
No. Of
Test
Primary Sedimentation
Sus. Solids 20
BOD 19
Trickling Filters
Sus. Solids
BOD
20
19
Trials
Control
Tanks
23
20
214-
21
Std.
Test
8.2
10.0
23.2
10.3
Deviation
Control
16.9
11.0
25.2
11.3
t P
5.61 .01
1.06 0.29
.971 0.33
137
-------
Table 55 C
Test Period
Control Period
Sample
Point
Raw
Grit EEf.
Prim. EEf.
Secondary EEf.
Pond BEE.
Phosphorus
mg/1
12.9
10.9
5.6
4.7
6.2
Reduction
2.0
5.3
0.9
-1.5
Percent
Removal
15.5
48.4
16.1
—
Phosphorus
mg/1
12.7
11.5
9.0
8.5
9.4
Reduction
1.2
2.5
0.5
-0.9
Percent
Removal
9.4
21.8
5.6
—
Percent Removal by the entire plant
Number of Samples
51.9
22
26.0
to
00
Test Period
Control Period
Sample
Point
Raw
Grit EEf.
Prim. BEf.
Secondary BEf.
Pond BEf.
Nitrogen
mg/1
55.9
55.7
42.9
30.5
38.9
Reduction
0.2
12.8
12.4
-8.4
Percent
Removal
0.35
23.0
28.9
Nitrogen
mg/1
56.0
49.1
43.5
34.2
46.6
Reduction
6.9
5.6
9.3
-10.5
Percent
Removal
12.3
11.4
21.4
Percent Removal by the entire plant
Number Of Samples
30.5
20
24
20.1
-------
Table 55 D
to
Statistical Analysis Of Full Plant
Scale Chemical Treatment Data
Composite Samples - Primary affluent
No. Of Trials
Mean
Std. Deviation
Parameter
Suspended Solids mg/1
BOD mg/1
Phosphorus rag/1
Nitrogen mg/1
Test
20
19
20
20
Control Test Control
24 101 157
22 269 357
24 42.9 43.5
22 5.6 9.0
Table 55 E
Test
39
65
10.8
1.54
Control
55
73
3.4
1.02
Statistical Analysis Of Full Plant
Scale Chemical Treatment Data
Parameter
Suspended Solids
BOD mg/1
Phosphorus mg/1
Nitrogen mg/1
Composite
No. Of
Test
20
20
20
20
Samples - Trickling Filter Effluent
Trials Mean
Control Test Control
24 52 68
22 66 107
22 4.7 8.5
24 30.5 34.2
Std.
Test
!l9.2
1:25.6
i 1.07
1 5.28
Deviation
Control
125.8
155.8
i .85
t 5.61
3.81 <.01
4.04 <.01
t p
2.17 .03*
41.6 <.01*
2.24 .03*
* Indicates significant difference in Mean at the 0.05 level
-------
Table 55 F
Statistical Analysis Of Full Plant
Scale Chemical Treatment Data
Composite Sanyles - Plant Effluent
Parameter
Suspended Solids mg/1
BOD mg/1 20 22 22 28 ± 6.78 ± 7.50 .446 .66
Phosphorus rag/1
Nitrogen mg/1 20 24 38.9 46.6 ± 4.41 i 5.43 5.10 <.01*
*Indicates a significant difference in the Means at the 0.05 level
No. Of Trials
Test
20
20
20
20
Control
23
22
22
24
Mean
Test
32
22
6.2
38.9
Control
50
28
9.4
46.6
Std. Deviation
Test
110.3
± 6.78
± 1.16
1 4.41
Control
±24.1
± 7.50
1 0.56
i 5.43
-------
Table 56
Flow And Chlorine Data
Test Period
Date
1970
June 26
July 1
July 2
July 6
July 7
July 8
July 9
July 10
July 13
July 14
July 15
July 16
July 17
July 20
July 21
July 22
July 23
July 24
July 27
July 28
Flow K
Influent
10.91
10.97
11.81
11.35
14.27
8.73
9.98
7.57
8.18
8.31
7.73
6.62
9.60
10.61
10.59
—
Mean 9.81
ft Of Trials 15
IUU
Effluent
9.47
10.80
10.05
8.34
10.97
11.00
12.26
12.10
6.57
8.57
6.91
7.45
7.41
3.82
7.55
9.63
9.55
10.00
9.04
10.79
9.11
20
Chlorine Data
Chlorine Demand rag/1
14.1
9.6
14.5
6.2
4.6
4.1
2.8
5.2
3.0
8.4
3.4
3.9
3.8
6.6
2.7
3.5
4.0
3.6
3.1
2.8
5.5
20
Pounds Used'
1040
944
1216
432
424
372
284
520
164
600
196
244
232
212
172
284
316
300
236
248
422
20
* Based on, actual measurement
141
-------
Table 57
Grab Samples
Data Of Full Plant Scale Chemical Treatment
Suspended Solids Grab Samples
Control Days Average
Time
10 a.m.
12 Noon
2 p.m.
4 p.m.
8 p .m.
Ave. Of
All Data
Time
10 a.m.
12 Noon
2 p.m.
8 p.m.
Ave. Of
All Data
Raw
Ind.
1095
1091
1041
846
432
902
Rax*
Ind.
1032
1145
1235
325
923
Raw
Dora.
675
780
773
874
500
720
Suspended
Test
Raw
Dom.
737
703
602
350
591
Grit
Sff.
543
526
506
699
269
508
Solids
Days
Grit
Bff.
529
525
427
246
470
Prim.
Eff.
140
204
214
221
161
188
Grab Samples
Average
Prim.
Eff.
71
109
115
88
101
Final
Eff.
48
63
79
86
69
70
Final
Eff.
34
42
49
42
42
Time
8 a.m.
12 Noon
4 p.m.
8 p .m.
12 Mid.
—
Time
8 a.m.
12 Noon
4 p.m.
12 Mid.
•*
-------
Table 58
Flow And Chlorine Data Summary
Test Control
Period Period
Influent Flow 9.81 10.31
affluent Flow 9.11 9.29
Chlorine Demand 5.5 13.9
Pounds of Chlorine Used 422 1113
Chlorine Cost At 7.5^/lb. $31.65 $83.47
Chlorine Cost/rag $ 3.23 $ 8.10
143
-------
Table 59
Settleable Solids Pond Effluent
24 Hour Composite ml/1
Control
Period
May 20
May 21
May 24
May 25
May 26
May 27
May 28
June 1
June 2
June 3
June 4
June 5
June 8
June 10
June 11
June 12
June 15
June 16
June 19
June 22
June 23
June 24
June 25
0.5
1.5
0.3
0.4
1.4
0.6
2.6
0.5
0.5
0.2
0.3
0.5
0.9
0.4
0.6
0.4
0.4
0.7
2.6
0.2
0.4
0.2
0.5
Test
Period
June 26
June 28
June 29
July 1
July 2
July 8
July 9
July 10
July 12
July 14
July 15
July 16
July 19
July 20
July 21
July 22
July 23
July 24
July 26
July 27
0.2
0.1
0.2
0.1
0.1
0.2
0.2
0.1
0.0
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.0
0.1
0.2
Average
0.12
Average
0.63
144
-------
About a week after chemical treatment a sharp drop was observed in chlorine
demand. This is one unanticipated beneficial result from using the
chemical treatment process. The average chlorine demand for the control
period was 13.9 mg/1 while for the test period it was 5.5 mg/1. This is a
substantial reduction in total use of chlorine and would help to offset the
cost of chemicals in a plant scale operation. The reduction in cost of
chlorination was $1.29 per million liters ($4.87 per million gallons).
This is shown in Table 58.
The chlorine demand rose slowly after the use of chemicals was
discontinued taking approximately a month to reach the level it had been
before chemical treatment was started. Again, this may not be entirely
attributable to the ferric chlirde since during the later part of the
test period, due to algae, there was dissolved oxygen in the effluent.
The presence of free dissolved oxygen no doubt also oxidizes the hydrogen
sulfide thus reducing the chemical chorine demand. It was only when the
effluent turned green that the D.O. tests were run. The effluent turned
green the last week of the test period and most of the D.O. tests were
run after the test period. Only two D.O. tests were run before this, one
during the control period and one during the test period. Both of these
tests showed a zero D.O. Of course, this presence of algae could have
been stimulated by the forming of an insoluble precipitate of ferrous
sulfide which reduced the immediate oxygen demand allowing the algae
to multiply and thus generate more oxygen by the photosynthetic process.
This latter condition seems to be what happened since the chlorine
demand started increasing slowly after discontinuing the test and by
September 22 was nearly equal to the chlorine demand prior to the chemical
feeding. Also, the green color due to algae slowly disappeared.
The overall results of the data collected during this test indicate that
no significant increase in BOD or suspended solids removal could he
145
-------
achieved at the South St. Paul plant by the use of this chemical treatment
process. If it were not for the anaerobic pond as the final treatment
process, chemical treatment might be worth considering as an interim
method of treatment since the BOD in the effluent of the secondary
sedimentation tank for the test period was significatly lower than that
for the control period. The BOD of the filter effluent was 66 mg/1 during
the test period and 107 mg/1 during the control period. This is a
significant difference in BOD and was the result of chemical treatment.
