iPA-660/2-74-055
June 1974
Environmental Protection Technology Series
'•'- '..-.•. v'i>'^IJfe!4iSlite^«^i^lSi^
••'Carbon-ate
I
55
\
\
LU
CD
Office of Research and Development
U.S. Environmental Protection Agency
Jgtshington, D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
Environmental Protection Agency, have been qrouped Into five
series. These five broacl 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.
This report has been reviewed by the Office of Research and
Development. Approval does not signify that the contents
necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
For sale by the Superintendent of Documents, U.S. Government Printing Office
Washington, D.C. 20402 - Price $1.90
-------
EPA-660/2-7^-055
June
PHYSICAL-CHEMICAL TREATMENT OF MUNICIPAL WASTES
BY RECYCLED MAGNESIUM CARBONATE
By
A. P. Black
A. T. DuBose
R. P. Vogh
Grant #12130 HRA
Program Element 1BB036
Roap/Task 21 AZV-21
Project Officer
R. P. Stringer
U.S. Environmental Protection Agency
Atlanta, Georgia 30309
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 204&0
-------
ABSTRACT
The applicability to municipal wastes of the recently discovered
magnesium carbonate-lime water treatment process has been investigated.
A sixteen-month laboratory study was conducted and was followed by an
eight-month pilot plant study. Four wastewaters with COD values
varying from 200 to 1,500 mg/£ were examined. Bench-scale coagulation
studies designed to compare the effect of added MgC03 with treatment
by lime only showed a 0%-30% greater reduction in effluent COD
residuals. Color and turbidity reduction by the magnesium-plus-lime
process averaged 50%-85% greater when compared to treatment by lime
only. A series of 72-hour pilot plant runs was conducted with the
magnesium precipitated increased after each three-day period. Effluent
characteristics improved as the amount of magnesium precipitated was
increased. Influent and filter effluent samples were collected every
four hours and analyzed for COD, TOC, total phosphorus, alkalinity,
hardness, calcium, and magnesium. Values for BOD were determined from
composited samples. The percentage reduction in chemical (COD) and
biological (BOD) oxygen-consuming substances ranged from a low of 70%
for no magnesium ion precipitated to a high of 90% for 30 milligrams
per liter of magnesium ion precipitated. Higher dosages have not
yet been investigaged.
This report was submitted in fulfillment of Project Number
12130 HRA by the City of Gainesville, Florida, under the (partial)
sponsorship of the Environmental Protection Agency. Work was completed
as of September 1, 1973.
ii
-------
CONTENTS
Page
Abstract , , ii
List of Figures , iv
List of Tables v
Acknowledgments ix
Sections
I Conclusions 1
II Recommendations 3
III Introduction 4
IV Metal Ammonium Phosphates 9
V Coagulation of Sewages 37
VI Pilot Plant Studies 79
VII References 106
VIII Appendix 110
iii
-------
LIST OF FIGURES
No. Page
1 Pilot Plant Flow Scheme 80
2 View of the Pilot Plant 82
3 Process Flow Scheme 85
4 Typical COD Reduction Curves 99
iv
-------
LIST OF TABLES
No. Page
1 Average Characteristics of Effluents 12
2 Average Characteristics of Effluents 13
3 Average Chacteristics of Effluents 14
4 Initial Test for N and P Removal 16
5 Coagulation - Flocculation of Trickling Filter
Effluent with Lime in pH Range 7.9-9.6 18
6 Coagulation - Flocculation of Trickling Filter
Effluent with Lime in pH Range 10.0-11.6 19
7 Nutrient Removal Utilizing Magnesium and
Phosphorus 20
8 Nutrient Removal as Magnesium Ammonium Phosphate 21
9 Nutrient Removal as Copper Ammonium Phosphate 22
10 Nutrient Removal as Calcium Ammonium Phosphate 24
11 Nutrient Removal as Calcium Ammonium Phosphate 25
12 Nutrient Removal as Calcium Ammonium Phosphate 26
13 Nutrient Removal as Calcium Ammonium Phosphate 27
14 Nutrient Removal as CaNH^PO^•H20 28
15 Nutrient Removal as CaNH4POtt-H20 29
16 Nutrient Removal as Ca(NHtt)2 (HPOi^-t^O 30
17 Nutrient Removal as CaCNHi^ (HPOlt)2'H20 31
18 Nutrient Removal as Calcium Ammonium Phosphate 32
19 Nutrient Removal as Calcium Ammonium Phosphate 33
20 Nutrient Removal as Calcium Ammonium Phosphate 34
v
-------
LIST OF TABLES—Continued
No. Page
21 Nutrient Removal as Calcium Ammonium Phosphate 35
22 Solubility of Calcium Ammonium Phosphate 36
23 Total Alkalinity Fluctuations 38
24 Coagulation of Raw Sewage 39
25 Coagulation of Raw Sewage 40
26 Coagulation of Raw Sewage 41
27 Coagulation of Raw Sewage 42
28 Chemical Treatment of Raw Sewage 43
29 Coagulation of Raw Sewage 45
30 Ammonia Removal 46
31 Gainesville Wastewater COD Removal by Lime
and MgC03'3H20 Plus Lime 47
32a Effect of Increasing MgC03«3H20 and Lime on
Removal of COD, TOG, and Total Phosphorus 48
32b Effect of Increasing MgC03-3H20 and Lime on
Removal of COD, TOG, and Total Phosphorus 50
33 COD Reduction With and Without Addition of
MgC03'3H20 51
34 COD Removal by Magnesium Carbonate Hydrolyzed
With Lime 53
35 COD Removal by Magnesium Carbonate Hydrolyzed
With Lime 54
36 COD Removal by Magnesium Carbonate Hydrolyzed
With Lime 55
37 COD Removal, Duplicate Samples to Check
Reproducibility of Results and Effect
Mg++ 56
VI
-------
LIST OF TABLES—Continued
No. Page
38 BOD, COD and TOC Reductions, Gainesville,
Florida, Sewage 57
39 North Miami Raw Sewage After Comminutor 59
40 N. Miami Sewage - Compared with Gainesville
Sewage 60
41 Comparison of Wastes from U. of Florida,
Gainesville and N. Miami 61
42 Mixture of Wastes of Gainesville, University of
Florida and North Miami to Check Effect of
Flocculation Time on Adsorption of COD 62
43 COD Reduction on Municipal Wastes of Montgomery,
Alabama, With and Without Addition of MgC03'3H20 63
44 COD Reduction on Municipal Wastes of Montgomery,
Alabama, With and Without MgC03«3H20 Addition 65
45 COD Reduction on Municipal Wastes of
Montgomery, Alabama 66
46 COD Removal by Magnesium Carbonate
Hydrolyzed with Lime 67
47 COD Removal by Magnesium Carbonate
Hydrolyzed with Lime 68
48 COD Reduction by Magnesium Carbonate and Lime
Using Montgomery, Alabama, Raw Sewage 70
49 COD Reduction by Magnesium Carbonate and Lime
Using Montgomery, Alabama, Raw Sewage 71
50 Treatment of Montgomery, Alabama, Waste with
Lime Alone, High COD Waste 72
51 Treatment of Montgomery, Alabama, Waste with
Lime and 50 ppm MgC03• 3H20, High COD 73
52 Treatment of Montgomery, Alabama, Waste with
Lime and 100 ppm MgC03'3H20, High COD 74
vii
-------
LIST OF TABLES—Continued
No. Page
53 Coagulation of Low Magnesium, High COD Waste,
Montgomery, Alabama 76
54 Coagulation of Low Magnesium, Low COD Waste,
Montgomery, Alabama 77
55 Dimensions of Main Units 79
56 Product Recovery by Carbonation of Mg(OH)2 86
57 Recovery of Product MgC03•3H20 87
58 Alkalinity 'as CaC03 of Selected Samples
of MgC03'3H20 88
59 Carbonation of Sludge 89
60 Carbonation of Thickened Sludge 90
61 Carbonation of Sludge 90
62 Carbonation of Sludge 91
63 Carbonation of Sewage Sludge 91
64 48-Hour Pilot Plant Run 94
65 72-Hour Pilot Plant Run 96
66 Effect of Increased Magnesium Precipitated 100
67 Lime Only Treatment 101
68 5 mg/& Magnesium Precipitated 102
69 10 mg/fc Magnesium Precipitated 103
70 20 mg/H Magnesium Precipitated 104
71 30 mg/£ Magnesium Precipitated 105
viii
-------
ACKNOWLEDGMENTS
The support of the city of Gainesville, Florida, for providing
the site for the project, the utilities, and the wastes is gratefully
appreciated.
Gratitude is extended to Messrs. R. P. Vogh, E. W. Ranew,
G. A. Sarver, A. J. Fowler, and other members of the City of
Gainesville Wastewater Plant for their tireless assistance and
valued advice in construction and operation of the pilot plant.
Acknowledgment is extended to the Department of Environmental
Engineering, University of Florida, for providing laboratory space
for this study and for the helpful assistance of the following:
Dr. J. E. Singley, Professor of Water Chemistry, Jeanne Dorsey,
Secretary.
The support of the project by the Office of Research and
Monitoring, Environmental Protection Agency, and the Project Officer,
Dr. R. P. Stringer, is acknowledged with sincere thanks.
ix
-------
-------
SECTION I. CONCLUSIONS
The treatment of municipal wastewaters by coagulation with
recycled magnesium bicarbonate and lime has been investigated. Due
to the present unabailability of magnesium carbonate tri-hydrate,
magnesium sulfate was used as the source of "make-up" magnesium. The
magnesium hydroxide formed by its precipitation with the magnesium
naturally present in the waste was converted by carbonation of the
sludge to the highly soluble bicarbonate and recycled.
The wastewaters studied may be grouped with respect to their
total COD values into three categories.
a. Wastes containing less than 400 mg/£ of total COD.
b. Wastes containing from 400-800 mg/£ total COD.
c. Wastes containing more than 800 mg/£ total COD.
1. The data showed from 10% to 30% more total COD removed from
category (b) and (c) wastes by coagulation with recycled
magnesium bicarbonate and lime than could be obtained by
treatment with lime only. This level of reduction was also
achieved for both BOD and TOC. Jar tests indicate that even
greater percentage removals of total COD should result from
the coagulation of very high COD wastes, that is, in the
range 1,200-2,000, but pilot plant runs in this range have
not as yet been made.
2. Values for total phosphorus in the clear, settled effluent
from the coagulation unit were normally less than 0.1 ppm
P except when sludge carryover took place. As reported by
Menar and Jenkins (53) and verified in this research, any
phosphate solids that escape the sedimentation basin
(pH 11.5) will be resolubilized in the carbonation basin
(pH 9.5). Filtration following carbonation is of no value
in removing phosphate.
3. Values for both suspended solids and color were very much
lower than when lime alone was used. Vaues for suspended
solids before carbonation were usually less than 2 ppm and
residual color after stabilization was usually less than
5 pcu.
4. Coagulation of the low total COD waste of category (a) with
recycled magnesium and lime resulted in small but measurable
-------
improvement over values when lime alone was used. This is
probably due to the fact that the percentage of soluble COD,
less likely to be removed by coagulation than COD due to
suspended solids, is normally highest in wastes having low
total COD values. However, the much greater percentage
reductions in phosphate, suspended solids and color found
for category (b) and (c) wastes were also found for category
(a).
5. Although, as noted above, the greatest percentage reduction
in total COD was found where suspended solids were high,
usually in the range of 175-250 mg/fc, the soluble COD
values of the wastes of categories (b) and (c) were reduced
by 60%-70%. Values for soluble BOD and TOC were reduced by
about 40%. This reduction was brought about by precipitating
at least 20 mg/£ of magnesium ion.
6. In the operation of the pilot plant, rapid mixing was not
used. Best results were obtained by adding the recycled
magnesium bicarbonate liquor to the raw waste effluent at
the splitter box and the lime at the influent to the up-
flow flocculator. A rotational speed of 2 rpm was found
sufficient for adequate mixing and the formation of large,
heavy floes, their density greatly increased by the presence
of the coprecipitated calcium carbonate. The flocculator
performed as a fluidized bed, retaining the occasional
larger particles.
7. The addition of about 4 ppm of activated silica or 0.10 ppm
of a strongly anionic high molecular weight polymer such as
A23 increased floe size and density and improved settling.
8. Due to the high buffer capacity of the coagulated wastewater
in the high pH range employed, it was found that more accurate
control of the treatment process could be obtained by dif-
ferential titration than by determining the pH value.
Hydroxide alkalinity in the range 140-180 mg/£ (corresponding
roughly to the pH range 11.4-11.6) was found to produce the
lowest values for residual COD.
-------
SECTION II. RECOMMENDATIONS
The municipal wastewater of the city of Gainesville, Florida,
is typically a weak-medium strength wastewater as defined by the COD
test. A very light industrial load is presently imparted to the
sewage. One of the main advantages of the magnesium process demon-
strated in the laboratory is the superior treatment of high-strength
wastewaters over lime alone. This fact needs to be evaluated on a
pilot plant scale by adding high strength wastes to the raw sewage.
Industrial wastes or digestor supernatant would serve as supplements.
While many chemical coagulants are available, lime coagulation
was the only one used for comparison during this study. Further
laboratory and pilot plant studies need to be performed utilizing
higher dosages of magnesium carbonate and lime to completely evaluate
the magnesium process.