In deciding whether or not to use chemical treatment in a trickling
filter plant it would be necessary to consider if this reduction was
sufficient and whether it could justify the cost of chemicals. For a
treatment plant treating packinghouse wastes with primary treatment only,
this dual system of chemicals could be of value in reducing suspended solids
and BOD.
Whether or not a system is practical depends, upon the cost of operation.
The plant-scale system tested at South- St. Paul did not prove to be practical.
However, if it were used at a chemical treatment plant not followed by biological
treatment the economic consideration would be of importance.
The cost of operations of chemical treatment facilities are dependent
upon plant size. The costs of operation were calculated on the basis of
a 438 I/sec ( 10 mgd) flow. Labor, capital construction, and utility
costs are based on information obtained during the plant-scale evaluation.
Chemical costs are based on dosage utilized in the plant investigation and
price quotes from the chemical companies. The plant evaluation
utilizing full-scale equipment offers reliable cost information. A summary
of these costs are shown in Table 60.
The cost of ferric chloride is $0.121 per kilogram C$0.055 per pound!, for
large quantities delivered to the nearby Metro plant. Both polyeletrolytes
were priced at $3.52 per kilogram C$1.60 per pound) plus $0.044 per kilogram
146
-------
Table 60
Economics
Based On A 10 MGD Plow
Item
$/Year % Of Total
Chemical and handling cost $117,416 71
Capital construction and maintenance 24,958 15
(piping, chemical building and
feeding equipment)
Labor (Operator, sampler, chemist) 19,100 12
Utilities (Gas, water electricity) 2,760 2.
TOTAL $164,23U 100
147
-------
($0.02 per pound) freight for 2270 kilogram (5000 pound) lots. The dosages
used were 50 mg/1 ferric chloride and 1.0 mg/1 polyeletrolyte for 14
hours and 30 mg/1 and 0.75 mg/1, respectively for 10 hours. Chemical handling
costs were estimated at $3,500 per year.
The capital construction was amortized over 30 years for the chemical
building, 5 years for the feeding equipment, and 20 years for the
special piping, all at a 57, rate of interst. Maintenance of these items
was estimated at $2,500 per year. Labor costs were figured at 3120 hours
per year at $5.00 per hour. This manpower is necessary to operate the
feeding equipment, chemical preparations, and sampling. Laboratory
equipment and 0.3 man-years of a chemist's time cost an additional $3,500
per year. The costs of electricity, gas, and water were estimated on a
monthly basis of $168, $20, and $42, respectively.
A reduction in cost of $1.29 per million liters (.$4.87 per million gallons).
was realized from reduced chlorine demand. This was subtracted from the
cost of chemical treatment.
The total cost of chemical treatment oyer and above the normal operating
costs of the plant came out at $10.58 per million liters (.$40 per million
gallons). Since it is considered likely that the reduction in chlorine
demand occurred in the pond, the cost of chemical treatment for a trickling
filter plant without an anaerobic pond as a final treatment step would be
$11.90 per million liters ($45 per million gallons^.
148
-------
SECTION XI
PUBLICATIONS AND PATENTS
1. Larson, K. D., R. D. Crowe, D. A. Maulwurf, and J. L. Witherow.
"The Use of Polyelectrolytes in the Treatment of Combined Meat-
Packing and Domestic Wastes." (Presented at the Water Pollution
Control Federation Annual Conference, 1970).
2. Patent pending: July 25, 1969. "Process for Improving the Treatment
of a Combined Mixture of Domestic and Meat Packing Wastes."
For: U. S. Department of the Interior, Environmental Protection
Agency, Water Quality Office.
Description: The use of a dual chemical system of ferric chloride
followed by a high molecular weight organic anionic polyelectrolyte
to effectively flocculate the waste, thereby improving the removal
of suspended solids and biochemical oxygen demand C.BOD) in the
primary clarification step of a waste treatment plant.
149
-------
SECTION XII
APPENDICES
A. Biological Studies 151
B. Buckbee Hears Waste Ferric Chloride Analysis 175
C. Sludge Studies 178
150
-------
APPENDIX A
Report On The
Biological Changes in the South St. Paul Sewage Treatment Plant Before
And After Treatment With Fed And Naclytle 675 Coagulating Agents
By Keith M. Knutson, Ph.D
Bioengineer Consultant
St. Cloud State College
Department of Biology
St. Cloud, Minnesota 56301
Project Initiated: June 15, 1970
Project Terminated: July 28, 1970
Completion Report Submitted: September 10 , 1970
151
-------
INTRODUCTION
The major objective of modern wastewater treatment is the removal of
putrescible organic matter, therefore reducing the biochemical oxygen
demand (BOD). At the South St. Paul Sewage Treatment Plant the initial
mechanical solids removal of domestic and industrial wastewater is very
efficient due to the fact that the primary tanks are preceded by pre-
primary tanks and flocculation units. Even so, due to the nature of th_e
predominately packing plant industrial sewage, a considerable load of
organic matter and complex inorganic colloidal particles remain in
suspension. Many of these particles are large enough, to settle out
and are removed by bottom-scrapers in pre-primary settling tanks. However,
as is well known, additional settling of even colloidal sized particles
can be effected by feeding chemical coagulating agents to wastewater.
Therefore at the South St. Paul Sewage Treatment Plant (as will be
expanded on in their general report^ a demonstration project was
planned to test the efficiency of using cations and polyelectrolytes.
METHODS AND MATERIALS
Ferric chloride (FeCl^) at 50 mg/1 and Nalcolyte 675 anionic polyeletrolyte
at 1 mg/1 final concentrations, were fed to the wastewater after pre~
primary settling from June 26, 1970 to July 27, 1970, with a minor
shut-down due to pump failure from June 27 to 29, 197Q. The coagulant
entered the wastewater in mechanically stirred flocculation tanks
prior to entering the primary settling tank. Sampling was conducted
before treatment (June 17 and 24} and during or after treatment commenced
(July 8 and 22) with an inspection trip concerning a peculiar algal
bloom July 29, 1970.
Trickling filter stones were collected from the surface and placed in
sterile 18 oz. Whirl-Pak (NASCO) hags. Most organisms were identified
152
-------
and original illustrations were made of each (Plate I) to acquaint
the reader with the biota. The association of organisms was carefully
tabulated so that any severe changes or treatment effects could be
singled out.
The anaerobic stabilization pond was sampled using a 12 foot boat where
six stations were established about 450 feet apart, and number seven station
at the pond effluent about 900 feet from station 6 (Fig. Al). At each
station samples were collected with a modified PVC dissolved oxygen
sampler (Hach, 1968). A 6 oz. sterile Whirl-Pak bag was placed into the
sampler after sterilizing the sampler cover by soaking it in 95% isopropyl
alcohol. The device was lowered to about 7 feet, opened and raised to the
surface, therefore obtaining an integrated (composite) sample
representative of the 7 foot column of water. The composite water
sample thus obtained was collected in a sterile container making
bacteriological analysis possible. Bacteria were diluted from 0.01 to
0.000001, to insure a plate count from 30 to 300 colonies, and each
dilution was plated in triplicate in Triptone Glucose Extract Agar (or
plate count agar) incubated for 48 hours at 35%F (Standard Methods, 1965).
A thin layer of sterile TGEA was poured over the culture agar to retard
the growth of fungi and strict aerobic bacteria since the pond was
anaerobic (o mg/1 dissolved oxygen from the Winkler test). Colonies
were counted and recorded as colonies per milliliter of pond water.
Plate II illustrates many anaerobic stabilization pond biota.
Algae were filtered in 1 to 5 ml aliquots on 0.45 micron Millipore-
fliters. Filters were placed on glass slides and cleared with- immersion
oil and stored with a cover glass over them. The algae on the filters
were enumerated according to the procedure of McNabb (1960) . All algae
were identified from wet mounts in the living condition, an absolute
prerequisite for the McNabb method. Protozoans were identified but not
counted, although many are illustrated on Plate I and Plate II. Water
temperature was recorded at each station with a thermistor and dissolved
153
-------
oxygen samples were taken periodically using the Azide modification of the
Winkler tehcnique (Standard Methods, 1965). Secchi disc transparency
was an additional physical measurement taken at each station on the
anaerobic stabilization pond. This measurement consists of lowering into
the water a 20 cm black and white painted disc to the depth which it can
no longer be seen and recording this depth as the limit of transparency
using the human eye as the sensitive photo cell. All sampes were kept
in a styrofoam cooler during transportation to St. Cloud, Minnesota and
were evaluated immediately, less than 6 hours after their collection.
RESULTS AND DISCUSSION
1. Biological Filter (or Trickling Fllterl microorganism analysis-'-'
A qualitative analysis of surface filter stones revealed a unique
association of organisms. Most abundant was the slime colonial
bacterium Zoogloea ramigera which, covered the stones. At the fringe
(contact with other stones) of each stone was the grey colored filamentous
fungus Geotrichum candidum that forms a felt-like mat in and upon which.
supported the growth of the grazers, nematode worms and Psychoda the
filter fly larvae. Surface stones, being exposed to solar radiation
also support the growth, of photosynthetic organisms (the algael, as
opposed to Zoogloea and Geotrichum that obtain their nourishment
heterotrophically or saprophytically from decaying organisms or the rich-
wastewater being continuously sprinkled on them.
The algae reponsible for the dull green to blue-green appearance of the
surface stones are the domiant algae Chlorococcum sp. and Nltzschia
palea along with Phormldium sp., Anacytis sp., Ulothxix sp. and
Uranema sp. No significant change in this association of species, with
respect to number of species and dominance, was noted after treatment with
the coagulating substance. However, each appeared to be in greater
abundance after treatment.
154
-------
2. Anaerobic Stabilization Pond Bacteriological Analysis—Table Al
and Figure A2 show the before arid after treatment effect on the pond and
bacteria density. In all cases bacterial numbers decreased from station
1 to station 7 (pond effluent). This is to be expected since bacterial
numbers are very high (especially coliform organisms) in raw sewage
and diminish as the wastewater passes through the plant. The magnitude
of decrease is not great in the pond and is not considered unusual.
However, there was a distinct increase in bacteria numbers after the
Fed and polyelectrolyte were added. The increase was by a factor
range of 5X to 8X at station 1 and 7X to 18X at station 7. The sudden
change in bacterial density after or during treatment is considered
significant. Therefore, since there was an Increase in density, hence
an increase in matabolism it is considered a beneficial change; more
saprobic biological activity increases the breakdown of non-refractory
organic matter into simple inorganic compounds and readily decomposable
bacteria cells.
3. Anaerobic Stabilization Pond Algal Analysis—Under anaerobic
conditions (absence of oxygen) one would expect to find none to a small
number of algal species. This was the case since only 6 species were
dominant and considered of value and autochthonous (produced in the
pond). They are: Chlorella vulgaris, C, ellipsoidea, Palmellococcus
protothecoides, P. miniatus, Nitzschia pales, and Pyrobotrys gracilis.
Nitzschia and Chlorella and common sewage algae and can temporarily
live heterotrophically, and the others need considerable research. The
importance of algae in this anaerobic stabilization pond that has a mean
BOD range of 218 to 492 mg/1 is probably not significant since the strong
anaerobic habitat restricts algal growth and restricts the number of
species that can tolerate anaerobiasis. This is in direct opposition
155
-------
to aerobic stabilization or oxidation ponds that are aerobic during the
day and become slightly anaerobic at night due to microorganism respiration,
BOD. Oxidation ponds have a 3 to 4 week detention period and rely largely
on algal photosynthesis to provide oxygen for aerobic bacterial metabolism.
Table A2 and Figure A3 show the abundance of algal cells for each station
for the four sampling periods. Algal numbers were always lowest at station 1
(clarified sewage influent to the pond) and generally increased to station
7 (pond effluent). Before treatment total algal populations ranged from
1,060/ml to 5,890/ml and averaged 2,720/ml, while one week after treatment
started algae ranged 2,060/ml to 7,490/ml and averaged 4,650/ml. This is
considered a minor fluctuation and represents a 1.7 fold increase after
treatment. However, two weeks later during treatment algal density
increased greatly ranging from 7,330/ml to 960,670/ml averaging 382,10Q/ml,
or a factor of 140X from before treatment values. On July 28, 1970,
Mr. Keith Larson notified me to check the algal bloom present in the
stabilization pond. Upon examination of the samples which recorded
1,800,100 cells/ml the same 6 species w,ere represented as listed earlier,
mostly Chlorella however. The fact that Mr. Larson has never noted algal
blooms in the stabilization pond since its construction in 19J53 makes
this bloom after treatment with FeCl3 and polyeletrolyte coagulating
agents highly significant. Since water temperature varied little
(Table A3), Secchi disc transparency remained about one foot and solar
radiation was not atypical, the possible direct or indirect effect of the
coagulating agents on the bloom is plausible. However, the bloom also"
coincided favorably with lower BOD values in the pond since Swift
ceased slaughtering operations and the lower pond BOD probably stimulated
algal growth. During slaughtering operations (1963-Nov., 19691, the
pond BOD average ranged 492 mg/1 from the influent to 218 mg/1 at the
effluent. Since December, 1969, to August, 1970, the average range was
156
-------
202 to 63 mg/1, repectively. Lowest values occurred during the bloom in
July, 1970, 22 mg/1 BOD. This climbed after the bloom occurred to the
August monthly average of 43 mg/1 BOD. The lowest pond BOD (22 mg/1)
probably also resulted from and stimulated algal growth and hence oxygen
production. During the bloom and to mid-August dissolved oxygen was
noted during day-light hours at the pond effluent (peaked at 2.5 mg/1)
to 5 August 1970). The samples taken by the sewage plant chemists
showed typical oxidation pond mid-day highs and early morning-late day
lows. No samples were taken at night, but they probably would show no
dissolved oxygen. The slight increase in algae was noted in the July 8
samples, when bacteria also increased greatly from the two before
treatment samples. Algae probably responded more slowly to the
environmental changes than the bacteria.
The reduction of colloidal particles and suspended solids by
coagulating agents should have reduced the turbidity in the pond, No
reduction was recorded since the Secchi disc transparency measurements
varied little (mostly 11-13 inches) throughout the sampling period
(Table A3). The pond is continuously mixed internally by upwellings of
soft organic matter brought on by trapped gas (C02, CH^, H2S) in the
bottom sediments. Reduction in turbidity would cause increased algal
bloom. It is probably related to the addition of coagulating agents.
Again, increased growth of microorganisms, algae and bacteria, is
beneficial and hastens the utilization and decomposition of non-refractory
organic matter.
Table A4 provides a systematic list of the biota found on surface stones
of the biolgical filters and from 0 to 7 feet of the anaerobic
stabilization pond during June and July, 1970.
157
-------
BIBLIOGRAPHY
Anon. 1970. Thirtieth Annual Report, Sewage Disposal Commission,
South St. Paul, Minnesota. April 1, 1969 to March 31, 1970. 33pp.
Hach Chemical Company, 1968. Water and Wastewater Analysis Procedures.
Hach Chemical Company. Catalog No. 10. Ames, la. 104 pp.
McNabb, C. D. 1969. Enumeration of freshwater phytoplankton concentrated
on the membrane filter. Limnol. Oceanogr. 5: 57-61.
Standard Methods. 1965. Standard Methods for the examination of water
and wastewater. A.P.H.A. 769 pp.
158
-------
GLOSSARY
AEROB (AEROBIC). Organisms that can prosper only when oxygen is present
more or less abundantly (Gr. Aer-air, bios=lif e) .
ALGAE (ALGA). Simple plants, many microscopic, containing chlorophyll.
Most algae are aquatic and may produce a nuisance when
environmental conditions are suitable for prolific growth.
ANAEROBE (ANAEROBIC) . Organisms which either by obligation or facultatively
thrive in the absence of oxygen (Gr. an-without) .
ASSOCIATION. An association in biology includes the entire organic P«pu-
latlon of a given habitat with two or more organism species
dominating the group.
AUTOCHTHONOUS. Arising in the biological system under consideration
(Gr. autos«self, same, chthon»lana; .
nourish) .
the bacterial breakdown of organic matter and oxidation of
certain compounds.
HETEROTROPH (HETEROTROPHIC) . Organisms that are dependent on organic
matter for food.
many fungi.
NON-REFRACTORY. Yield readily to treatment, easily oxidized and hrofcen
down.
chlorophyll in the presence of light.
SAPROBE (SAPROPHYTE). Any organism living on dead or decaying organic
matter.
159
-------
SIGNIFICANCE. Used only in connection with statistics, and in general,
it is used in connection with the rejection of a hypothesis and
observable variations that cannot be explained by the use of
a confidence interval.
TURBIDITY. The degree of opaqueness produced in water by suspended
particulate matter which in turn is generally responsible for
water color quality, and the concentration of substances,
if sufficiently high, determines the transparency of the water
by limiting the light transmission within it.
160
-------
TABLE 1A. The abundance of bacteria in the anaerobic stabilization pond
during June and July, 1970, for seven stations.
[ON
(PLATES)
1
2
3
Mean
1
2
3
Mean
1
2
3
Mean
t 1
2
3
Mean
> 1
f
2
3
Mean
61
x
2
3
Mean
7 1
r
2
3
Mean
17 June
59
56
56xl05
340
280
280
300x10^
86
118
108
104x10*
73
98
99 4
90x10*
102
75
66 4
81x10*
61
51
44
52x10*
45
40
38 4
41x10*
24 June
206
203
214
208x10*
93
88
99
93x10^
38
70
62
57x10
56
36
52
48x10*
225
180
300
235xl03
320
350
360
34 3x1 O3
150
134
160
148x1 O3
8 July
100
104
106
103xl05
70
80
85 5
78xl05
420
400
420
413xl04
250
380
400
343xl04
240
250
230
240x1 OT
420
350
270 L
346x10
300
310
165 ,
258x10'
22 July
208
120
110 ,
170x10-
160
150
135 t
148x10"
100
100
120 ,
106x10-
55
60
50 ,
55x10'
42
65
65 t
57x10-
380
390
400
390x10
290
270
280
280x10
161
-------
TABLE 2A- The abundance of algae in the anaerobic stabilization pond
during June and July, 1970, for seven stations.
STATION SPECIES OR GROUP
1
2
3
14.