The sewages studied had a ratio of insoluble to soluble COD
(and BOD) of three to one. Wastewaters which have a low insoluble
and high soluble COD (BOD) need to be evaluated in both jar tests
and the pilot plant in order to determine the effect of the magnesium
coagulation on dissolved constituents.
Phosphorus and heavy metals are removed by the coagulation
process and, therefore, are contained in the sludge. The reclaimed
magnesium and calcium values need to be evaluated to determine the
presence or absence of phosphates and heavy metals. A specific
problem is the rate of phosphorus accumulation in the calcium carbo-
nate prior to recalcination. The possibility of separation of the
phosphate fraction by flotation requires evaluation.
The amounts of organic and ammonium nitrogen and heavy metals
removed by the lime magnesium process should be determined and
compared with removals obtained with other coagulants.
-------
SECTION III. INTRODUCTION
Over the last few years it has become apparent that conventional
"secondary" biological sewage treatment processes do not provide the
degree of treatment required for effective water pollution control.
Well-operated biological treatment processes can provide at best
approximately 90% removal of suspended solids and biochemical oxygen
demand with little or no reduction in nitrogen and phosphorus levels.
This level of performance will not meet the increasingly stringent
demands for better water quality and more effective pollution control.
As a result, considerable effort is now being devoted to the develop-
ment of physicochemical processes capable of accomplishing the degree
of treatment required by more exacting effluent standards (1,2). The
traditional approach to the application of physicochemical processes
to wastewater treatment has centered on providing "tertiary" treat-
ment for wastes which have already undergone conventional "secondary"
biological treatment. This increment of tertiary level physico-
chemical treatment to conventional biological processes results in
significant additional treatment cost. In addition, the effective
operation of a tertiary treatment system depends on consistent and
effective operation of the biological secondary process. Because
of these factors the effort is now being made to develop successful
physicochemical treatment processes which can be applied to municipal
primary wastes (3).
Review of Literature
The first attemps to chemically treat sewage were made in
Paris in 1740 (4). In the next 100 years chemical treatment
processes became well established in England. Most of these plants
used iron and lime salts as coagulants. The promotion of several
such processes was done on the basis of the supposed value of the
sludge as fertilizer. The chemical processes gradually lost favor
because they were expensive, did not produce a stabilized effluent,
and yielded larger quantities of sludge. By 1910 most of them had
been discarded in favor of plain settling followed by biological
processes. Chemical treatment never generated much interest in the
United States during this period.
In 1929, Rudolphs et al, (5) revived interest in chemical
treatment by describing the increased settling rates of sewage
solids brought about by the addition of small doses of ferric
chloride. The most widespread use of chemical treatment was in
improving the degree of treatment achieved by sedimentation. Many
combinations of chemicals were tried (1,6-9).
-------
Between 1936 and 1941, Rudolfs and Gehm published a series of
papers dealing with coagulation of sewage (10-17). They found the
optimum pH ranges for coagulation with iron salts to be 2.5 to 3.5
and 9.5 to 10.5 with some slight shifting of the optimum pH caused
by variations in septicity, quantity and type of industrial wastes
present and quantity of iron coagulant used.
In England a number of attempts were made to improve the
quality of effluent from chemical treatment of sewage.
The Laughlin process (18) consisted of adding ferric chloride,
lime, and paper pulp to raw sewage, settling for 1 hour and then
filtering. Suspended solids removals of 85% to 95% and BOD removals
of 65% to 85% were obtained. Results comparable to those achieved
by the Laughlin process were achieved at Great Neck, N.Y., by passing
coagulated and settled sewage through two vacuum filters, using paper
pulp as a filter medium (19).
The Guggenheim process consisted of the screening of raw
sewage, coagulation with lime and ferric sulfate, flocculation,
sedimentation, and zeolite filtration (20). A final effluent with
1 mg per liter suspended solids, 5 mg per liter BOD, and 2 to 3 mgs
per liter total nitrogen was produced.
A scheme was also proposed using the Aero-Accelerator
manufactured by Infilco (21). Calcium carbonate sludge from a water
softening plant was added to raw sewage suspended solids in a 1:1
ratio. Ninety-five percent BOD reduction was obtained.
The Landreth process consists of coagulating raw sewage with
lime and then subjecting the sewage and lime floe to electrolysis in
a basin containing iron electrodes (4).
Ferric chloride was used as the coagulant in the Stevenson
process (22,23). Recovery and reuse of the coagulant was attempted.
Alum recovery was practiced in Holland, and about 80% recovery
was achieved (20). The work done at Lake Tahoe has shown that an acid
alum recovery scheme is only feasible if the chemical coagulation is
not designed to remove phosphates (24). Lime recovery by recalcination
was conducted at the sewage treatment plant at Syracuse University (25).
This study indicates that coagulant recovery can be achieved by
recalcination.
Activated silica has also been used in sewage treatment.
Hurwitz and Williamson (26) used copperas and silica for chemical
sewage treatment. Rudolfs (27) also did some preliminary work using
acid activated silica.
Studies have been made on the use of proteins as coagulant
aids (28). Compounds of gelatin and ferric chloride ("Ferrigel") and
of gelatin and aluminum chloride were found to be effective coagulants.
-------
Background Information
Until 1957, a recalcination of sludge produced by softening
high magnesium waters by the lime soda process was considered impos-
sible. Wide adoption of recalcination had been retarded by the fact
that no successful method for the physical separation of the calcium
carbonate from the other components had been developed. In that year,
Black and Eidsness (29) developed a process to selectively dissolve
the magnesium hydroxide from the calcium carbonate using carbon dioxide
gas. This gas would be readily available from the lime kiln.
At Dayton, Ohio, which softens well waters high in magnesium,
a high quality quicklime has been produced using this process since
1958. The supernatant from this process, containing the magnesium
which had been converted to the soluble bicarbonate form is then
discharged to the river.
In 1968 the necessity of meeting new and rigorous standards
for such waste discharges required another method of disposal. This
impetus led to the discovery of a relatively simple and inexpensive
method of recovering the magnesium as a carbonate. In this process,
which was developed by A. P. Black (30), all wastewater is recovered
and recycled. The only waste material, when present, is clay, which
may easily be landfilled.
Briefly, the process employed at Dayton is as follows. Sludge
produced from the softening operation is pumped to a sludge recarbo-
nation basin where it is mixed with scrubbed kiln gas containing about
20% C02. The magnesium hydroxide is dissolved from the calcium carbo-
nate. This slurry then passes to a thickener from which the clear
supernatant containing the magnesium, now in the form of soluble
magnesium bicarbonate, will overflow and be passed to a heat exchange
unit where it will be warmed to 40°C. The solution will then flow to
an aeration basin equipped with mechanical stirrers. Precipitation
of the magnesium carbonate as the trihydrate is rapid and essentially
complete in 90 minutes. The snow-white product will then be vacuum
filtered, dried and bagged for shipment. The thickener sludge is
sent to a kiln and calcined to a high quality quicklime. For every
1 ton of lime fed, 1.3 tons to 1.4 tons are recovered. The excess
is derived from the calcium carbonate in the raw water. In turbid
waters, the sludge consists of CaC03, Mg(OH)2, and clay. The Mg(OH)2
is dissolved and recovered as described above. The clay is separated
from the CaCOs by flotation and the purified CaC03 calcined.
Calculations made on the basis of data supplied by a major
midwestern municipal softening plant treating 75 MGD of hard, turbid
water indicate a saving in chemical costs of approximately $340,000
a year. This saving results from:
-------
1. Elimination of alum.
2. Reduction in cost of lime due to recalcining.
3. The use of kiln stack gas both for carbonation of sludge
and finished water pH adjustment.
Not included in this figure were the profits from the sale of excess
magnesium carbonate and excess lime and the reduction in demand for
chlorine used for disinfection. Elimination of prechlorination and
reduced postchlorination are possible since bacterial disinfection
and virus inactivation occur at the process operating pH of above 11.
Certain intangible benefits would accrue such as prevention of
precipitation of alum in the distribution system and the advantage
of producing and stocking your own chemicals for treatment in event
of strikes and national emergencies.
During 1970, coagulation studies were carried out at the
University of Florida1s Environmental Engineering Laboratory by
Black and Thompson (31,32) comparing magnesium carbonate and alum
as coagulants for organic color and turbidity removal. Water from
approximately 20 major cities along with several synthetic solutions
were evaluated utilizing the jar test procedure. These waters
represented a wide range in physical and chemical characteristics.
In summation, the following conclusions were reached:
1. Magnesium carbonate is superior to alum for the removal of
both turbidity and color.
2. The floes formed are larger, heavier, and settle faster than
the alum floes. Therefore, the capacity of the plants will
increase.
3. Color is much more significant than turbidity in determining
the necessary chemical dosages.
4. Release of the coagulated color during the sludge carbonation
step is not a problem when the color of the water is less than
150.
5. The use of magnesium carbonate produces a treated water with
superior physical and chemical characteristics compared to
alum treated waters.
Waters high in magnesium can be treated at a much lower cost then
waters low in magnesium since no make up in magnesium is required
and a lower coagulating pH is possible.
-------
Scope of Investigation
The purpose of this investigation is to evaluate the feasibility
of treating domestic sewage and industrial wastes with magnesium carbo-
nate hydrolyzed by lime. This process appears attractive for several
reasons. First, a new water treatment process using magnesium carbo-
nate hydrolyzed by lime as the coagulating agent has been developed
in which the magnesium carbonate is recycled (31,32). For hard waters
the process will lead to a surplus of magnesium carbonate. This excess
could be used for chemical treatment of sewage. In addition, the use
of magnesium carbonate hydrolyzed by lime has the potential of removing
a greater quantity of COD, BOD and phosphorus than lime alone for
sewage treatment. Bench scale tests will be designed to determine
the magnitude of this potential. The possibility of removing ammonia
nitrogen by this process will also be investigated. Recovery and
recycling of the magnesium and lime from the sludge will be a primary
consideration. If the results are encouraging, a pilot plant will be
constructed.
-------
SECTION IV, METAL AMMONIUM PHOSPHATES
Nutrients in wastewaters are important because, upon discharge
to a water course, they promote biological responses that interfere
with the desired uses of water by man.
Over the last few years it has become apparent that conventional
secondary biological treatment processes do not provide the degree of
treatment required for effective water pollution control. Secondary
biological processes do not reduce the total level of nutrients, but
merely convert them from one form to another. Moreover, when digester
supernatant liquor is recycled within the plant, a two- or three-fold
increase in the total nutrients being discharged occurs.
The addition of tertiary chemical treatment for removal of
nutrients has thus become necessary. An estimate of the relative
costs of the different stages of waste treatment has been reported (1)
as follows:
Primary treatment 3C to 5C/1,000 gallons
Secondary treatment 8c to llC/1,000 gallons
Tertiary treatment to
remove nutrients 17£ to 230/1,OOP gallons
Total 28C to 39C/1,000 gallons
Unfortunately, the use of chemicals has been a constant
consuming process with no prospect of recovery and reuse. In
addition, the sludges produced by these tertiary processes consti-
tute a difficult waste disposal problem in themselves. Therefore,
only a few plants have provided such treatment.
Recently, a new water treatment process using magnesium
carbonate hydrolyzed by lime as the coagulating agent has been
developed in which the magnesium carbonate is recycled and reused
(31,32). Its effectiveness as a coagulant and the savings to be
achieved by recycling both the lime and the MgC03 dictate that
studies should be carried out to determine its effectiveness for
treatment of municipal and industrial wastes and mixtures of the two.
While phosphate may be removed by a number of different
processes, the removal of ammonia nitrogen is much more difficult.
One attractive possibility would be its removal with phosphate as a
metal ammonium phosphate, the best-known compound of this type being
magnesium ammonium phosphate, MgNH^POi, -6^0. A gravimetric method
-------
for the accurate determination of either magnesium or of phosphate by
precipitation as the double phosphate has been employed for many years
and its solubility under the conditions of its precipitation is less
than 1 milligram per liter (33).
Literature Survey
An extensive literature search was conducted regarding the
preparation and properties of the compound, magnesium ammonium
phosphate. Theoretically, the compound reaches its minium solubility
at pH 10.7 (34). A number of investigators have reported solubility
data for magnesium ammonium phosphate, but a comparison of the various
data reveals a lack of agreement among the published results (35-38).
Bridger (35) 180 mg/£
Bube (36) 170 mg/£
Szekeres (37) 160 mg/£
Uncles (38) 140 mg/Jl
All solubility data were based on phosphate analyses except
that of Uncles and Smith, which were based on magnesium analysis.
The rounded average of the values reported is 160 mg/£. Expressed
as phosphorus (P) the solubility is 20 mg/£, as magnesium (Mg) the
solubility is 16 mg/£, and as ammonia (NHs) the solubility is 11 mg/£,
or as ammonia nitrogen (NHs-N) 9 mg/£.
The solubility in the range of values listed would preclude
its precipitation from either untreated municipal wastewater or from
trickling filter effluent unless special conditions resembling those
used in analytical procedures are employed. The use of excess Mg"1"1"
ion and pH variations appeared to be the only variables worthy of
study and they were investigated.