5
6
7
Chlorophytes
Diatoms
Oscillatoria
TOTAL
Chlorophytes
Diatoms
Oscillatoria
TOTAL
Chlorophytes
Diatoms
Oscillatoria
TOTAL
Chlorophytes
Diatoms
Oscillatoria
TOTAL
Chlorophytes
Diatoms
Oscillatoria
TOTAL
Chlorophytes
Diatoms
Oscillatoria
TOTAL
Chlorophytes
Diatoms
Oscillatoria
TOTAL
17
& Others
rileyi
& Others
rileyi
& Others
rileyi
& Others
rileyi
& Others
rileyi
& Others
rileyi
& Others
rileyi
June
810
290
20
1120
2240
300
50
2590
4570
710
20
5300
2240
710
30
2980
5180
710
0
5890
3560
710
0
4270
5180
710
0
5890
ALGAL CELLS/ML
24 June
530
530
140
1200
810
370
200
1380
710
370
140
1220
630
370
60
1060
920
530
60
1510
710
630
140
1480
1280
710
200
2190
8 July
1690
370
0
2060
2680
630
0
3310
3640
450
0
4090
5510
630
10
6150
6950
530
0
7480
5510
530
0
6040
1540
1850
0
3390
22 July
6990
200
140
7330
43280
450
60
43790
139800
530
200
140530
299500
730
60
299930
522900
370
0
523270
698800
370
0
699170
960300
370
0
960670
28 July
____
____
-_-_
....
____
••.••>«•
___—
_-__
<^tWW«B
1800000
1000
0
1801000
162
-------
TABLE 3A. Secchi disc transparency and water temperature profiles in the
anaerobic stabilization pond during June and July, 1970 for seven stations,
June 17, 1970—1:30 p.m.
Depth (feet)
0
1
2
3
4
5
6
7
Secchi Disc (inches)
June 24, 1970—10:15 a.m.
0
1
2
3
4
5
6
7
Secchi Disc (inches)
July 8, 1970—9:30 a.m.
0
1
2
3
4
5
6
7
Secchi Disc (inches)
July 22, 1970—10:00 a.m.
0
1
2
3
4
5
6
7
Secchi Disc (inches)
STATION NUMBERS
(1)
76
76
76
76
75
75
74
74
12
74
74
74
74
74
74
74
74
12
74
74
74
74
74
74
74
74
12
76
76
76
76
76
76
76
76
11
(2)
75
75
75
75
74
74
74
74
13
74
74
74
74
74
74
74
74
12
74
74
74
74
74
74
74
74
11
76
76
76
76
76
76
76
76
11
(3)
75
74
74
74
74
74
74
74
12
74
74
74
74
74
74
74
74
12
74
74
74
74
74
74
74
74
11
76
76
76
76
76
76
76
76
12
AND TEMPERATURE °F
(4)
76
76
74
74
74
74
74
74
12
74
74
74
74
74
74
74
74
12
74
74
74
74
74
74
74
74
11
76
76
76
76
76
76
76
76
12
(5)
76
76
76
76
74
74
73
73
16
74
74
74
74
74
74
74
74
12
74
74
74
74
74
74
74
74
13
76
76
76
76
76
76
76
76
11
(6)
76
76
76
75
74
74
74
74
16
75
75
75
75
75
75
74
74
12
74
74
74
74
74
74
74
74
12
76
76
76
76
76
76
76
76
11
(7)
76
76
76
76
14
75
75
75
75
12
74
74
74
74
13
76
76
76
76
12
163
-------
TABLE ^A. Systematic list of the biota found on surface stones of the
biological (trickling) filters and from the surface to 7 feet in the
anaerobic stabilization pond during June and July, 1970.
GENERAL GROUP
1. BACTERIA
SPECIES OR GENUS
Sphaerotilus natans
Zoogloea ramigera
Begglatoa sp.
Thiothrix sp.
Chroma tium sp.
HABITAT
FS I/
FS
FS
FS
AP 2/
FUNGI
Fungi-Imperfecti
ALGAE
Chlorophyta (green a
Volvocales
Spondylomoraceae
Chlatnydomonaceae
Ulotrichales
Ulotrichaceae
Chlorococcales
Oocystaceae
Geotrichum candidum Link, ex Pers. FS
emend. Carmichael
Igae)
Pyrobotrys gracilis
Ch 1 atnydomonas sp.
Ulothrix sp.
Uronema sp.
(Korsch.)
AP
FS+AP
FS 3/
FS Zf/
Chlorococcaceae
Scenedesmaceae
Zygnematales
Desmidiaceae
Palmellococcus miniatus (Kutz.)Chodat AP
P. protothecoldes (Kruger)Chodat AP
cTTlorella vulgarls Beijerinck FS+AP
C. ellipToidea Gerneck AP
Ankistrodesmus falcatus (Corda)
Ralfs var. acicularis (A. Braun) AP
Chlorococcum sp. FS 5/
Scenedesmus acuminatus (Lag.) Chodat ~
var. minor G. M. Smith FS+AP
Closterium sp.
AP
I/ Filter stones on surface of biological filters (FS)
?/ Anaerobic stabilization pond (AP)
and 5/ sent to experts for identification, these particular
"~ species are not listed in the flora of this region and
are presently in liquid and agar culture for further
examination.
164
-------
SYSTEMATIC LIST OF THE BIOTA (Continued)—
GENERAL GROUP
SPECIES OR GENUS
HABITAT
Euglenophyta (euglenoids)
Suglenaceae Euglena gracllls Klebs
Myxophyta (Cyanophyta, blue-green algae)
Hormogonales (Oscillatoriales)
Oscillatoriaceae Oscillatoria rileyl Drouet
01. geminata Meneghin1
PKormidjjum sp.
Chroococcales
Chroococcaeceae Anacystis marina (Hansgirg) Drouet
and Dailey
Chrysophyta (yellow-brown algae)
Bacillariophyceae
AP
AP
PS
FS
FS
Pennales
Centrales
Nitzschla palea (Kutz.) W. Smith
HantzschTa amphioxus (Ehr.) Grun.
Navicula gregaria Donkin
Fragillaria sp^
Diatoma vulgare Bory
Surirella angustata Kytz.
Gomphonema olivaceum (Lyngbye)Kutz
Pinnularia gp.
Synedra ulna (nitzsch) Sir.
Cymatopleura solea (Brebisson) S.SmithAP
Caloneis amphisbaena (Bory) Cleve AP
Melosira' granulata CEhr.) Ralfs AP
Cyclotel'la MeneghTniana Kutz. AP
Stephanodiscus niagarae^ Ehr. AP
FS+AP
AP
FS
AP
AP
AP
AP
AP
AP
PROTOZOA (single celled animals)
Ciliata Colpoda sp._
Epistylis rotans Svee,
VorticefTa sp.
Buplotes sp.
Sarcodina
Stylonychia sp
Lionotus sp.
Valkampfia sp.
5. ZOOFLAGELLATES (flagella locomotion, no chlorophyll)
Bodo sp.
Oicomonas Ocellata
AntHophysa vegetans (O.F. Muller)
Stein
Gyromonas ambulans Seligo
FS
AP
AP
AP
AP
AP
AP
AP
AP
AP
165
-------
SYSTQiATIC LIST OF THE BIOTA (Continued) —
GENERAL GROUP SPECIES OR GWUS HABITAT
6. NH4ATODA (pseudocoelomate round worms)
unidentified nematode FS
7. ARTHROPODA
Insecta
Diptera
Psychodidae Psychoda alternata Say FS
Crustacea ""
Cladocera
Daphnidae Moina sp. AP 6/
6/ Observed during the late July algal bloom when dissolved oxygen was
cTetected in the anaerobic stabilization pond. The specimen was sent
to an expert for species identification since it was not listed in the
fauna of this area.
166
-------
SOUTH ST. PAUL, MINNESOTA, WASTEWATER TREATMENT PLANT
Mississippi
e*
ers
-influent after grit-
Pst -primary settling
P -pumping station
f -biological filfi
st -storage tank
fst -final settling tanks
cct-chlorine contact tank
e -treated effluent to river
0 feet 300
ANAEROBIC STABILIZATION
POND
AREA
VOLUME
DEPTH (maximum)
DEPTH (mean)
PERIMETER
27.4acres
240 acre-ft.
8.5ft.
6.5ft.
6100 ft.
sample stations
FIGURE
-------
^Tr-.crO 17 JUNE
24JUNE
10T
0 (feel) 450
1 i lalioni 2
1350 2250
456
FIGURE 2 A.
3150
7
O 28 JULY
22 JULY
17 JUNE
34JUN1
10
0 (l««0 450
1 stations 2
FIGURE 3 A.
168
-------
DESCRIPTION OF BIOTA ILLUSTRATIONS FOR PLATE I.
1. Phormidium sp. Common blue-green alga, shows several filaments,
sheaths cbnfluent, apical cells rounded to obtuse conical.
2. Zoogloea ramigera. Very abundant slime colonial bacteria covering
the entire filter stone, shows cells within a mucilage secreted
by Zoogloea, attached to a fungus filament.
3. Sphaerotilus natans. Uncommon on the filter stones observed,
shows many cells attached forming a filament.
4. Oscillatoria geminata Meneghini. Common glue-green alga, shows a
typical filament with constricted cross walls.
5. Uronema sj)A Common attached green alga, shows three filaments
wTEVpointed terminal cells, cup-shaped perital chloroplasts,
and basal attachment cells.
6, Anacystis marina (Hansgirg) Drouet and Dailey. Common blue-green
alga attached to filter stones, usually forming a dense bright
blue-green mass of cells, illustration shows a portion of a
colony with surrounding secreting mucilage.
7. Beggiatoa sp. Commonly observed motile sulfur bacteria, shows a
filament with sulfur globules.
8. Thiothrix sp. Commonly observed non-motile sulfur bacteria, ahows
~a "colony" attached to a fungus cell with sulfur globules.
9. Geotrichum candiduro Link, ex Pers. emend. Carmichael. Very abundant
filamentous Fungi-Imperfect! attached to filter stones, shows
two septate filaments with two exogenous conidia.
10. Ulothrix sp. Common filamentous green alga, shows single filament
with cup" shaped perital chloroplasts, hold-fast basal cell, and
smaller terminal cells when compared to long basal cells.
11. Chlorella vulgaris Beijerinck. Frequently observed coccoid alga,
shows three cells with perital cup chloroplasts, and reproductive
mother with four non-motile daughter cells.
12. Nitzschia palea (kutz.) W.Smith. Very common yellow-brown alga
(diatom) on surface filter stones, shows a single cell.
13. Chlorococcum sp. Very common coccoid green alga, responsible for the
"green appearance of surface filter stones, shows coenobia of^
nine cells with many motile zoospores in one, and small opening
of perital chloroplast with single pyrenoid.
lit. Chlorococcum sp. Shows a mature, bi-flagellated zoospore with eye
pigment spot.
15. Colpoda sp. Frequently observed ciliate protozoan, shows
cHaracteristic "bean" shape of this species.
16. Efcistylis rotans Svec. Very common sessile, stalked, protozoan,
Found attached to the posterior end of Psychoda mostly, shows
two animals on striated and segmented stalk.
169
-------
DESCRIPTION OP BIOTA ILLUSTRATIONS FOR PLATE I. (Continued)~
17. Chlamydomonas sp.^ Uncommon on the filter stones observed, shows a
single cell with two flagella, eye pigment spot, and perital
cup chloroplast.
18. Psychoda alternata Say. Very common "filter fly" larva, always
observed within the mucilage and fungus filaments, shows a
single segmented larva.
170
-------
PLATE 1. BIOLOGICAL FILTER BIOTA
If
171
-------
DESCRIPTION OF BIOTA ILLUSTRATIONS FOR PLATE II.
1, Gyromonas ambulans Seligo. Uncommon zooflagellate, shows single
cell with four flagella.
2. Surirella angustata Kutz. Frequently observed diatom, shows
single cell.~
3. Oicomonas ocellata. Common colonial zooflagellate, shows colony
with one flagella per cell.
4. Oicomonas ocellata. Shows two cells attached.
5. Oicomonas" ocellata. Shows single, more rounded cell.
6. ValkampfTa sp.Frequently observed amoeba, shows single cell
with large pseudopod or "foot".
7. Anthophysa vegetans (O.F. Muller) Stein. Common colonial,stalked,
zooflagellate, shows colony with unequal flagella, and stalk.
8. Lionotus sp. Frequently observed ciliate protozoan, shows
single cell.
9. Bodo sp. Commonly observed zooflagellate, shows single cell with
one anterior and one posterior trailing flagellum.
10. Colpoda sp. Common ciliate protozoan, shows single animal.
11. Buglena gracilis Klebs. Frequently observed flagellated alga,
shows single cell with long flagellum and plate-to-cup
shaped chloroplasts.
12. Vorticella sp. Uncommon stalked ciliate protozoan, shows animal
with stalk" attached to substrate.
13. Nitzschla palea (Kutz.) W. Smith. Most common diatom observed in
the pond, shows single cell with characteristic markings on
the frustule.
14. Scenedesmus acutninatus (Lag.) Chodat var. minor G. M. Smith.
Frequently observed green alga, shows characteristic attachment
of four cells.
15. Pyrobotrys gracilis (Korsch.) Commonly observed colonial green
alga, shows typical colony of about 16 cells, bi-flagellated.
J^* Pyrobotrys gracilis (Korsch.) Incorrectly called Chlorobrachis
gracilTima but is a zygote of Pyrobotrys, shows four flagella
and torpedo-shape.
17. Pyrobotrys gracilis (Korsch.) Full grown zygote without lateral
arms.
18. Palmellococcus protothecoides (Kruger) Chodat. Common coccoid
green alga shows oval shape of cells and lobes of perital
chloroplast.
19. Palmellococcus miniatus (Kutz) Chodat. Common coccoid green
~alga, shows spherical shape of cells and lobes of perital
chloroplast.
20. Buplotes sp. Frequently observed ciliate protozoan, shows
characteristic shape of the animal and cirri-like "feet".
21. Btiplotes sp. Side view, shows how this animal can walk with its
cirri, or modified cilia.
172
-------
DESCRIPTION OF BIOTA ILLUSTRATIONS FOR PLATE II. (Continued) —
22. Euplotes sp. Cross section of the animal showing dorsal ridges.
23. OscillatorTa rileyi Drouet. Also keys out to Schizothrix
calcicoTa (Agardh) Gomont from Drouet's latest revision.
Frequently observed highly gas vacuolated single planktonic
filaments, shows single filament.
2^. Closterium sp. Uncommon destnid, shows cresent shaped cell with
two chlorop1as ts.
25. Chiore!la vulgaris Beijerinck. Common coccoid green alga, shows
single cells with perital cup chloroplasts.
26. Chlorella ellipsoidea Gerneck. Common coccoid green alga, shows
"oval shape of cells with perital cup chloroplasts.
27. Chlamydomonas sp. Uncommon bi-flagellated green alga in the pond,
" shows single*"cell, eye pigment spot, and perital cup chloroplast.
28. Chromatium sp. Commonly observed motile sulfur bacteria, shows
single cells with flagella and sulfur globules.
29. Ankistrodesmus falcatus (Corda) Ralfs var. acicularis (A. Braun).
Frequently observed green alga, shows long cell with pointed
ends.
173
-------
PLATE 11. ANAEROBIC STABILIZATION POND BIOTA
174
-------
APPENDIX B
EXPLANATION OF THE SOURCE OF WASTE FERRIC CHORIDE SOLUTION
USED IN FLOCCULATION TESTS AT THE SOUTH ST. PAUL SEWAGE TREATMENT PLANT
JULY, 1969 to SEPTEMBER, 1969
This ferric chloride solution waste is generated at Buckbee Hears
Manufacturing Company in St. Paul during the dissolving of steel
sheets to make television screens, electric shaver heads, steel sieves,
and other items where small uniform high density holes in steel plate
are needed.
In the process for making television screens (which, is typical
for the other items), the sheet metal to be used is coated with a
substance resistant to ferric chloride and the image of the television
screen with the millions of holes is printed with heat and light on
this coating. This printing step removes the coating for each, hole
down to bare metal. The sheet metal is then run through a strong
146°F ferric chloride solution fortified with muriatic acid.
A continuous testing of the solution is carried out and as the ferrous
iron content gets above certain limits, chlorine gas is added along
with water. During testing, if certain conditions prevail, some
muriatic acid is also added; these additions bring up the ferric
chloride content to the proper level for uniform rapid etching. This
addition of iron (the iron etched away from the sheets of steell the
muriatic acid to furnish C12 ions, the chlorine gas and the water keeps
converting the raw iron to ferric chloride and thus keeps increasing
the volume to a point that several hundred gallons of excess ferric
chloride waste is being generated daily. This waste is then pumped
over to other etching tanks for etching stainless steel sheet (nickle,
chrome, steel) and cooper plate. Again, when the ferric chloride is
spent in these etching tanks, it is dumped into large holding tanks
for ultimate disposal. Due to the large size, a fairly uniform waste
product is produced.
175
-------
At the present time, this ferric chloride waste is being hauled by
tank truck to the Metropolitan Sewage Treatment Plant to be utilized
as a supplemental supply of ferric chloride sludge conditioning. This
solution has an active FeCl3 content of approximately 45% (See analytical
data of representative sample tested by the Twin City Testing Laboratories)
At present, as far as can be determined, there are only two other
sources of this waste material and they both are manufacturers of tv
screens. However, there are places such as the branch office in Chicago
of the company from which we get our testing material where the demand
for ferric chloride etching at present is insufficient to justify
putting in the chloride rejuvenating equipment. In these cases the
spent ferric chloride is a serious waste problem. Since the pR of
this waste is low enough to be very corrosive to sewers, in some
communities these wastes are being prohibited entrance to the city sewers.
It is conceivable that now that this method of etching holes and
slots in steel has been developed to exact a science that many other
items where minute holes are made in metal will be manufactured in
this manner which will increase the amount of this waste and therefore
its availability.
176
-------
645-3601
TWIN CITY TESTING AND ENGINEERING LABORATORY. INC.