Background Data
In the initial studies using the municipal waste treatment
plant of the City of Gainesville, Florida, the chemical and bio-
logical treatment processes were constantly monitored. The two
processes (activated sludge and trickling filter) are operated in
parallel and treat approximately the same total volume. The influent
to the trickling filter plant is by gravity flow while the activated
sludge plant influent is from force mains and lift stations. During
the course of this research, the digestor supernatant liquor was
alternately recycled to each of the processes on a four-month basis.
The reasons for this procedure were:
10
-------
1. The loss of a prime location for disposal of a considerable
volume of digester supernatant liquor.
2. The desirability to investigate the increased nutrient
concentration that would result from in-plant recycling
of digester supernatant liquor.
3. To evaluate the physical, chemical, and biological problems
that would prevail as a result of recycling the digester
supernatant liquor.
Data from three different recycling situations were compiled.
The first situation (before October 11, 1971) was prior to losing a
prime disposal site and, consequently, only half of the total digester
supernatant liquor was recycled to the activated sludge process
(Table 1). The second situation (before February 18, 1972) was
brought about by the loss of this disposal site for digestor super-
natant liquor, and resulted in a total volume of 30,000 gallons of
digestor supernatant liquor being recycled to the primary clarifiers
of the trickling filter process (Table 2). Case three (after
February 18, 1972) consisted of the total flow from the digestor
being recycled to the activated sludge process (Table 3).
The majority of jar tests for nutrient removal as a metal
ammonium phosphate were carried out under situation two.
Analytical Methods
The methods for analyses conducted on the wastewater streams
were in accordance with Standard Methods for the Examination of Water
and Wastewater (33), Methods for Chemical Analysis of Water and
Wastes (39), and "Methods for Analyses of Selected Metals in Water
by Atomic Absorption" (40).
All nutrient analyses were performed using the Technicon
AutoAnalyzer.* Initial comparisons between ammonia nitrogen concen-
trations obtained by distillation and titration with the Technicon
phenate method were widely divergent. However, upon addition of the
catalyst, sodium nitroprusside, the phenate method (39) yielded values
which checked with those obtained by the distillation procedure (33).
Phosphate analyses were performed using the single reagent method (41).
Total phosphate was determined by treating samples with a strong acid
solution and ammonium perfulfate, and subsequently heating in an auto-
clave for 30 minutes at 12l°C (15-20 psi).
Magnesium and calcium were determined titrimetrically (33) and
by atomic absorption (40). The atomic absorption unit was utilized in
Technicon, Tarrytown, N.Y.
11
-------
rH
0)
rH
t-Q
cd
H
X~N
CM
5>fc
CO 4-1
•U 13
C cd
0 C rH
rl rH
CO 3
O 4J 1-1
•H a> a)
4-1 P^ JD
co o
•H 4-1 4J
rl (3 0
0) cd O
4-1 4-1
a co o
Cd C 4-1
M M
cd 0) JH
43 & O
0 3 -H
co s-i
O) PM
00 >-i
cfl O
)-l 4-1
ai co
> 0)
•^ 00
•H
Q^
4-1
C
O 4J
4J Cd
W (3
(1) M
00 0)
Q 3
CO
4J
4J 13
13 01
cd 3
ft . U 1
M-l
C3 W
O
CN -H M
=tt= 4-1 0)
cfl -H
4-1 N <4H
13 -H -H
cfl rH !-(
rH -H Cfl
PM rO rH
cfl C_3
4-1
CO >i
fc
4J cd
CJ 13
cfl a
4-1 0
13 0
O 0)
U CO
4-1
a
cd (-1
rH (U
PM -H
M *r-t
rH 0) J-l
=Sfc 4-1 Cd 4J
rH rH (3
4-1 -H O 0)
C3 rM 3
Cd ^» rH
rH 00 }-l 4-1
PM C3 cd <4H
•H T3 W
rH C
M o
a a
•H Q)
l-l CO
H
0 0
1 1 1 0
r>- 1 I i so
rH ITl CM OO «N
• rH VO CN •
r^» r^»
rH CO iH v£> f^
• iH 00
en
en ^" r~^
o cr>
^d" rH
PM
CO
cfl
J- 4J
O -H CO
PM C W
1 -H (U
O rH |3
43 cfl t)
4-1 AS IH
M r-l «J
^^ "^5 *]£
12
-------
CM
0)
rH
,0
cd
H
/•-s
i-H
-fc
csl
CO 4-1 I--
4-1 C Ov
C Cd rH
CU rH
3 PU •>
rH 00
MH O rH
I4H 4-1
W >.
•d rl
MH cu cd
o c 3
LJ t-t
rH ^H
co 3 .a
0 4-1 CU
•H cu fn
4J <*!
CO O
•rl 4J 4J
rl 0
cu cd «
4-1 4-1 i-H
o cd r~
cd C cr>
rl rl rH
td cu
J2 (X •
CJ 3 i-H
CO rH
(!)
00 l-i l-i
td o cu
S-l -U ,O
cu co o
> 0) 4-1
< 60 O
•H O
P
*^^
4J
e
M cd
O 4-1
4-i td
to C
cu $-i
00 CU
•H CX
P 3
CO
4->
•W C
C 0)
cd 3
rH rH
Pu MH
MH
CTvl
r*4
o
CM -H rl
=«= 4-) Q)
td -rl
4-1 N IM
C -H -rl
Cd rH M
rH -H Cd
Pk J3 rH
cd cj
4-1
CO >-,
M
4-1 Cd
CJ -r)
cd C
4J O
C 0
o cu
O CO
4J
c
cd Vj
i-H CU
pti -H
14-1
rl -H
i-H 0) IH
=Sfe 4J Cd 4J
rH rH C
4J -rl U CU
C Pn 3
Cd f>-.rH
i-H 60 ^i MH
Pu C! cd MH
•H T3 W
«-i a
* 0
0 0
•H Q)
I-l CO
H
0 O
I I t I %o ao
1 1 1 1 "I
i-H 00 ul O 00 -d"
rH 1^ CM
r-»
CM o in m m
« CO iH m CM
r^ rH
PM
co
cd
!^ -*
4J 0
•H PH
TJ 1
•H Z O
J3 1 4=
IH P Q cr> 4-1
sa 3 o o w u
CX H 0 « a O
O 0
o m
CM 1^.
«\
CO
in m
rH O
rH rH
0 O
m I-H
rH rH
•^
?"l
4J
iH CO
a CQ
•H 0)
rH a
td T3
^S r.
5! a
13
-------
ro
CU
iH
,0
cd
H
CO
4J
C
cu
3
rH
M-l
M-l
W
14-1
O
CO
O
•H
4-1
U)
•H
W
QJ
4J
O
cd
s-i
cd
JC
CJ
cu
00
cd
i-i
cu
3
/~\
CM
=$£:
4-1
C
cd
rH
PM
O CM
4J r-~
T3 rH
cu
C -
i-l 00
3 tH
4J
cu >>
ftf h
cd
4-1 3
a ^
Cd U3
4J CU
cd fn
C
M M
a) cu
CX 4-1
3 U-4
w <3
>-i
o
4-1
CO
CU
GO
•H
O
**^
M
O
4J
CO
01
00
•H
Q
4-1
G
cd
j
l~1
PH
a
0
CM -H
=tfc 4-1
cd
4-J N
C -rl
Cd i-l
rH -H
PM 43
cd
4-1
CO
4-1
CJ
cd
4-1
c
o
CJ
4-1
fi
Cd
tH
PM
M
rH CU
% 4-1
,-H
4J -H
C PM
cd
rH 00
PM 0
•H
rH
Jd
0
•H
h
H
4J
C
cd
4-1
cd
C
t-l
0)
a.
CO
4-1
0
OJ
3
fH
m
M-l
W
M
0)
•rl
IW
•H
M
Cd
rH
CJ
t>>
M
cd
•a
e!
o
u
cu
to
M
a)
•H
4-4
•H
M
cd
i-H
CJ
>,
n
cd
T3
C!
O
O
0)
w
4-1
CJ
cu
rH
M-l
U-l
W
o
I oo
I m
CM
o
CSJ
ro
oo
rO
vO
OO
o
CM
o
o
cs
CM
CM
00
O
m
vO
O
CO
O
o
•H
T3
•H
Pw
CO
cd
o
PM
O
43
4J
CO
CO
-------
combination with a DBG grating spectrophotometer and potentiometric
recorder with scale expander.
Prior to analysis for total alkalinity, total hardness, calcium,
magnesium, nitrogen, phosphorus, chemical oxygen demand, and biochemical
oxygen demand, all chemically treated wastewater samples were filtered
through Whatman No. 2 paper.
Magnesium Ammonium Phosphate
In the initial phase of the jar tests, wastewater analyses
showed a surplus of ammonia nitrogen over phosphorus on the order of
2 to 1. Therefore, phosphate had to be added to the wastewater in
order to achieve the proper N/P ratio for precipitation of magnesium
ammonium phosphate. The calculations to assure the proper proportions
are as follows :
NH,-N =1 P = 3P;97 = 2.2 x NH,-N
3 14 o
NH,-N - 1 PO, = i^il = 6.8 x NH3-N
3 k 14
NH,-N =1 Mg = j = 1.7 x NH3-N
d ° 14 3
NH3-N = 1 MgC03'3H20 = ' = 9.9 x NH3-N
The phosphorus was added as potassium dihydrogen phosphate (1 mA = 10
mg P).
All jar tests were conducted using an improved version of the
multiple stirrer.i" All pH measurements were made using a pH meter
with a combination glass and Ag/AgCl electrode.** The pH meter was
calibrated daily.
The first jar test was conducted on August 12, 1971 (Table 4).
The alkalinity of the waste was significantly higher than the total
*Beckman Model 1301, Beckman Instruments, Inc., Fullerton, California.
Coffman Ind., Inc., Kansas City, Kansas.
Corning Model 7, Corning Glass Works, Philadelphia, Pennsylvania.
15
-------
o
e
os
a,
nj
H
f-
rH
Oj
J3 •»
4-1 O
O
E-H
to
2
Q
O
U
O
0
S
o
u
in
UJ
z
o
IE
<
X
^
"
m
<
V)
^
"
9
Z
Ul
Q
1
4
ro
0
C
CO T3
rH d>
d-, rH
r. 4-1
r< (1)
4-1 O C/>
rH -H
•H rH U
U- p. 0
i— <
BO CM LH
C S *
•H (X MH
rH Pi 2 C 0
M O-, rt
U O OJ 6 r->
•HO 4-1 (3
>- T-H 0 05 U] 3
H to JS .H O
ca> ^ t/1 g
j-> r< O
m >H c 3 js O-, .H .
W g -H HI 4-1 MH X .
G -H T3 -H «H J-i OO CTl OO Tj-
Xo 6inC CEoj LOOI LO
r-l rH O HI Oj bO (P
03 CM O rH .H £ rH 1 1 1 1
T3r< tOPL.- If) O U
COri 62 rH rH rH i— 1 rH
OMnOrnCd "tn-j-J oJoJrtnj
U MHOO~ (D G >>>>
CUX MH O bObcol OOOO
W-HX tso ^c4-i eses
S-'-f'OcDO nj-HoS rH D O 3 rH
LO -H 12 4-> 4-1 4-1 O T3 O «*> o\° Q\» o\»
4-lfiO4-lrH O fX Oa,2O
COirHCU-H rH 3O O
(OOSCOWtL, U. CO U «
O
U
a:
lu
r-
J CM Tf
jlft-vOrHLOvO <* rHl>-
OTtOCTlLO'*00 rj- rH
$ ^
-, «
? °
2 ^
OOOT-Q
»<«o.. ......
<0 ,. O "
0 w o IU_lCD,m_lm "
K<<0Z<3^5<5 0
-------
hardness, representing a situation of "negative" noncarbonate hardness.
This is typical of domestic wastes and required higher lime dosages
than were anticipated. The data indicated that more phosphorus was
removed as pH was increased, but that less ammonia nitrogen was
removed. Theoretically, the compound, magnesium ammonium phosphate,
reaches its minimum solubility at pH 10.7 (34). This test demonstrated
an increase in ammonia nitrogen once this pH value was exceeded.
Approximately 60% of the COD and 54% of the BOD were removed by coagu-
lation with magnesium carbonate and lime.
The second and third jar tests (Tables 5 and 6) were designed
to study the lime demand of this wastewater to achieve a desired pH
level. Simultaneously, a study of the ability of lime to remove
phosphorus was carried out. Phosphorus removal, of course, increased
with increasing pH values.
These tests were then repeated using only added MgC03 and
phosphate (Table 7). The magnesium dosage was extended to include
a three to one excess of magnesium. No significant removals of ammonia
nitrogen were observed. Phosphorus removal increased with increasing
pH values.
Table 8 shows jar tests performed to precipitate magnesium
ammonium phosphate in the pH range of 9.2 to 9.7. Although ammonia
nitrogen was not removed, phosphorus was precipitated as pH increased.
As mentioned previously and shown in Table 1, the ammonia
nitrogen concentration in the secondary wastewater is less than the
minimum ammonia nitrogen value as magnesium ammonium phosphate.
Therefore, these tests apparently show that removal of ammonia nitrogen
is not possible in this concentration range.
Other Metal Ammonium Phosphates
The preliminary findings suggested a further literature search
aimed at exposing other metal ammonium phosphates with lower solu-
bilities (35).
Copper ammonium phosphate, solubility 9 mg/fc, would produce
an ammonia nitrogen concentration of less than 1 mg/fc. A jar test
was conducted in order to precipitate this compound (Table 9). The
failure to produce significant nitrogen removal and the cost of CuSO^
precluded further consideration of this compound.
Iron ammonium phosphate, solubility 95 mg/£, and manganese
ammonium phosphate, solubility 38 mg/£, were two other possible
compounds for consideration. However, the cost, lack o . recycling
probability, and high nitrogen solubility eliminated tiise compounds.
17
-------
ai
i
01
a
nl
p,
c
rJ
Si
ci
C
r.
H
C
O
3
U
o
o
4-1
rt
be
0)
o
u
U
x •*
•M O
r< a,
O
o»
X
0
o
HARDNESS
X
_l
<
to
=1
«f
0
r-
L>
Z
0
10
o
'1
10
o
u
0)
<0 I
}— ^ Q.
V)
*
_J
<
in
o
0
X
o
X
a
\
?
z
Ul
o
<
V)
o
o
9UIT1
d
O
O
rH
O
(Nl
^3-
Ol
(Nl
OO
rH
1^.
(Nl
LO
\O
(Nl
Ol
rH
ro
OO
i— t
i-H
1O
Ol
LO
(Nl
r-t
O
(Nl
\o
HI
O
u
10 O
o
o
-------
O
o
0)
be
ts
3;
(X
C
0)
3
•s
H
Ci
C
<4H
O
3
U
o
o
rt
rH
3
cfl
O
U
A t
4-> O
!-. PL,
O
s
6
u
HARDNESS
<
•I
F)
U
r-
O
Z
u
10
o
to
o
u
Sox
r-K a
-J
10
0
o
X
o
X
f
DOSAGE IN
aiutq
d
O
0
rH
00
rH
r-
00
LO
0
rH
CM
LO
rH
0
O
rH
O
LO
rH
O
CM
rH
O
to
rH
r-
00
0
0
rH
OO
0>
CTi
O
rH
O
O
OJ
0
CM
04
o
to
0
LO
o
rH
c-
K
tn
O
rH
0
Lf)
O
C^J
to
to
o
rH
to
rH
LO
to
rH
LO
O
to
rH
rH
rH
O
O
to
o
rt
CM
O
rH
LO
OO
rH
OO
rH
VO
LO
Ol
rH
i-H
rH
O
LO
to
O
CM
LO
04
0
O
LO
CM
LO
CM
CM
LO
CM
CM
rH
rH
O
O
O
vD
CHARACTERISTICS OF RAW WASTEWATER Gravity Flow Raw Sewage NO^-N 0.09 mg/1
bo
6
to
O
Z
4->
C
at
3
rH
4H
<44
H
4-J
C
rt
!i
(U
4->
rH
•H
OH
be
P!
A!
O
VH
E-
10
o
u
10 O
o m
o w • o
o < « O
<» o °
»|S
?!'
Z I
u
X
_J
<
=: u
X r- J <» H
^ g S 5 ?
9
8
5
O
t-
•
° 8 8
°
u • ui
5 I
O
0.
O
10
19
-------
t-^
rH
XI
ca
E-
in
3
(H
O
X
(X
in
o
. CI-
TS
n)
e
oc
rt
O
ui
z
0
IE
X
*
<
i
CD
U>
It
<
^.
?
2
IU
0
o
a
4
-3
IO
'I
.cT
"5
i-
^
u
m
O
i
10
0
u
g*
n
8
S
X
a
)ZH£-
-OD^W
d
i
0
o
(Nl
CO
m
i^
i/i
t~
o
LO
O
tvl
t— (
0
t-~
rH
00
LO
t^
00
f^
o
o
rH
0
CM
(Nl
O
LO
r- 1
O
LO
f^
rf
oo
0
LO
I-H
O
CXI
to
0
rH
rH
00
LO
r^
00
oo
o
o
r-j
0
CM
^~
O
rH
rH
00
LO
r^
o
CTl
0
LO
CM
O
CM
LO
0
o
1-1
CM
LO
r~-
rH
Ol
O
0
to
o
CM
VO
C
IU
3*0
W CD
r. C
.H
C
ca
E
J3 in
S X
rH
j2 ca
MC
3 ca
o
ID
-H
r<
(U P
r. 3
rH U M P. (-1 P, P,
•H C o E
r. .H O «
H O. O, W
c
(U
E
o
u
o
E-
r*j
o
rH
O
U
a
u
3
>
»-
z
_l
4
X
_l
<
"*
O>
l<
O
u
Q
CJ
If,
ARDNESS <
I
_j
<
r-
o
LO
LO
t
»
O
u
•
«.
9
a
=
c
_
2
0>
to
|C
O
r> U
0
U
tfl
4
=
iC
4
»
O
c
u
••
<
H
?
0
c
u
111
_1
s
3
-J
0
M
U
o
K
^
P
e
u
e
u
_l
n
—
g
a
c
•
•J
<
^
g
o
s
111
_
0
3
_.
0
M
O
rH
00
2
1
r
z
z
rH
LO
^•
4
O
o a.
t
0
20
-------
oo
O
S-J D,
O
IO
X Z
2
d»
Z
a
o
HARDNESS
*
<
IO
ss
•f
u
i-
X
r. )= D 4->rH
CU -H C rH Ctf
•t-l g -H -H C
rH g MH CD
•H LO
m iH O in -M
tO Q) £
BO X rH 0>
C -H X pvH
•HS-H 6 rn
U -H ^ C
•H ft O rH
H Cd rH rH
H Pi CO <
O
u
£
o
iH
O
•!-> -1-1
cd -
i-
ALKALIN
vO
01
H
O
u
o
u
in
<
in
ARDNES
X
-i
4
|
\0
LT3
">
»
o
*J
5
a
X
5
3
u
o
»3-
1"
o
n u
o
u
in
«
MAGNESIU
•>
a
c
u
_
£
c
s
u.
I
_
o
If
u
o
1-
I
u
e
u
5
S
o
o
•
J
4
1
0
s
u
_l
0
s
c
CO
CXI
I-l
oo
z
1
r
X
z
0
oo
•*a-
4
o
D Q-
1
O
21
-------
o>
vU
rt
f-
*->
rt
O
o-
§
•H
C
i
Cx
CX
o
o
6
o>
3
2
to
az
v
x
d
o
HARDNESS
x
_i
<
ro
'i
•f
>-
o
z
o
fO
o
g
ro
O
o
m
vD
*£>
t-~
LO
rt
0
-a-
r-H
00
1-4
LO
r*j
vO
»3-
r~-
o
*t
o
VO
tH
O
CJ
<£>
c
(U
4H 3 4->
W C 3
j) g
4-> O
i-H CM O
•H tO
O in
es«4H o
C 4H
•H 2
4-1
c
(U
o
u
c
• H
0>
in
13
§
u
o
i-H
4H
O Oi !-.
•rt O O
>H O O O
H i-H to CL,
CM
i-H
it
O
U
a
u
VI
1-
z
-1
<
X
-
4
CO
01
II
o
(J
a
) u
u
Ul
(fl
u
a
o
K
3
_
i
1*0
LT)
f)
|
8
'
3
s
(.
•
LO
^
|l
0
O
-------
Calcium Ammonium Phosphates
According to Lange (42) calcium ammonium phosphate,
CaNHi+POif«H20, is insoluble. Lehr et ail. (43,44) have prepared
several different fertilizers containing this compound.
A series of jar tests were conducted over a one-week period
similar to those shown in Tables 5 and 6. Twenty-four jars were
dosed with the same amount of phosphorus and slightly increasing lime
dosages. The entire pH range, from 9.2 to 11.5 was covered with the
pH of each jar increased by only one-tenth of a pH unit. Tables 10
through 13 show no significant ammonia nitrogen removal. Primary
effluent to the trickling filter beds was used as the wastewater
source during these jar tests.
Tennessee Valley Authority (45) has described the properties
of over 200 fertilizers as well as the methods of preparation. The
calcium ammonium phosphates described in this publication by TVA were
prepared by adjusting the pH prior to introduction of the calcium
compound. Up to this point in this research, lime had been used to
adjust the pH value in the jars.
Tables 14 and 15 are examples of jar tests conducted by
adjusting the pH with trisodium phosphate and disodium hydrogen
phosphate prior to adding a calcium compound. These tests were
performed in an effort to prepare the compound CaNH^PO^'l^O. The
experiments were not successful. Efforts were then made to prepare
the compound Ca(NHit)2(HPOlf)2*H20, dimorph A, at very low pH values.
Tables 16 and 17 show the data obtained. An elevated temperature was
utilized to avoid interference from the more soluble dimorph B. The
lack of evidence for nutrient removal suggested examination of the
pure salts and evaluations of nutrient removal from synthetic solutions.
The next approach employed was to use solutions of ammonium
chloride in distilled water. The concentrations of ammonium ion were
10 mg/£ and 20 mg/Jl, respectively. Tables 18 and 19 depict the jar
tests conducted in an effort to precipitate the ammonium as
CaNH^PO,- H2
-------
CL)
•M
«
"p.
Oi
0
e
3
c
0
E
^
E
o 3
u
O rH
rH CS
-0 U
C8
H in
03
T-H
>
o
e
C
ID
•H
>-t
4-1
2
to
— Z
d>
2
o
<_>
tn
UJ
z
o
CL
X
*
<
CD
<
r-
tn
X
<
X
9
2
UJ
CO
o
Q
0
4
s§
o
sS
0
r-
O
Z
o
K)
o
u
X
ro
o
0
£*
IO
8
X
o
i
0.
suiiq
MH
•H (ft £*
u, ou in
4-> 0) CM
M p *J at
•H -H C C
rH E -H CO
M e e
u o +j
f-> CM ^3
H H S w
O r< -H
i LH o ' in
<-> S T3 rH
C CL, S: 0 rt
HH rH CM -H P!
W MH OD
XX -H
X'H «H in rM
MS 2 o> +J
Cfl rH 3
S -P S P- P!
•H m o 6
^ 0) rH CU
O, PL, CO CO
in
•1-1
C
CJ
e
e
0
u
cc
UJ
<
> O ^£> CM t-~
1" rH 01 rH
OT ^
_. K)
1 °
2 o • • • . . . .
o o g • o • •
" 4 K> O
CO ° ,„ O "
. . CO fl (J .
ii co w " • •
fc " g ° 3 -8 -g §
s *: S a Q°o^o§
£t<*»0 0 g
G?r»s°^K^^z
«-J_i:suj_icn,iB_iin i
tc<?-i?j?j n <»•
X_JO4
-------
•p
P.,
l/l
JS
O-,
g
3
•rH
C
O
rH E
rH 3
•H
0) U
i — t i — 1
n o3
a u
i/>
03
n3
O
6
&.
•P
•H
3
Z
E 2
Z
£
o
r< fvl ,£
O rl -H
i 4n o i in
HH X
•PS T3 rH
CCX S 0 t«
>4H O O rH P
MH rH CM -H (3
X H -H
X'H >H I/I r>
tti rH 3
E +J S ftC
•H in o e
V-t Gj rH nj
O, tL, CO CO
in
•p
C
o
U
en
Ul
h-
4 vO M
S N •* rH vo'
| rH 01
. ro
* 0
J o
t o
OUW-Q
0 * 10 0 ...
s°«85:. . . :
u OT OT u
fc " 2 S < 8 S I
SilSS o°o°o*
5t4«5fO 0 g
r-zI»2°'"i-',uaim
43JlaJo_1^Jiz i
K44GZ43<3<3 ' O
4*r-J<»r--l?J»-J 100.
I_IO4
-------
aj
t-
oj
,G
c
t/i
o
.c
a,
6
•H
c
o
e
U
,—I
re
O
E
C
0)
K)
Z
ty>
o
0
HARDNESS
x
_l
<
»o
-•§
U
•!I
K
o
z
U
10
o
U
I
tft
o
0
CD
< O X
I-H tt
V)
*
_l
<
ro
O
0
z
o
Z
&
\
9
Z
Ul
o
<«
in
o
a
3UITT
00
o
o
*t
rj
i~-
o
ai
en
o
o
LO
*
c G 3
•H -H G G
iH E -H CS
x e G
O O +J
•H i—t O Clj
(-1 CM ,fi
H !-> fe wi
Oh -H
i MH O i t/i
<4-l X
4-> 2 13 rH
G o. S
a> os PL, s-. e
3 oS N-H -H tn V*
f- S S O 6
ii rt rH rt
O. U, CO CO
e
o
u
oo
o
1— 1
K
O
U
o
o
>
(-
z
_l
<
X
_1
<
^3-
Ol
1*
O
U
0
) U
V
<
in
in
Ul
z
0
(£
«
Z
U
<
K
o
»-
1
t
o
s
•
3
3
:
t.