REPORT OF:
PROJECT:
REPORTED TO: Sewage Disposal System
South St. Paul, Minnesota 55075
(6) Attn: Mr. K. D. Larson
ENGINEERS AND CHEMISTS
662 Cromwell Avenue - St. Paul, Minn. 55114
CHEMICAL EXAMIHATI08
DATE: September 19, 1969
FURNISHED BY:
COPIES TO:
LABORATORY No. 2-10353
SAMPLE IDEHTIFICATIOH:
Ferric Chloride Solution
TESTS AHPRESULTSi
Ferric Chloride as Fed (percent)
Nickel as Ni (percent)
Manganese as Mn (percent)
Molybdenum as Ho (percent)
Calcium as Ca (percent)
Cobalt as Co (percent)
Copper as Cu (percent)
44.90
0.24
Less than 0.01
0.03
Hil
0.04
0.0002
REMARKS: Sample received September IS, 1969.
MUTUAL PROTECTION TO CLIENTS THE PUBLIC AND OURSELVES. ALL REPORTS ABE SUBMITTED /.» THE CONFIDENTIAL PROPERTY OF CLIENT*. AND AUTHOR-
- IUi ^.^T^.0.! o. nATeuENTS CONCLUSIONS OH EXTRACTS FROM OR REGARDING OUR REPORTS IS RESERVED PEMPIM5 OUR WRITTEN APPROVAL
Laboratory, Inc.
177
-------
APPENDIX C
SLUDGE CHARACTERISITICS
In November and December.of 1969 chemical treatment tests were
made on separated industrial and domestic wastewater received at the
South St. Paul sewage treatment plant using a dual system of ferric
chloride and Dow A-23. During this time some sludge studies were run
to determine if the characteristics of the sludge produced by the
chemical treatment process were changed in any way.
The characteristics of sludge tested were percent moisture, percent
volatile solids, heat of combustion, filterability, and digestion
characteristics.
Sludge samples were taken of sludge from both the test and control
tanks. Grab samples of sludge were taken several times a day on
those days when sampling was being done. Samples of the test and
control tank were taken at approximately the same length of time after
pumping of the tank had started. Sludge samples were taken from
a pit from which the sludge was pumped. The sludge was pumped
from the bottom of the pit and flowed into the pit through, a pipe
near the top. Samples were taken of sludge as it fell from the pipe
into the pit. A large enough number of grab samples of sludge were
taken to average out individual differences between the test and
control tanks and produce a reliable average.
Percent moisture, percent volatile and filterability tests were
run on grab samples. Heat of combustion and digestability were run
on composite samples made of equal proportions of grab samples taken
for the day on which the test was run.
178
-------
The percent moisture test was made by drying an approximate
100 g sample of sludge overnight at 105°C in a drying oven. Volatile
solids were run by igniting the dried sludge sample at 6QO°C.
until all the organic matter was volatized.
Heat of combustion tests were run using a Parr peroxide bomb
colorimeter. Each sample on which the heat of combustion was run
was prepared from one liter of the composite sample taken from
total composite sample for the day after thorough mixing. The one
liter sample was dried in a 105°C oven. It was then ground to
a fine powder by mixing it dry in a'blender'run'at high speed.
An approximately sized sample of this was taken and weighed for the
actual test.
Digestion characteristics of the sludge were tested using scale
model digesters. Two digesters were used for sludge from the test tank
and two digesters were used for sludge from the control tank. Each.
digester unit consisted of two stages.
The individual stages consisted of five gallon polyethylene bottles
The first stage was stirred continuously by an electric stirring
device to provide thorough mixing throughout. The second stage was
quiescent which allowed the sludge to separate from the supernatant.
The first stage was provided with a sludge inlet tube, a sludge
withdrawal tube, a mechanical stirring device with a water seal, and
a gas line to the gas collection unit. This unit consisted of two
three gallon glass carboys. The first carboy was filled with water.
The gas from the digester entered this and displaced water into the
second one where its displaced volume could be measured thus affording
an accurate method for measuring the volume of gas produced. Both th£
179
-------
first and second stages of all four digestores were immersed in a
constant temperature water bath held at 35° +_ 1.0°C. All four
digestors were identical in design.
The digesters were run for a period of over a year on primary
sludge from wastewater to which no chemicals had been added. This
produced a state of equilibrium of constant digestion, gas production
and digestor pH. Determined by the digesters tests were cubic feet
of gas produced per pound of volatile material destroyed and percent
of the volatile material which was digested. Parameters measured
directly were gas produced by each stage, percent moisture and percent
volatile solids of the sludge added to the first stage and sludge
removed from the first and second stages as well as the supernatant
removed from the second stage. The pH was also measured on the
influent and the effluent of both stages and the supernatant to insure
consistancey of operation. See Tables C66-C72.
The results of the percent moisture analysis for the sludge
produced by the industrial wastewater are shown in Table C61. The
average percent moisture for the test tank sludge was 96.2%, while the
control tank sludge showed a percent moisture of 95.5%. The average
percent volatile solids of the test tank sludge was 76.2% compared to
76.1% for the control tank sludge. Apparently the addition of chemicals
to separate industrial wastewater had no effect on the percent moisture
or percent volatile solids of the sludge as it was removed from the
settling tanks.
The percent moisture and percent volatile solids content of the
sludge produced during tests on domestic wastewater are shown in
Table C62. The average percent moisture of the sludge removed from the
test tank was 95.6% while the average percent moisture of the sludge
removed from the control tank was 94.5%. The average percent volatile
180
-------
solids of the sludge removed from the test tank was 80.7%, while the
average percent volatile solids removal from the control tank was
82.8%. Again on tests on separated domestic wastewater there was no
large difference indicated between sludge samples taken from the test
and control tanks in terms of percent moisture or percent volatile
solids.
The results of heat or combustion tests are given in Table C63.
During the period of testing of industrial wastewater the average
heat of combustion of sludge taken from the test tank was 9,922 Btu
per pound while the value of sludge taken from the control tank
was 9,102 Btu per pound. Considering the number of samples run and
variation between samples, the average of the results are of the
same order.
Filterability of the sludge was tested using the buchner funnel
method. Two hundred ml of sludge was filtered through, a number one
Whatman filter paper in a nine cm buchner funnel. The volume of
filtrate and the length of time required to draw enough, water out
of the sample to break the vacuum were measured. The percent
moisture remaining in the filter cake was determined. The results
of filterability tests on sludge samples taken during industrial
wastewater tests are given in Table C64. There was no significant
change in the average ml of filtrate for the test and control tank
sludges nor was there any significant difference In the percent
moisture remaining in the filter cake of the test and control tank
sludges.
There was observed, however, a significant difference in the time
of filtration between the test and control tank sludges. Contrary to
what might be expected, the filtration time of the sludge taken from
the test tank was substantially greater than that taken from the
181
-------
control tank. The average filtration time for the test tank sludge
was 53 mintues, while the average filtration time of the control tank
sludge was 24 minutes.
The results of the filtration tests run on sludge taken during
the testing of chemical treatment on domestic wastewater are given
in Table C65. Again there was no large difference in the volume of
filtrate. The filtration time for the domestic sludge from the test
tank was 33 minutes and the filtration time for the sludge from the
control tank was 29 minutes. This is not as large a difference as
observed in tests of chemical treatment of industrial wastewater and
is not as meaningful an indication of poorer filterability
characteristics. An explanation of the fact that a much longer time
was required for the filtration of test tank sludge as compared to the
control tank sludge during the industrial period was that ferric
chloride was added to the wastewater at a dosage of 50 mg/1 during
the test on industrial waste while during the greater part of the tests on
the domestic waste only 20 mg/1 of ferric chloride was being added to
the wastewater.
The data collected for sludge digestion tests is shown in Tables
C66 through C72. A summary of the results of the sludge digestion
tests for sludge taken during the industrial test period are shown
in Table C66.
The volume of gas generated per pound of volatile solids
destroyed for the test tank was somewhat less than that generated
per pound of volatile solids destroyed for the control tank.
Considering the limited amount of data collected and the fact that
slightly greater amounts of volatile solids were added to the
182
-------
control digesters these results are felt to be somewhat inconclusive,
but they do indicate the results of the chemicals did not have a
substantial detrimental effect on the operation of the digesters.
In terms of percent volatile material destroyed the digesters
operated on sludge from the control tanks showed a reduction of
68.3% of volatile solids. Again considering that a somewhat
higher amount of volatile solids were added to the control digestors
this lower pecent reduction is somewhat inconclusive. Again the
data indicates the digestores were not substantially inhibited.
A summary of the results of the digestor tests on sludge
obtained during trials of domestic wastewater are given in Table C67.
The volume of gas production per pound of volatile materials destroyed
was greater for sludge taken from the test tank in this case.. Again,
considering the limited number of test days it was not felt the. results
were conclusive but again were indicative, that the chemicals did not
have an inhibitory effect on the digestion process.
In terms of percent of volatile materials destroyed the test
sludge digestors and contol sludge digestors. showed very little
difference. The test tank digester showed a reduction in volatile
solids of 74.2% while the control tank digester showed a reduction of
69.8% volatile solids destroyed. Again this indicated the presence of
chemicals in the sludge did not inhibit digestion.
Table C68 gives the characteristics of the sludge added to the
digestors. Tables C69 through C72 give characteristics of the sludge
and supernatant withdrawn form each, of the digestors. While this data is
limited, a study of the data indicates that there was no drastic upset
of any of the digestors due to the use of chemicals.
183
-------
The volume of sludge produced during chemical treatment of
combined wastewater was approximately 50% greater than produced
without the addition of chemicals. When no chemicals were added
120,000 gallons per day of primary sludge was produced in the three
sedimentation tanks while with the addition of chemicals about
180,000 gallons per day of sludge were produced.
The following conclusions can be drawn from the tests on sludge;
(1) No change was made in the percent moisture and percent volatile
solids contents of the waste by the addition of chemicals. C2) The
heat of combustion was not substantially effected by the use of
chemicals. (3) Addition of ferric chloride to the industrial waste-
water in concentrations of 50 mg/1 substantially slowed the rate of
filtration of the sludge, however, the addition of 20 mg/1 of ferric
chloride to domestic wastewater had slight effect on the filtration
characteristics. (4) The data from the sludge digestion tests were
somewhat inconclusive but it did indicate that digestion was not
substantially inhibited , nor were the digestors drastically upset by
the addition of sludge from sewage treated with the dual system of
ferric chloride and organic polyelectrolyte. C5) Sludge volume was
increased by about 50% by the addition of chemicals.
184
-------
Table 61C
Sludge Moisture And Volatile Solids Content
November Industrial Waste
Percent Moisture
Percent Volatile
Nov. 3
Nov. 4
Nov. 5
Nov. 6
Nov. 12
Nov. 13
Nov.
Nov. 18
Nov. 19
Test
98.7
96.4
97.4
98.2
94.2
96.9
98.0
95.7
95.2
97.5
97.3
97.4
94.3
94.4
97.3
93.1
95.8
94.7
99.1
95.7
95.0
99.0
95.5
98.9
99.3
93.5
96.3
96.9
93.7
Control
94.2
94.4
98.6
95.0
96.4
95.2
94.1
94.3
91.8
94.7
94.3
93.6
95.2
95.6
95.2
95.3
98.5
93.9
95.0
96.7
99.6
99,5
94.5
99.6
99.3
92.6
94.4
94.3
92.4
Test
73.8
80.7
72.6
66.8
77.8
79.4
75.0
86.3
81.0
83.8
77.0
76.6
80.2
83.7
77.4
78.8
90.0
76.6
78.7
65.8
77.6
75.0
64.3
77.7
68.9
56.8
67.9
79.8
77.2
75.3
Control
81.7
83.6
77.0
74.0
82.7
83.3
83.4
76.7
85.1
86.0
85.4
82.2
82.7
83.0
79.8
82.7
81.3
69.3
80.5
82.3
74.5
48.8
53.5
79.4
50.0
55.0
65.6
82. 8
81.9
70.0
185
-------
Nov. 20
Nov. 25
Mean
Standard
Deviation
Table 61 (Continued)
Percent Moisture
Test
93.6
97.0
96.0
94.2
95.8
94.6
96.1
Control
99.4
95.5
94.8
94.5
94.2
93.8
97.6
96.2
11.77
95.5
12.14
Percent Volatile
Test
76.0
76.2
77.8
66.3
79.0
81.3
79.3
Control
57.8
81.9
80.6
72.5
83.2
80.4
76.8
76.2
16.62
76.1
+10.44
186
-------
Table 62C
Sludge Moisture And Volatile Solids Content
December Domestic Waste
Percent Moisture
Percent Volatile
Date
1969
Dec. 1
Dec. 9
Dec. 10
Dec. 11
Dec. 16
Dec. 17
Dec. 23
Mean
Standard
Deviation
Test
96.4
98.8
96.2
98.7
94.7
95.6
97.7
92.7
93.8
98.4
98.5
96.7
91.8
92.7
93.2
97.4
92.4
99.2
94.7
93.4
94.7
Control
95.5
97.1
95.2
93.4
94.2
93.5
93.4
99.1
94.3
95.5
94.8
95.1
91.3
92.4
92.2
92.7
93,6
92.4
99.8
95.1
94.5
Test
75.2
76.2
80.2
76.6
84.0
81.1
77.2
83.4
81.3
75.0
78.2
81.1
89.9
91.6
83.4
82.2
85.5
76.3
76.0
85.0
85.8
Control
79.2
80.5
84.8
84.3
84.7
84.9
87.6
73.8
80.7
83.9
85.7
85.3
90.2
85.6
85.4
85.4
86.9
85.8
53.0*
84.7
86.3
95.6
12.42
94.5
±2.14
80.7
±4.45
82.8
13.80
* Excluded from calculations
187
-------
Table 63C
Calorific Tests
BTUAB
Industrial Tests Sludge
Date - 1969
November 18
November 19
November 20
November 26
Mean
Standard Deviation
Date - 1969
December 9
December 11
December 16
December 17
December 23
Mean
Standard Deviation
Test Tank
9,413
10,216
9,676
10,385
9,922
* 455
Domestic Waste Sludge
Test Tank
10,093
11,448
11,954
11,139
12,785
11,484
t 971
Control Tank
8,683
9,446
7,496
10,783
9,102
± 1,376
Control Tank
11,074
11,712
11,059
11,264
11,650
11,352
i 312
188
-------
Table 64C
Buchner Funnel Tests On Sludge
Industrial Wastewater Tests
A-23 And FeCl3
Date
1969
Nov. 3
Nov. 4
Nov. 5
Nov. 6
Nov. 12
Nov. 13
Nov. 18
Nov. 19
Nov. 20
ML
Of
Filtrate
Test
193
189
177
177
191
175
160
191
178
189
159
163
188
181
167
197
150
176
163
173
185
151
188
165
153
182
154
174
Control
146
163
154
175
168
167
145
168
165
155
185
174
173
171
156
168
182
163
158
155
163
160
171
162
179
167
159
187
Filtration
Minutes
Time
Test Control
105
- 84
65
64
22
62
60
55
41
21
72
55
36
58
83
43
49
37
58
23
32
43
20
17
23
25
20
28
25
23
32
21
21
35
19
17
20
15
17
13
15
26
25
73
Filter Cake
%
Test
66.0
73.0
71.6
69.8
66.0
69.2
76.7
67.8
74.5
59.5
72.5
70.5
56.0
66.4
70.6
60.2
85.5
68.7
69.9
70.8
69.6
76.1
64.6
72.2
70.0
57.2
76.0
66.8
Moisture
Control
71.6
67.6
72.2
65.0
68.0
62.6
68.0
70.0
70.5
72.1
62.2
67.3
64.2
65.9
72.0
66.7
66.2
71.0
66.3
75.6
73.0
63.4
63.4
73.8
69.6
69.8
70.0
64.2
Mean
Standard
Deviation
175
±14.0
166
±10.6
53
121.2
24
±12.2
69.2
±6.1
68.2
±3.7
189
-------
Table 65 C
Buchner Funnel Tests On Sludge
Domestic Wastewater
Treatment FeCl And A-23
ML
Of
Filtrate
Date
1969
Dec. 1
Dec. 9
Dec. 10
Dec. 11
Dec. 16
Dec. 17
Test
174
198
178
192
169
170
188
153
154
196
194
184
139
162
156
189
159
199
Control
171
187
176
152
160
160
163
194
156
170
165
167
148
148
148
156
164
144
Filtration
Minutes
Time
Test Control
70
57
36
12
60
58
38
15
24
20
25
28
13
8
38
30
25
43
24
36
34
40
51
20
27
31
30
30
36
19
9
23
39
38
21
26
Filter Cafce
%
Test
71.0
62.8
73.0
54.8
68.1
70.3
60.4
78.6
76.0
66.0
49.5
64.0
71.2
64.2
68.8
59.5
68.3
66.6
Moisture
Control
71.6
65.2
67.0
61.8
72.2
70.0
60.7
61.8
75.0
72.3
72.2
73.2
57.8
72.8
69.8
69.8
69.9
71.9
Mean
Standard
Deviation
175
162
i 18.3 ± 13.1
33
t 18.2
29
± 9.8
66.2
± 7.3
68.6
± 5.0
190
-------
Table 66C
Gas Production
November Industrial Waste
Digester
Test Sludge
ft 1
Test Sludge
ft 2
Control Sludge
ft I
Control Sludge
tt 2
Ave. Volume
L/Day
A 13.5
B 3.4
A 16.5
B 1.4
A 16.7
B 4.2
A 19.0
B 7.0
Cu. Ft. For
Month
14.3
3.6
17.5
1.5
17.7
4.5
20.2
7.4
Average
Average
Cu. FtAb Of Volatile
Material Lost In
Two Stages
17.5
17.4
20.6
25.6
Test 17.4
Control 23.1
Volatile Material Destroyed
Digestor
Test # 1
Test ft 2
Control tt 1
Control t 2
Vol. Solids
Grams Added
575
575
715
715
Vol. Solids Vol. Solids
Grams Removed Grams Destroyed
113
82
227
227
462
493
488
488
%
Red.
80.5
85.7
68.3
68.3
Average Of Test 83.1
Average Of Control 68.3
191
-------
Table 67C
Gas Production
December Domestic Waste
Digestor
Test Sludge
tt 1
Test Sludge
tt 2
Control Sludge
ft I
Control Sludge
tt 2
Average Of Test
Average Of Control
Ave. Volume
L/Day
A 10.3
B 4.5
A 19.5
B 2.9
A 11.6
B .7
A 21.1
B 4.9
Cu. Ft* For
3 Week Period
8.4
3.6
15.9
2.4
9.4
.6
17.2
4.0
29.4
18.7
Volatile Material Destroyed
Cu. FtAb. Of Volatile
Material Lost In
Two Stages
24.7
34.1
13.4
24.1
Digestor
Test ft 1
Test it 2
Control ft 1
Control ft 2
Vol. Solids
Grama Added
310
310
526
526
Average Of Test
Average Of Control
74.2
69.8
Vol. Solids
Grams Removed
90
67
189
128
Vol. Solids %
Grams Destroyed Red.