3
t.
r
o
o o
a
O
in
<
,
~
I
«
a
0
§
t
c
l-
c
c
U
=
) C
• «
U
, e
i t
> 0
> H
u
e
b.
a
c
u
0
c
«
-I
4
>~
0
> t-
Q
C
BE
U,
a
_
G
V
CN]
CNJ
rH
2
'
2
> '
LT>
O
*
^
o
p °-
i o
26
-------
a
"ex
o
ex
E
oniu
E
<
cium
c
v*,
«
r-H
.emova
<->
c
|H
!-,
4->
£*
£
o
u
VI
LU
z
o
rr
i
^
(D
4
r-
10
^
"*
?
z
UI
i
a
«
O
1*1
O
r-
(J
Z
o
ro
O
u
1.
ro
O
0 I
ro
O
I
O
I
ami']
»
t-n
00
i — i
O
VO
01
(-H
i-n
00
rH
0
00
CN1
o
LO
00
rH
0
O
I-O
^
LO
00
t— 1
to
r^
Csj
o
CTi
VJ
r-H
O
LD
s
o
en
LO
I— 1
O
INI
ID
•H
M
•H
u
H-J
2
i-H
w
X
cfl
E
• H
t/1
C
ID
g
E
O
K
LU
4
tu
to
a:
O
u>
K
IO
to
(-
u
4
ft
4
X
U
(D 0
•p a
P +
•H f
E
o
r
0 >
M-l C
£ 5
Pi 0
c
o
0 C
i— 1 0
X >
•H T
S 5
+j ;
01 C
ffl r-
«-. -
t-
z
_)
_l
4
I)
O
IU
n
U CNl
H C£
= £
-4
3 '
- O
4J
^ t— 1
-J -H
<
H t/)
: a>
rH
2 E
H rt
: t/>
un
r"
O
o
o
> u
in
4
V)
u
z
o
4
X
1
m
•rH
in
^
t— i
ttf
C
ca
4->
C
o
!H
<-l
3
C
•>
«
0
•
U
2
1
_
K
O
0 O
a
0
C/1
1
4
27
-------
JD
OJ
O
o
a,
O
E
o
OS
to
K Z'
Z;
d>
2
D
DOSAGE IN
auiiq
Hd
d
|i
rH
CM
in
00
0
^
rH
OO
0
LO
r- 1
I— 1
r*j
t^
oo
o
LO
iH
OO
O
L/1
(M
O
C^J
o>
00
o
0
r-H
00
o
LO
to
C3
CXI
!—)
CTl
O
t^
tH
OO
O
in
•*
0
fNJ
LO
Ol
o
Ol
i-H
00
0
un
LO
0
OJ
r~
01
o
i-H
i-H
rH
00
O
If)
v£>
(U
4J
O rH -P
rH O MH rH
X X U)
•H X r< C
S -H OJ -H
STSrH
13 C M
•H > O O
p.0 O.H
") rH Ct> (H
Oi W CO H
6
O
o
ft.
13
03
K
0
u
0
U
in
4
>•
t-
ALKALIh
K
O
u
0
>
i
<^
_
<
=
C
8
Ud
=
C
V
u
o
L_
I
u
u
s
C
V
Q
0
m
;i
Q
g
..
li
U
OD
3
J
G
V.
bO
CN1
2
'
I
00
r
\
O
in a.
: i
: o
28
-------
rt
LJ
(S
o
r.
4-"
M
E Z
2
o>
5
0
o
<
r-
(/>
*
4
\
r
z
IU
<
W
o
o
a
<
ro
s!
o
(fl
O
""^
0
r-
o
I
0
ro
0
2
ro
O
0
OX
1- a-
ro
O
o
I
O
X
a
Z13V3
Hd
d
o
o
rH
o
i — i
0
LO
O
O
*^
O
cvi
06
00
o
LD
CN]
O
rH
O
U/1
tn
t^
i — i
LO
GO
0
o
to
o
rH
O
LO
^O
•H 3
6 R
•H
O g
CN1
O
o
o
C
0)
3
o r^ l
o '
rH O I
e. o
(fl rH
Pi CO
e
o
u
O
e.
K)
a!
T3
rt
a,
<
U
r-
OT
IE
V)
U
H
v>
(-
U
<
IX
<
10
o
m
o
u
-------
•s
E-
O
(NJ
a
O
(X
CO
U
IB
O
(U
c
u
•H
^
•P
Z
to
a z
z
*»
D
O
HARDNESS
b£
_l
<
••!
•f
o
K
0
Z
o
10
o
¥
m
o
o
Soi
KH 0.
]C
<
10
8
X
o
X
a
V
r
2
in
I
o
-OD^O
d
It
t^
rg
00
LO
O
LO
rH
O
•O
i-H
t^
CM
00
LO
o
00
i-H
O
^0
f-J
t--
(M
00
LO
O
iH
1X1
O
VO
to
r^
OJ
-
CO
LO
0
»s-
o
U
O
o
<0
4J (fl
3 eo
•*.* 3
•p o e c
0) CO -H
CMLO 6
X KrH
4-> O
18 ^ !H rH
J3 & O
M-f f-!
(-1 00 O
(U • S M-l
is ee s 3 tu
>0 O, rH 4->
•M O Bi 4H i-H
C o m -H
•H T) iH O W u_
v to
•O 4-> X X M
I-H tn -H x ^ «
(D 3 S -H 18 -H
2 T3 fH
I/I 18 -H S O <->
(-1 & O U -H
18 33 13 i-l •
t-
z
3
<
*
_i
<
K
0
a
in
4
in
TOTAL HARDNES
)
f
O
•
U
2
X
=
a
u
.8
a
U
V
a
!
i
0
n 0.
i
0
30
-------
CS
H
O
OH
V - /
rt
O
g
c
0)
>-
4-1
3
M
X Z
9
X
0
u
HARDNESS
x
-1
<
K)
=1
in
»O
s<3
u
t-
u
X
u
n
o
£
10
o
o
CD
CM
LO
CM
00
LO
0
oo
0
VO
to
LO
CM
oo
LO
o
Ol
0
vO
•5J-
UO
CM
oo
LO
0
o
rH
o
>o
LO
LO
CM
00
LO
0
rH
t-(
O
*O
vO
3 i/>
C (U
•^tCJ -H 4->
oo e 3
c/i ^o C
CMLO O -H
^01 4->
X r. , LO C
rO 3 ^
oo rt (X 4-> P-,
• t- oi 2 c
LO O (X I) rH
fxo pi 3 O
o g o t-i 4-1
4-> ^-1 X X M
II
0
u
J
~
y
o
H
o
0 U
a
U
V
X
3
V)
U
3
a
•>
Q
O
U
_j
<
O
h
£
C.
U
O
«f
U
0
r-
^
H
0
-------
rt
JC
ex
U)
o
e
o
g
§
• H
U
I—I
«
O
V
a:
•I-J
3
to
= Z
•z.
»
D
U
HARDNESS
x
_)
<
*$
O
F)
•5
»-
U
z
u
m
o
2
m
o
u
&*
rH
i-H
t~-
I~~
LO
cn
o
ro
tn
vo
rH
(M
r-
r-
cr>
LO
to
LO
vO
iH
K)
to
00
O>
01
0
Tj-
LO
vO
i-(
Tf
O
CTl
i— 1
0
i-H
LO
*t
LO
VO
rH
in
o
01
'd-
o
rH
0
in
LO
VO
iH
S-i
O
C
o
cs
•H
O
e
o
o
(X
sc
01
•O
13
cfl
OH
H
O
U
o
o
3
>-
t-
3
-1
<
^
_l
<
f
O
u
a
> O
V
in
in
u
z
a
K
<
I
_
I
^)
»
C
•
U
3
J
;
C.
3
u
H
O
n u
a
O
in
<
,
3
ai
u
z
1
i
C
C
u
1
C
C
u
u.
_
a
=
C
V
u
g
1
u
e
'l!
_
C
V
o
C
«
1
1
i
a
s
b
g
:c
«
to
o
1— 1
2
1
3
• —
^
o
o a.
i
0
32
-------
Nutrient Removal as Calcium \mmonium Phosphate
~7-
2
O
HARDNESS
5
0
fO
* o
-
o
z
u
K>
O
ro
0
O
m
*t O I
l- t- a
CO
4
ro
O
O
I
O
X
E
z
UI
CO
o
o
*OSBD
d
< 5
-3
O
O
r-J
O
Ot
O
o
to
o
to
to
-*
o
r-
rsj
O
O
O
ro
to
C-J
O
t— 4
.
CT-.
0
O
to
0
to
to
to
0
01
o
o
to
0
to
to
^
T-H
CO
en
o
o
to
o
to
to
">
0
LD
1 — 1
O
C
i — I
C
o
to
c
to
to
\o
o a
~~ c
O O C3
T- --^ O
+J U ~ T3
CHARACTERISTICS OF RAW WASTEWATER Comments:
ALKALINITY AS CaCOj
TOTAL HARDNESS As CoCO 3
CALCIUM AA C.COj
MAGNESIUM AS CoCOj
TOTAL COO
SOI UB( F TOO
o
a.
33
-------
CJ
4->
C.
G,
3
•rH
C
e
<^
o E
M 3
•rH
rH rH
X> (d
R U
t/>
n
O
e
OS
4-i
0)
• H
rH
3
2
to
— 2:
1
D
U
W
UJ
z
o
(E
4
Z
„
ID
<
K
^
X.
z
UJ
4
V)
o
o
c
10
u
•5
r-
4-1
e
E
0
u
a:
u
t- r^
4
* s ^
in
» "°
? °
Z y
Oo«'O
" * " u
<" o, 0 0
o <" o o . . . . . .
~ ?Ijt-_l 100.
I_1O4
-------
0)
4^
ca
x:
C-
V)
o
E
C
o
E
E
3
'H
I — 1
a
u
a
ca
o
E
+-i
CD
• r-l
i-l
2:
— 2*.
5
0
u
(ft
UJ
z
o
K
Z
^
•*
CD
tn
^
"*
>
Z
tu
DOSAG
a
«
M o
o o
tfl
u
1-
0
z
u
10
o
¥
K>
0
(J
0 X
h- »
IO
o
0
X
o
z
a
(X
!i
o
o>
0
r— 1
C
O
ai
o
rj
LO
CO
1 — 1
cjT
vO
to
un
00
rH
Ol
C
-
00
1 — (
00
en
o
LO
0
CXI
o
0
1 — 1
o
VO
«
D
• H
SH
O
i — 1
E
3
• H
0 0
E co
e ^ -=t
< p
C rt
rt T3 U
CD
cu u) ca
+J 3
^ T3 O
ca ro
(X -0
ca ~ a
H C. (1.
vi
4->
O
E
e
O
tc
UJ
4
UJ
in
J 0
2 *
K» 0
o o « o
" < 10 U
« ° « ° °
0 W> (i O
0
- > * « "
K i_ o: S _ o
UJ - < * C
o — ? co
I _l O < ^ C
g
8
35
-------
that the solubility data obtained for CaNH4POtf«H20 might not be
equilibrium values, the reason being that the solution produced by
the incongruent dissolution process shifts progressively in compo-
sition toward the mono- or di-ammonium phosphate fields (44.) . As
this shift occurs, new solid phases can appear in the following
sequence: apatite, octacalcium phosphate, dicalcium phosphate
dihydrate, and finally Ca(NH4)2(HPOit)2'H20. Once apatite forms,
it tends to persist as a metastable solid phase, thereby preventing
solubility measurements of CaNHltPOtt«H20 under true equilibrium
conditions.
Table 22 shows the analytical results obtained by adding
various amounts of CaNH^PO^-l^O to distilled water and intermittently
stirring over several time periods. A specific ion electrode was used
for these ammonia determinations.* The high ammonia content of these
solutions preclude further consideration as a means for nutrient
removal.
Table 22
Solubility of Calcium Ammonium Phosphate
(Distilled Water at 22°C)
Parameters
CaSH,PCVH20
Calcium
Phosphate
Ammonia
mg/£
5000;
12
90
18
mg/Jl
500^
5
60
22
*g/*
5,0002
12
900
150
mg/A
5,000^
5
600
170
^Stirred for 2 minutes.
^Stirred for 120 minutes.
The removal of both ammonia and phosphate in a one-step process
based on the formation of an insoluble metal ammonium phosphate was not
found to be feasible. While these compounds can be quantitavely precipi-
tated under laboratory conditions, their solubilities are too large to
meet effluent standards for municipal waste treatment.
Orion, Inc., Cambridge, Massachusetts.
36
-------
SECTION V. COAGULATION OF SEWAGES
Upon finalizing the metal ammonium phosphate production phase,
the main emphasis of work turned to the evaluation of other parameters
in the coagulation of sewage. Initially, secondary effluent was
selected to be used. However, after a few jar tests, three problem
areas were readily apparent. These were:
1. The removal efficiencies for BOD and COD were on the order
of 55% to 65%.
2. The production of a highly buffered system by the biological
treatment process complicates the coagulation procedure.
3. The most vexacious problem in sewage treatment (production,
treatment, and disposal of biological sludges) would not be
eliminated. This would be true since the coagulation step
was being utilized as a tertiary treatment step.
In addition, the costs of adding another step in sewage treatment must
increase the overall cost regardless of the process. Faced with these
formidable objections, the decision was made to examine the magnesium
carbonate-lime process as the primary treatment step. Consequently,
samples of the raw sewage after comminution were subjected to
coagulation.
Jar Testa
The collection of background data on the Gainesville raw
sewage showed a wide variation in alkalinity (Table 23). Thus, the
amount of lime required for treatment would correspond to these
variations. Corroberation of this type and level of fluctuation was
received from sewage plants in Orlando and Ft. Lauderdale. Initial
jar tests (Tables 24-26) proved highly successful. The BOD and COD
analyses were performed on the wastewater after coagulation and
filtration through Whatman No. 2 filter paper. Although carbonation
was carred out, BOD and COD analyses did not include this step.
Additional removals are possible from the carbonation step as shown
in Table 32b. Table 27 shows the first Total Organic Carbon* (TOC)
analyses conducted. All analyses were conducted before carbonation,
although normally a treatment plant will include this step. Phosphorus
removal was extremely efficient (99%), and the residual is less than
Beckman Total Carbon Analyzer.
37
-------
Table 23
Total Alkalinity Fluctuations
(24-Hour Composite)
Date Time
3-14-72 10 A.M.
11 A.M.
12 M.
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.
10 P.M.
11 P.M.
Alkalinity
190
206
194
188
188
184
180
178
146
148
136
132
. 132
132
Date Time
3-15-72 12 P.M.
1 A.M.
2 A.M.
3 A.M.
4 A.M.
5 A.M.
6 A.M.
7 A.M.
8 A.M.
9 A.M.
Alkalinity
132
130
90
104
102
102
110
120
230
206
Note: Total alkalinity - 154
Total hardness - 100
Calcium as CaC03 - 58
Magnesium as CaCOs - 42
pH - 7.5
38
-------
a
H
oj
DC
CS
03
a:
C
o
a
rH
3
a
o
u
Q
O
U
O
O
05
D»
2
0
u
(ft
bj
Z
O
CE
<
Z
X
<
m
<
K
V)
x
<
\
r
z
LU
SI
O
O
a
<
in
sf
O
in
= 8
u
r-
O
Z
u
10
O
X
in
O
u
O x
I- o.
in
O
o
X
X
a
OUIIT
£032K
1*
CO
rH
^o
\0
*t
oo
CO
CO
Tf
Cxi
o
Ol
CO
CO
t~-
I— 1
r- t
in
VO
rxj
o
LO
t— t
CO
(X)
(XI
*o
o
a>
CO
en
*t
(xi
o
o>
VO
01
CO
!-l
iH
O
t~-
CM
O
\o
tvl
O
(NI
O
CO
o
o
i-H
O
rj
t-4
(Nl
r-l
r-
CO
^t
o>
CO
rH
Lfl
t^
(Nl
O
1^.
rt
r-i
VO
i-H
rvj
CO
(XI
0
vO
CO
CO
vO
0
rH
vO
i-l
CO
CO
vO
a>
CO
rH
rH
0
CO
(XI
o
CO
•*
to
LO
CTl
rH
\D
(XI
•&
LO
(XI
OO
t
o
rH
vD
i— t
CO
CO
O
O
rH
CO
rH
rH
0
CO
(XI
o
oo
LO
UO
LO
rH
tM
"»
(M
(M
OO
O
O
rH
*O
(M
rH
*0
rH
CO
00
CO
0
rH
VO
rH
rH
LO
OO
(XI
O
G>
VO
00
C
•H
rH
LO 4J
+J
O co to
•P tn
!-< in
O T3 (U
•H -P
!-i n3 3
£X T3 p d
ni 1/1 .H
in (U £
M a)
0 ess '
P5 C rH
O rH VO rt
_
E;<«"2-§ " Q •
" ^ o * 8 o o
S^SS oo'-o*
2t<**0 0 g
S?1'!01!!1-^"^,, ,
-------
*
0
C£
^?
5
c/;
un S
rsj 03
«
t-H 4-
£ 0
0)
o
•rH
•M
i — I
3
be
rt
u
o
o
u
tr
o
U)
LU
Z
O
K
X
X
4
ID
4
t-
00
o
o
r-<
O
«3-
O
O>
•*r
UO
o
ro
i-<
i— t
o
^D
(Nl
o
*±
I— 1
00
r^
•ri-
oo
CNJ
OO
o
0
1— t
o
^r
o
CTl
eg
LO
00
M
i-H
i— <
i — I
O
1^
<%!
*o
r^
o
r-j
(Nl
t^
o
00
^j-
t
0
CTl
O
tn
o\
to
rt
4->
c
O
o
u
•^1-
LO
IT
O
o
o
m
4
>-
(-
Z
4
V
_J
4
C
O
•H
rbona
CO
o
CJ
^
o
<4--
cy
JD
l/l r
C) 4
t/i
•-I t
C r-
C
3 c.
O >
5 LL
O
O
o
u
o
o
(0
4
in
Ul
z
o
4
I
4
t-
O
1-
in
u
4-»
3
• H
e
o
c
• H
J
-< C
H -H
i — I
3 t-J
H cn
J -d
J o
< 0
oo rg
to -^1-
K
O
10 0
0 o
0 "
Q
O
in
o
o
CD
o
in
i o
io a.
x i
z o
40
-------
o
c/:
3
oc
CO
o
Q
c
u
Q
O
ec
(7
2
o
O
HARDNESS
*
_l
<
rA
=3
si
o
K
-H
rH 4H
•H M ca
x £ E
1/1 CU O
0 J3 4H
3
r-H CD J-i
ca to o
^> CX4H
ca
XrH to
•P S
•H
-------
*
O
cc
r-- .£
cxi «
c
o
JS
ZcJ
o
u
o
H
to
X
O
o
u
Q
O
ca
c*
o
o
c
to
f— *
oo
(NJ
1 — 1
<*
r-H
LT)
1^
(NJ
O
o
IT)
i— 1
to
«3-
U1
^
(NJ
>c
c
(NJ
to
C:
•*±
r— '
C
^1-
i-n
rH
O
C
to
o
o
*0
o
T-H
e.
e
t/)
X
t-<
K".
OC
•K
OC.
UJ
<
u
tf>
4
$
(t
o
u>
i-
w>
lu
K-
u
K
4
X
U
o
CTi
'E,
O
4-J
•u
o
M
r— t
'j^
a
*j
«
\o
^t
o
1—j
d
E
«
u~:
v>
C
0
c
E
O
o
r-\
'
X
• H
C
•H
^-t
a
^
r— H
re
r-H
OJ
•I-J
O
E-
O
tn
.
w
i/;
O
c
f-.
rt
r^
t— t
r;
*->
C
L_
C
1 — 1
.-H
O
U
y.
o
to
u
0
^-i
'—•
at.
C
• H
r-*
+->
+->
O
V.
"3
O
O
10
o
00 OJ O O C O
(NJ ro to L^ OC CO
I (NJ tO ^- LT, \C
4
K
_l
3 2
8
O
£
O
8
u
o
o
o
I-
O
o
o
to
o
o
o>
O
a.
42
-------
any level yet reported in the literature (Table 28). Other parameters
are shown for comparative purposes. An even higher removal efficiency
is assured since adsorption by granular activated carbon would be part
of this treatment process.
Table 28
Chemical Treatment of Raw Sewage
Authors
Villiers (46)
Weber (47)
Smith (48)
Smith (48)
Hannah (49)
Bishop (50)
This Research
Treatment
Method
Lime
Clarification^
Ferric
Chloride^
Limec
Lime +
Ferric Floe"
Limee
Lime +
Ferric IronJ
MgC03«3H20
+ Lime^
Residuals
Effluent Quality (mg/&)
COD BOD TOC NH3 P
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
187
84
—
420
125
420
1 92
265
66
.. 347
66
387
54
78
39
65
15
100
20
100
20
139
28
142
31
136
19
79 —
36 —
70 35
30 20+
—
—
78
23
118 —
26 —
134 33
29 22
9.2*
0.3
80*
<5
6**
0.7
6**
0.7
10*
0.4
8.7*
0.3
4.5**
0.02
^Dosage - 150-300 mg/X, (lime form not reported).
^Dosage - 200-350 mg/fc.
GDosage - 150 and 300 mg/X, (lime as Ca(OH)2).
^Dosage - 100 and 175 mg/X, (lime as Ca(OH)2).
^Dosage - 350-400 mg/X, (lime as CaO).
•'Dosage - 350 mg/X, and 5 mg/£ (lime as CaO).
^Dosage - 100 and 275 mg/£ (MgC03«3H20 and 98% Ca(OH)2)
**:
Total P.
Ortho P.
43
-------
Drum Tests
Table 29 shows the results of the first drum test. The
wastewater was not carbonated to pH 9.0 prior to analyses for BOD
and COD. The purpose of this drum test was to recover MgC03*3H20
from the sludge produced. A recovery of 97% was obtained.
Clinoptilolite
Clinoptilolite is a natural exchange material selective for
the ammonium ion. This material was used on bench scale tests by
Sullivan (51) at the University of Florida. Ammonia residuals from
feedwaters containing 2 to 5 mg/X, NH3-N averaged 0.12 to 0.32 mg/Jl
NH3-N.
Mercer (52) further studied this compound on a laboratory and
pilot plant scale. Laboratory tests showed 99% removal, while pilot
plant studies produced 97% removal (16 mg/£ to 1.5 mg/Jl) at 6 gpm/sq ft
flow rate.
Lime is the regenerant for the ammonia saturated Clinoptilolite.
The volume of liquid waste from the regeneration step is small and, of
course, high in ammonia. This waste liquid was air stripped by Mercer,
but low efficiency was observed for ammonia removal.
If succesful, this process would be adaptable to the Gainesville
project, since
1. Lime is recovered and recycled in the normal process, and a
portion of this lime could be used for regeneration.
2. The concentrated ammonia waste stream could be treated with
sulfuric acid to produce ammonium sulfate, a valuable
fertilizer.
Table 30 shows the initial ammonia removal test using the
_ i material.* The
specific ion electrode.'''
exchange material.* The analyses were performed using the Orion
Benefit of the Magnesium Ion
Numerous jar and drum tests designed to study the several
variables involved in the coagulation, flocculation, and settling
of the untreated raw waste of the City of Gainesville have been
presented. The data have indicated but have not clearly proved
*W. R. Grace Co., Clarkesville, Maryland
^Orion, Inc., Cambridge, Massachusetts.
44
-------
•3
H
ca
£>
CM
o
CM
r-t
vD
-3-
r-l
vO
r-^
CM
0\
co
t-H
rH
O
m
CM
Vf
r~
t-4
r-i
o-
in
CM
O
i-H
0
CO
l-(
o
<•
1-1
CM
r-
O
O
i-l
-3-
r-l
rH
in
VD
CM
,£* CU
y T\ cu
11 3 "o
T3 CO
O u 42 cd
B Cd 4-1 O p
td C r-l
o o 1-1 co cu
cu & v-i cd £
•a t-i 3 4J
cd m
co O o X C
cd M T-I
o
o
•• 3
vr cd
VI
-•g
CO
iH
P.
cd M
CO -O
Cd
CJ
o
• cd
w
01 o
o o
O -. cu 3 c
CO J3 CO B
> 4J 60 M
O T3 01
O CO 3 4J
CU 4^ r-l CU
M 4J 0) "O
C
CO
ca
cu •
U 4J
P. C
cd
B M
3 3
•H 60
w o
if
^
|i
O
C
u
CJ
o
r
a
r> O.
1
o
-------
Table 30
Ammonia Removal
Source mg/£ as NH3
Raw sewage 12
Coagulated and filtered sewage 9
After clinoptilolite <0.6
that the addition of MgCC>3 as a recycled coagulant provides settled
effluents superior to those where lime alone is used, and superior
to the effluents produced by other investigators employing lime alone.
However, most waters contain some magnesium which may or may not have
precipitated during the work of others and which may have influenced
the data obtained. It was decided, therefore, to begin a series of
jar tests designed to definitely establish, if possible, the fact
that the addition of recycled MgC03 is capable of producing results
superior to those produced with lime alone. The problem was to secure
a municipal wastewater low in magnesium since both the untreated
Gainesville and University of Florida wastewaters contain from 28-48
ppm magnesium expressed as CaCO^. Untreated municipal wastes were
shipped to Gainesville from North Miami and Montgomery, Alabama.
Waste from North Miami contained almost as much Mg as Gainesville
waste, but that from Montgomery, which has a very soft river water as
its source of water, contained only 8-12 ppm magnesium as CaC03 or
1-3 ppm as Mg"1"1".
Gainesville, Florida, Sewage
Table 31 presents the data obtained from an exploratory jar
test in which increasing dosages of lime only were added to jars 1-3
and the same dosages of lime plus 80 ppm MgC03 were added to jars 4-6.
Increasing lime dosages showed higher removal efficiencies. The same
lime dosages when combined with the magnesium ion removed less COD
until a hydroxide alkalinity of 140 mg/£ was reached. At this point,
the lowest COD residual of all jars tested was observed. Of course,
one must realize that the wastewater being tested naturally contains
about 8 mg/£ magnesium ion. Upon observing this action, a series of
tests was designed to show the effectiveness of increasing the
magnesium ion while maintaining a hydroxide alkalinity of 140 mg/&.
Table 32a demonstrates the observed results. The first two jars
were set up to determine the effect of lime alone at this hydroxide
alkalinity and also to show the reproducibility of our testing method.
46
-------
E
•H
_4
to
O
u
•o
c
rt
(U
e
re
>
o
e
cu
c:
Q
o
u
0)
^
V
4->
cfl
cfl
^
0)
4)
C
Q
O
U
c»
o
u
in
UJ
z
o
a
<
i
*
<
CD
<
1-
*
<
vl
f
Z
UJ
<
V)
o
0
0
•
TO
••I
Ifl
«o
S<*
o
I-
o
z
u
IO
o
*
io
o
u
0 I
1- 1
in
o
o
I
O
I
a
auifT
£03^N
li
*o
r-
o
•*t
o
oo
I— (
o
I— 1
o
o
r— 1
<
I— 1
to
[^
o
vO
O
1^
N
rH
O
O
rj
i
CM
un
VO
0
oo
o
00
r-t
VO
I-H
0
0
N1
1
to
^0
00
o
oo
Csl
o
*
o
o
o
i-H
O
OO
*3-
CsJ
00
o
o
CsJ
o
t^
rH
O
o
CN]
o
oo
LO
I— t
vO
c
o
1 — t
c
•*J
1 — I
LO
I— 1
o
o
Kl
o
oo
U2
4
UJ
t-
in
*
a.
u.
o
en
o
u
<
a.