220 70.1
243 78.4
337 64.1
398 75.6
192
-------
Table 68C
Raw Sludge To Digesters
November Industrial Wastewater
Test Tank Sludge
Nov.
1969
13
19
26
Ave.
Percent
Moisture
96.4
97.9
95.5
96.6
Percent
Volatile pH
78.3 6.02
72.6 6.32
80.0 5.96
77.0
Control Tank Sludge
Percent
Moisture
95.4
96.8
95.5
95.9
Percent
Volatile
80.3
76.6
80.7
79.2
pH
6.21
6.43
5.95
December Domestic Wastewater
Test Tank Sludge
Dec.
1969
2
10
18
Ave.
Percent
Moisture
97.0
96.8
95.8
96.5
Percent
Volatile pH
77.0 6.12
80.8 6.20
83.7 6.00
80.5
Control
Percent
Moisture
96.2
93.5
93.2
94.3
Tank Sludge
Percent
Volatile
80.8
85.0
85.6
83.8
PH
6.20
5.81
5.80
193
-------
VD
1st Stage Sludge
Table 69C
Sludge Digesters
Test Tank Sludge
Industrial Wastewater
Digestor ff I
2nd Stage Sludge
2nd Stage Supernatant
Nov.
1969
13
19
26
Ave.
Nov.
1969
13
19
26
Ave.
Percent
Moisture
97.7
98.7
97.9
98.1
1st Stage
Percent
Moisture
99.4
98.8
97.8
98.7
Percent
Volatile
72.8
64.2
72.2
69.7
Sludge
Percent
Volatile
51.5
56.0
60.8
56.1
pH
6.89
7.29
6.90
PH
7.50
7.42
7.53
Percent
Moisture
97.1
97.9
96.9
97.3
Digester #
2nd Stage
Percent
Moisture
98.2
97.7
97.2
97.7
Percent
Volatile
64.4
61.6
66.4
64.1
2
Sludge
Percent
Volatile
60.0
60.2
59.7
59.9
pH
7.22
7.36
7.12
pH
7.39
7.50
7.52
Percent
Moisture
99.6
99.2
99.6
99.5
2nd Stage
Percent
Moisture
99.6
99.6
99.6
99.6
Percent
Volatile
51.0
50.0
43.8
48.3
Supernatant
Percent
Volatile
47.0
36.5
29.5
37.7
pH
258
7.53
7.60
PH
7.50
7.50
7.61
-------
Ul
1st Stage Sludge
Table 70C
Sludge Digesters
Control Tank Sludge
Industrial Wastewater
Digestor # 1
2nd Stage Sludge
2nd Stage Supernatant
Nov.
1969
13
19
26
Ave.
Nov.
1969
13
19
26
Ave.
Percent
Moisture
95.7
98.9
98.8
97.8
1st Stage
Percent
Moisture
96.6
97.8
97.3
97.2
Percent
Volatile
73.2
68.5
65.8
69.2
Sludge
Percent
Volatile
72.8
69.2
73.6
71.9
pH
7.29
7.27
7.25
PH
7.03
7.30
6.98
Percent
Moisture
95.8
95.8
97.5
96.4
Digestor ff
2nd Stage
Percent
Moisture
96.2
96.9
96.6
96.6
Percent
Volatile
71.8
65.0
62.5
66.4
2
Sludge
Percent
Volatile
66.3
65.8
65.6
65.9
pH
7.33
7.46
7.50
pH
7.15
7.00
7.38
Percent
Moisture
95.4
99.5
99.6
98.2
2nd Stage
Percent
Moisture
98.2
98.9
97.2
98.1
Percent
Volatile
66.7
35.0
43.9
48.5
Supernatant
Percent
Volatile
65.6
60.0
66.7
64.1
pH
7.50
7.50
7.62
PH
7.41
7.32
7.30
-------
1st Stage Sludge
Table 71C
Sludge Digestors
Domestic Wastewater
Test Tank Sludge
Digestor # 1
2nd Stage Sludge
2nd Stage Supernatant
Dec.
1969
2
10
18
23
Ave.
Dec.
1969
2
10
18
23
Ave.
Percent
Moisture
99.2
97.4
96.3
98.3
97.8
1st Stage
Percent
Moisture
99.4
99.6
99.4
99.1
99.4
Percent
Volatile
47.0
75.3
81.3
73.4
69.3
Sludge
Percent
Volatile
44.8
43.0
43.8
52.7
46.1
pH
7.40
6.83
6.50
7.35
pH
7.47
7.70
7.60
7.52
Percent
Moisture
95.0
95.7
97.9
96.2
Digestor #
2nd Stage
Percent
Moisture
95.6
97.3
95.6
98.5
96.7
Percent
Volatile
63.4
56.5
63.5
61.1
2
Sludge
Percent
Volatile
55.4
56.7
61.7
47.3
55.3
pH
7.30
7.20
7.58
PH
7.53
7.62
7.40
7.60
Percent
Moisture
99.5
99.4
99.2
99.3
99.5
2nd Stage
Percent
Moisture
99.6
99.6
99.5
99.6
Percent
Volatile
40.5
44.5
53.5
54.0
48.1
Supernatant
Percent
Volatile
32.5
38.4
39.0
36.6
PH
7.62
7.71
7.70
7.61
PH
7.56
7.67
7.49
7.60
-------
1st Stage Sludge
Table 72C
Sludge Digestors
Domestic Wastewater
Control Tank
Digestor 0 1
2nd Stage Sludge
2nd Stage Supernatant
Dec.
1969
2
10
18
23
Ave.
Dec.
1969
2
10
18
23
Ave.
Percent
Moisture
99.1
95.3
95.2
96.0
96.4
1st Stage
Percent
Moisture
98.6
97.8
96.2
97.3
97.5
Percent
Volatile-
55.0
78.7
70.8
70.3
68.7
Sludge
Percent
Volatile
67.2
77.8
77.4
74.6
76.2
pH
7.53
6.85
7.43
7.61
pH
7.20
7.02
7.00
7.32
Percent
Moisture
95.1
94.5
94.8
95.8
95.1
Digestor #
2nd Stage
Percent
Moisture
94.0
98.9
96.5
96.3
96.6
Percent
Volatile
63.0
67.6
72.2
70.2
68.2
2
Sludge
Percent
Volatile
59.8
61.6
66.2
71.0
64.6
pH
7.45
7.32
7.50
7.55
PH
7.39
7.41
7.50
7.40
Percent
Moisture
99.5
99.6
98.5
98.5
99.0
2nd Stage
Percent
Moisture
99.5
99.5
97.9
98.9
98.7
Percent
Volatile
34.5
43.0
65.1
65.4
52.0
Supernatant
Percent
Volatile
54.6
49.0
70.8
63.5
59.5
pH
7.61
7.70
7.63
7.62
pH
7.49
7.50
7.39
7.42
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
I h Rep,
w
4. Title Evaluation of Polymeric Clarification of Meat-Packing -5- *-
and Domestic Wastewaters ; $
s.
Larson. Keith D. & Douglas A. Maulwurf
9. Organization
Metropolitan Sewer Board
South St. Paul, Minnesota
*Apr1l, 1974
' r«tmin OrgHK "t
IV. Project No.
12130 EKK
12. Sponsoring Organization
IS. Supplementary Notes
U.S. Environmental Protection Agency
No. EPA-660/2-74-020» April
13. Type < ' Repoi znd
Period Covered
is. .Abstract
Laboratory tests were conducted to determine which system of chemicals
would be most effective on the combined packinghouse and domestic waste. A dual
system of chemicals was found which was effective in the treatment of this combined
waste. This was a combination of ferric chloride and an anionic polyelectrolyte. This
system was effective in forming a floe which would settle out under the dynamic
conditions of the overloaded primary sedimentation tank. It was demonstrated that
treatment with this system could effectively reduce suspended solids in the efflunet
of the primary sedimentation tank over what could be achieved without the use of
this dual system. This was accomplished by running: (1) a parallel system of identical
test and control tanks and, (2) full plant scale investigation for both test and control
periods, kept as identical as possible.
When full plant scale tests were run, laboratory data was collected from various
sampling points throughout the plant during the test and control periods. -A significant
reduction of BOD and suspended solids was obtained in the primary sedimentation tanks an
a change in efficiency was not observed on the trickling filters resulting in an overall
reduction in these parameters in the effluent from the secondary sedimentation
tank. The cost of chemically treating the combined 10 mgd of wastewater would be
approximately $45 per million gallons and would be less for strictly domestic wastes.
37a. Descriptors
*Polymeric clarification, *Polyelectrolytes, Packing plant wastes, Flocculation,
Chemical Treatment of Wastewater, Ferric Chloride Treatment.
I7b. Identifiers
*Polyelectrolyte Treatment, Packinghouse Wastewater.
17c. COWRRFieldti Group
IS. Availability
]$. ST -vrity C '•ssa.
(Kvpoit)
20- Security Ctess,
(P ?«J
21. it
------- |