<
x
to
o
u
10
o
(O o
8 5
47
-------
fsl
to
in
3
V
O
o.
in
O
CL,
a
•p
o
H
g
CJ
o
a
o
o
0)
ci
o
e
•H
_l
to
O
ac
in
CB
o
C
CL,
H
U
O
H
Q
O
U
9
O
O
V)
111
z
0
o:
<
X
V
<
m
<
i-
x
_i
4
^
?
z
Ul
o
Q
1
<
A
•!
-f
»-
u
z
o
m
1
in
O
o
e*
IO
8
g
X
&
aunq
£008N
N
o
\D
rH
»*
CT1
f^
1^-
fsj
IO
f-a
rH
O
LO
i— 1
>0
I—I
oo
IO
to
I— 1
rH
LO
Ol
CM
O
i— 1
^f
•*
00
to
^O
f^
r^
N
to
Csl
I— 1
0
LO
rH
O
(M
OO
IO
to
rH
r-l
LO
O)
C-l
0
r-J
vo
rH
IO
to
«3-
t~-
LO
»
(O
C-4
i— t
to
to
o>
\o
CVJ
rj
oo
rj
0
LO
I— 1
vO
rH
fvl
Tf
«a-
rH
rH
O
LO
«0
0
IO
*t
o
rH
LO
to
OO
\0
vO
i-H
00
to
•*
LO
i-H
0
o
o
VO
«t
rH
rH
LO
f-
to
LO
f~
LO
CTi
O
to
to
LO
•£>
LO
rH
to
*!•
00
LO
rH
NO
o>
tM
1^-
*3-
rH
rH
0
0
"3-
O
O
rH
\O
in
0
0
rH
t4H
avier
(U
,c
*o
c
rt
M
H
rt
rH
^,
rH
§
• H
V)
V)
O
U
K
U
5
r-
m
1
o Cxi
Ctf r<
tn o)
H 60 U
o) C
•r->'H LH
rH O
C <-•
(I) 4-1 4_>
a> a> u
? «
4-1
4) 4J rH
O 4) O
C^i MH
(D
(H X^3
OJ rH CM
m
•H *H O
t) W rH
m .a
is aj cd
rV! «
r< 0 0)
« (H 00
rH N K) ^ LO VO
fl
' O
10 U
8 5
5
a
o
O
i 5
a s
o
8
o
3
-I
o
o
3
8
* w
i" '
48
-------
The results showed generally good agreement. The last four jars were
designed to maintain a constant pH while increasing the magnesium ion.
Lime had to be increased also in order to maintain a constant pH and
precipitate the additional magnesium ion. A measurable increase -of
COD removed occurred with increasing magnesium precipitation. At the
same time, samples for TOC and total P were collected and sent to a
private laboratory for analyses. The results still show significantly
lower residuals for total P and better removal efficiencies for TOC
when compared to other methods (Table 28).
Another point of interest has been whether carbonation would
effect a further reduction in the parameters under consideration.
Table 32b shows the results obtained by carbonating the supernatant
from the previous jar test (Table 32a). While no significant decrease
in TOC or total P occurred, a measurable reduction in COD did take
place in all jars. The TOC and total P results were provided by a
private laboratory. The total hardness was reduced from 180 mg/£ to
52 mg/£ in the highest magnesium dosed jar.
Table 33 shows the effect of increasing pH on COD removal
for lime alone and magnesium plus lime. A significantly lower COD
residual was obtained using magnesium plus lime at all comparative
pH values. In addition, when comparing lime alone at pH 11.5 with
magnesium plus lime at pH 11.4, a much better COD removal is observed
at the lower pH value. Thus, magnesium does contribute to higher
removal efficiencies as judged by COD reduction.
Tables 34-37 show four series of jar tests designed to evaluate
the mg/& of COD removed per mg of magnesium. Two Gainesville waste-
waters were examined: (1) a COD of 374 mg/£ , and (2) a COD of 520
mg/£. Samples were run in duplicate from pH 11.1—11.6 for each 0.1
change in pH. The COD removal/mg of magnesium precipitated was
different for each set of duplicate jars. However, if the 12 jars
in each series are averaged, the removal of COD/mg of magnesium for
the 374 mg/£ COD waste is 17.6 mg/£ COD removed/mg/£ magnesium
precipitated and for the 520 mg/& waste the removal is 20.1 mg/£
COD removed/mg/£ of magnesium precipitated. The total magnesium
ion present in each jar (present naturally plus added) was almost
the same, namely, 28.8 mg/& magnesium for the 374 mg/£ COD and 29.8
mg/£ for the 520 mg/£ COD.
Table 38 was designed to investigate the reductions in values
for soluble, insoluble and total COD, BOD, and TOC values by use of
combinations of MgC03 and lime.
The data indicate that the dosages of MgC03'3H20 idded differed
too little to significantly affect the parameters investigated. The
slight improvement was, however, proportional to Mg prec:pitated.
The stabilized and filtered effluents were passed through
granular carbon and ion exchange columns, with the samf dramatic
49
-------
XI
rt
f-
3
h
O
o
a.
a
4->
o
n!
U
O
E-
O
8
O
E
tu
OC
o
(U
c
n)
to
o
w.
c
• H
in
it
V
L,
u
c
o
IU
w
a,
£-
u
0
E-
Q
O
U
d>
2
d
u
in
UJ
z
a
-
u
z
u
10
o
¥
10
o
u
01
1- tt
10
8
§
z
a
SUITq
£03%
N
to
to
1-^
to
VO
vo
oo
rH
0
t^
oo
CO
IN]
Tt
^f
•3-
O
CT.
IO
01
1^1
O
t— I
to
to
01
to
i--
vO
«*
Ol
o
Tl-
^r
•<*
o
CTi
IT)
Ol
N
0
rj
•^r
1-1
>o
to
^r
vO
oo
rH
(NJ
•a-
o
O
rH
(NI
T
00
LO
VO
0
vO
•«*
O
o>
0
LO
to
o
u^
<*
t— i
I— I
to
to
o
VO
to
rH
O
^t
to
LO
•O
0
«*
f
O
o>
in
t*~
to
LO
t-~
LO
1— *
1-1
rj-
to
o
VD
(NJ
r-l
O
•^l-
Csl
LO
vO
O
^r
ra-
o
CT>
o
o
"9-
O
O
1— t
VO
Q
O
U
i— t
ea
C
o
• H
4-1
T)
"0
n)
V)
(U
£>
o
6
V*
c
o
rt
c
o
42
r-i
a
u
/ — i
to
o
u
rt
U
!/)
n)
V ^
T3
4_)
P.
a
4.
4.
M
^
LO LO r-j (NJ r*3 I-H
i-H rH tO LO r- O)
.
O
Z
t-.
C3
•-3
p-H CNJ tO r± LO vO
g
O
u
m
O
10 o '
o 10
" 2 ' o
O < IO O
0 « 8 5
3 23 «
50
-------
o
Csl
I
to
N1
O
CJ
QC
c
O
TJ
T)
O
TJ
C
O
3
•o
4)
Q
O
o
4) V
OS >
O
<*> E
Q
O
U
f
o
u
«n
m
u
z
o
K
4
X
X
<
•
M
V
4
^
r
z
0
«
«
1"
•f
t-
g
u
to
O
X
m
O
o
?i
»
8
§
X
a
3UITT
^OO^H
N
00
r~
M
oo
«»
ro
O
^-
O
r^
oo
o
CM
O
t-~
1— 1
l-l
1— 1
o
o
CM
1
1
l-l
o
00
vO
f~
CM
t^
^-
C4
vO
o>
oo
«M
NO
l~-
I-H
l-l
r-(
O
iH
rg
o
t~.
r-i
oo
t~
<-<
oo
00
«M
»»
ff>
N
VO
«•
00
rH
IH
F-4
in
«M
N
i
i
K)
P4
OO
OO
«
r-j
vO
O
CM
tM
00
C-l
a>
1-4
O
O
f-H
10
I-l
l-l
U)
ro
h-
VD
t^
00
CM
tM
rt
in
«M
CM
VO
l~-
rf
VO
IH
00
I-H
iH
»*
l-(
p-l
t/)
NO
CM
O
t-.
00
c«
a.
o
u
u
3
K
ALKALII
f
C
U
3
n
:
5
_
i
*
g
•
3
i
j
^
«
o
V O
a
^
3
5
»
§
o
-1
i
§
U
5
8
o
1
S
!
i
1
0
s
X
1
.
.
^
o
l> ft
0
51
-------
0)
3
C
C
O
U
to
o
0
0
o\« E
Q
o
u
c>
2
o
L>
HARDNESS
*
-J
<
A
s§
u
S3
o
r-
O
Z
u
ro
O
X
10
o
0
Sox
t- 1- tt
«
*
_i
4
ro
O
u
x
o
X
a
^
r
2
UJ
3
VI
o
0
3UITT
^oogw
t/1
CD
C
• H
rt
(J
CD
i— 1
t
E
aj
t/i
n
n)
^H
tj:
•K
•H -H
e E
T3 ?
•H O
P-rH
rt t/)
i
2
4-J
3
C
•H
E
LO
tfi
4->
C
O
E
E
0
u
2
Cu
CC
LO
ro
CSJ
t/)
O
4-1
3
C
• H
g
o
ro
-H
s£
in rt
rt " — i
CD rH
60 i-H
r< rt
rH C
• H
VI
rt ^-^
2
rt
(D
a
a.
4_>
o
C
t/i
a>
o
13
U
O
1 — 1
tL,
p.
«
0
o
rH
- — '
*
LH
E
T3
• H
P.
ra
IH
!H i-H
rt o 4->
U H (U
to to
T3 •
C tor)
rt o C
U rt
U bo
O 2 U
rH O
UH X rH
bo^w
<4H 3
O O r.
C
tD O
o
c -a
rt 'O
r* rt
rt
(1) 4-*
P- O
P* C
rt
T3
4) -H
•M
0)
E S
O
ti_
0
bo
f-,
rt
r-H
JS
o
3
E
t
t/i
i-<
rt
•r—
be
2
much clearer on
-o
C
rt
cc
C
• H
^
(U
rH C
V< -H -H
0) MH rH
4_) 4_t
Ul r- *->
rt (D O
<4H 4-> w
tn
rt
U-
TJ
4->
P.
d
•f
4-
b£
S
O
rS
]^|
rt
i— ^
Or—t \O rH O <
0
o
oo
CN]
O
u
•o
rH
oo ^r ^j- rvi
CTl vO to oo
to
o
0
10 0 •
o 10
o • o
" < n o
w ° o
2 f ¥ °
* I 5 < §
5 1 ' ' § u, 1
- * "> -1
s* i- j 2 i- -i ?
J O < < O O O
< h- 0 X r- CO H
BOO
SOLUBLE
- N
»
o
Q.
52
-------
O
i-
•d
4-"
rt
C
O
e
>H
in
0)
e
X
.0
o
Q
O
U
0
o
CJ
£
o
u
in
iu
z
a
E
5
^
4
f
V)
X
"
^
f
s
i
a
«
IO
FJ
O
r-
O
z
o
X
IO
o
u
ox
r- 0.
IO
8
X
o
X
a
Z(HO)
SODSW
jg
s
LH
CM
IO
f-
to
Ol
Ol
1
r-
o
rH
rH
vD
CM
O
Ol
o
00
o
oo
m
rH
rH
O
to
CM
rH
rH
t—
in
CM
to
to
01
01
1
r-
o
1-1
rH
CM
O
01
00
CM
m
i-H
rH
O
to
in
CM
rH
CM
Ol
vO
OO
to
5
o
CM
0
rH
•=)-
O
01
OO
OO
vD
10
**
rH
rH
O
O
in
CM
rH
to
r-
oo
to
CM
o
T1
CM
O
CM
rH
CM
O
Ol
00
s
^p
f-H
rH
O
O
CM
rH
"*
M
rH
0
K)
CM
CTi
f^]
OJ
rH
rH
s
o
Ol
0
o
rH
OO
0
to
rH
rH
0
to
in
CM
rH
in
to
CM
rH
O
to
CM
vO
01
CM
CM
rH
rH
vO
CM
0
01
vO
Ol
CM
iH
to
rH
rH
0
to
in
CM
rH
*
a.
&.
0
I/I
(U
minu
in
i
X
• H
T)
P,
a
&.
in
(U
o
u
K
Ul
i
*
u.
0
in
v>
la
K
<
X
U
t
11
r-
<-
t-
a
u
2 i
a,
x
o b
£
CS) ,£
in T:
0) F
•P n
minu
0 L
to
u
i a
r-
X f.
£p
I/
O r-
rH i-
to <
00
rH
U
a
u
;
3
-i
<
j
)
^
i
t
>
j
i
s
1
,
1
!
j in
(U
) -4->
3
> C
> -H
E
"in
C
«H
CM
Ol
II
0
Cl
a
V
in
M
C
IT
\-
r<
O 0)
tH rH
T3 6
K in /— v
•H frO
E ^3 O
rt U
tn r< ca
4) O C_)
rH
ft, 2
e < CB
« ^-/
in o
O T3
TJ •• 4J
4) O> &
r- C 0.
4) O 3
4J -H a) +
rH V OS +
•H rt 04
^ t4 -H .H
0) J3 >
U rt in o
rH in C
CL, .H J-.
3 rt c«
Q C3 *~3
Tt 00 Tt
VO CM t^
t
0
IO O
S 5 . .
35 . §
9 o u
§ 8 u, g
s »> zj
J |jj J ffl j
o x ? 8 t?
o t— to
rH O O
rH i— t rH
CMTt \O
I I I
rH K) in
o • o
e o s . .
0
u • u
3 -> 1 * *
3 Z 3 » 2
8 g 8 i o
53
-------
X!
(0
E-
OJ
g
•O
(U
O
H
Tfl
rt
(3
O
J3
rt
I
0)
C.
ol
a)
O
a
o
Q
o
C_)
9
x
o
o
HARDNESS
X
j
<
ro
=s
«f
0
i-
u
z
u
to
O
¥
10
O
U
ID
~
00
o
r- 1
o
oo
rH
(Nl
O
Ol
vO
LO
o
Ol
rH
0
rH
to
U5
(NJ
to
o
00
o
*o
oo
oo
o
rH
(NJ
t^
1
OO
o
rH
C
OO
I-H
(Nl
O
ai
o
\o
o
01
I—t
o
rH
to
IO
CN1
>*
Ol
00
(NJ
oo
(NJ
Tt
to
rH
Tt
r-.
^t
to
rH
OO
O
(NJ
1 — 1
o
CT>
Tt
oo
^t
00
o
0
oo
(NJ
LO
CNJ
LO
rH
01
(NJ
OO
(NJ
•*t
to
1 — 1
*3"
t-~
1
<*
to
i — 1
00
o
(NJ
rH
0
Ol
(NJ
Ol
o
oo
o
o
oo
(NJ
LO
(NJ
vO
ER Comments: Rapid mix @ 70 RPM - 5 minutes
RAW WASTEWAT
0
in
u
t-
v>
K
IU
t-
u
K
4
X
U
Slow mix e 40 KFM - .50 minutes
ALKALINITY AS CaCO. 1SH D,^-;^i,r oQ<-ti^r,^ ^1 ^^ . c m^,,<-Q<:
j~>
in
cs
o
o
rH
i-l
.
*->
JC
w
• H
rH
(/)
r<
ca
t<
O
o
o
u
ffl
4
VI
(/]
u
a
c
K
<
I
J
4
E
X
13
3
O
rH
O
OJ
N
C8
X
O
rH
rH
V
X
VO
„
LSi
f
to
!/)
r<
rt
^
^>
vO
r>
1
s
s
*
2
c.
^
<
c
•H C
O -r
-IT:
« a
> -i-
o (-
6 a
O r-
r- M-
a
—i i.
CS
t-1 >
O i-
4-J r
CT
•M a
t/i r-
0 L
E
— 1 U
m -r
-H +
O +
t
^ ^
c
H "C
O Q
CUT
•^
(/) 0
H
js
oo
(Nl
F
O
0 0
0
0
1
I
TOTAL COD ^^ appearance of Jars 1-2 compared to
XI
rt
r<
O (NJ
O t—
CT\ LT>
rH
O -V.
U.LO
rt
3 -o
0) <1J
co *-"
u
3 P.
in E
o *
C in
•H
OS
o
o
u
o
v>
u
o
to
O
U
u
rt
V 1
•a
a
P.
tO LTi tO
r^ LO to
l ^j- \o
( IO LT.
o
o
a
O
in
o
O
a>
O
UI
I O
10 a.
-------
to
•
V)
1C
_J
<
CO
o
u
s
X
a
X
r
2
Ul
o
5
o
o
auiTq
£03§N
|i
vO
LO
CM
CM
f!
rt
tO
O
O
,-H
^1-
CM
vO
r~-
CM
vO
i-H
vO
i-H
i-i
O
to
f
tn
CM
l-H
i-H
vO
r~-
o
vO
i-l
vO
iH
rH
0
to
•*
LO
CM
i— 1
CM
GO
LO
*O
\O
10
CM
t
O
rH
i-H
O
CM
O
00
o
f
1— t
LO
iH
i— 1
O
O
"3-
LO
CM
iH
to
00
f-
vD
'*•-,
t-H
LO
rH
i— 1
0
O
^3-
UO
CM
i-l
^1-
i— t
^O
00
«*
to
CM
^t
>O
i-4
i—4
VO
i— 1
O
OO
VO
CM
r-t
•>a-
i— i
iH
O
t~-
to
LO
CM
i-H
LO
O
vD
«*
00
VO
CM
i-H
^J-
i-H
i— 1
O
f^
to
LO
CM
i-H
vO
o
to
00
(U
00
oj
Oi
tn
is
ca
fa
0)
I— {
I-l
•H
fc^
f-*
I/I
0)
c
rt
CM
f-
— ^
|>-
1—4
LO
T3
m
m 4->
i-H U
ex m
E •-!
n) I-H
V) O
u
*o
ca a)
^ i-H
bo ex
CD
'O
0)
o
e
Q
O
U
_
b-
g
§
U.
_
ffi
-
_J
o
V
£
;
p
g
UJ
^
™
s
o
0
a
^
H
g
o
.«
u
-
:
c
M
Z
1
r
X
z
o
0 0.
1
0
55
-------
to
41
U
13
O •
J-,
P.
an-
ft,
o
4> 4->
X U
U 4)
O <+H
tn 13
V C
rH CIS
P.
6 in
CO rH
4) in
4-> 4)
CO OS
u
ft
Q
O
6
4)
a:
o
o
u
Q
O
u
fi
X
o
O
(0
UJ
z
Q
K
<
X
*
<
»
4
t-
l/>
X
<
\
f
z
UJ
4
8
O
in
•I
P)
SI
»-
u
z
o
10
o
X
in
o
u
21
IO
§
X
&
auiTq
£003w
N
CM
0
\o
CM
CM
M
OO
L/I
CM
Ol
1
OO
L/}
o
fl
rH
O
CM
CM
OO
cn
rH
rH
o
1-
tO
uo
CM
rH
r-~
*3-
VD
CM
CM
O
rH
rH
rH
O
"3-
to
tn
CM
I-I
CXI
OO
VD
CM
•t
•*
•t
(U
+->
3
C
•H
S
LTi
1
X
• H
E
'O
• H
P.
CO
K
in
4J
C
E
6
O
U
3
C
•H
e
0
rO
£
O
>H
4J
CO
rH
3
U
O
O
rH
H.
1 jars <5 minutes
rH
CO
C
•H
T3
•H
P,
CO
g_t
bt
C
•H
rH
+J
4->
4)
c/5
between duplicate
feet of ApH
r-t
CO
4>
rH
O
4-1
K
a
4->
n)
C
r<
4>
PL
3
CO
4-1
C
4)
S
4)
4)
rH
bo
CO
4J
C
4)
rH
rH
CD
U
X
w
<4H
4)
13
C
CO
in
4)
3
rH
CO
£>
Q
O
U
4J
C
4>
13
'H
£>
4)
^
rH
rl
CO
4)
rH
U
O
CO
to
0 O
•H 6
rH 4>
•H M
£>
in M
D S
C
• H
CO
U
CM
CO \O
^- c —
0 i
1
1 CM
O rH
CO rH
ID
•O 13 C
c c a
CO CO
rH
t~- 01 rH
in in to
!-l rH r-c
rt cO ctt
^^^
vO
OO
r-l
K
0
U
O
o
3
^
z
4
4
*£j-
cn
N
0
u
o
1 O
in
4
in
at
u
z
0
a
X
_i
4
r-
CM
VO
1
•
8
c
„
2
t
.
4
U
CM
to
r
.8
a
u
V
9
2
«
UJ
a
s
a
o
CM
in
f»
c
C
<.
1
V-
c
£
u
i
c
• 0
o
o
1 r-
J
! ^
i p
> o
> t-
u
^"
u
0
c
u
Q
O
ii
i V-
o
i
u
a
«.
c
«
1 2
I ,
I 3
^
o
n a.
i
'• o
56
-------
11
ot.
s
D
C/)
00
to
0)
r—<
^
H
a>
c
c
o
T3
0)
13
C
IS
O
o
Q
O
B5
o
o
p
0
CD
Q
O
u
o
a
u
HARDNESS
*
_;
<
K)
s§
O
F)
„ o
S
O
*
IO
o
u
?g*
*
_j
<
10
o
0
§
x
a
"^
?
Z
UJ
8
o
Z(HO)
ED
O^H£
£OD8^
0
to
o
CT1
*0
CTl
^J-
(Nl
O-J
LO
t— 1
rH
LO
C~-
to
o
vO
rH
CO
CT.
to
*fr
^f
to
LT)
O
r— t
vO
^t
^O
in
o
a>
o
o
i-H
0
to
rg
Ln
i — I
i— t
O
o
•*r
o
OJ
to
*3-
GO
to
-H
3 P,
C oj
•H r-.
E
O
Ln 2
in
4-J
C
(U
E
o
U
to
o
u
B)
u
in
n)
T3
>
O
E
Q)
OS
W
S
O O CM
oo Ol O
rH
1 1 1
rH tO to
V
_l_l
• H
rH
O
rH
• H
+J Q CM O LO
P, 0 . . .
O P3 rH rH O
C
C
•H
C
o
•H Q
4->0
U(_>
3
T3
O
Oi
t*p
CtJ O^ O iH i-4
^J 00 CTi Q\ C_2
O p \o o m
r^ T3 O • • •
r C U rH i-t 0
CC
v) \o r-- oo
C oo oo oo p;
t-( O
& u
• J-i O LT> u^ LO
— ) r** o frO rt E— «
O LO \O vO ^J
w
^4
O
« O
*~) IS
rHtOLO UHCd rHtOLn
"^ ^~3
O
00
rH
rt
O
u
a
u
I
H
Z
4
if
Ol
H
O
u
a
> o
V
U
a
(j
V
<
X
z
u
(!
O
to
Ln
D
a
c
u
_
s
CM
to
rH
g
U
Ul
_
a
3
C
LO
O
rH
o
O
h-
1
1
00
LO
u
e
Ul
-
:
-i
c
rH
vO
rH
O
c
o
t^
a
s
Ul
a
c
M
Z
1
r
X
z
o
n a.
i
0
57
-------
reductions in criteria values observed in other jar tests. The fact
that final TOC values are higher than those for both COD and BOD
could be due to the presence of an unusually stable organic compound
or compounds not oxidized by hot dichromate or biodegraded in the
BOD environment.
North Miami Wastewater
North Miami withdraws water very low in magnesium from shallow
wells. However, upon receiving and analyzing this sample, the magnesium
content was found to be very similar to the Gainesville wastewater.
The magnesium present in the sewage probably results from the presence
of a small amount of sea water entering the sewer lines. A test
(Table 39) very similar to a previous jar test (Table 33) was con-
ducted. The results again showed magnesium plus lime superior to
lime alone in resultant COD residuals.
Table 40 compares the effectiveness of the magnesium process on
two wastewaters run side by side. Sufficient MgC03 was added such
that both contained the same amount. The North Miami wastewater
behaves very much like waters observed in water treatment plants,
in that alkalinity is less than total hardness. However, Gainesville
exhibits a high negative noncarbonate hardness which is not affected
by coagulation.
Table 41 continues the comparison of wastewaters run side by
side. Again, the negative noncarbonate hardness is observed in the
University of Florida sample, although at a much lower level. Finally,
all three wastewaters were composited (Table 42) into one sample and
the effect of mixing time and contact time at high pH were evaluated.
All six samples were dosed similarly and subjected to identical times
of rapid mixing. Then, at six different time periods of slow mixing,
samples were withdrawn and COD determinations were made. The results
show that 15 minutes slow mix is probably all that is required. The
samples were then allowed to stand an additional two hours to evaluate
the benefit of contact time at high pH. A significant reduction in
COD was observed in all jars.
Wastewater Low in Magnesium
On May 18-19, 1972, two samples of raw wastewater from the Catoma
Sewage Treatment Plant, Montgomery, Alabama, were obtained. One sample
was collected at 3:00 A.M. and the other at 8:30 A.M. Analyses of the
wastewaters showed only 3 mg/& and 2 mg/£ of magnesium ion naturally
present. Tables 43-47 present the jar tests conducted on these waste-
waters. The 3:00 A.M. sample had a COD of 500 mg/£ while the 8:30 A.M.
sample had a COD of 1,500 mg/&, the highest COD encountered to date,
and several times higher than normal municipal waste.
58
-------
3
C
•H
to
o
O
c/.
o
2".
Q
O
U
9
0
i
>
to
o
CM
0
LO
O
1-H
00
O
t— 1
•3-
LO
•
LO
OO
CXI
1
-1
(N)
r-j
00
r-*
rH
O
v£>
t— (
CNJ
t— (
s
•
o
0
o
ex]
CXI
f.
LO
O
LO
LT.
VO
OO
o
I— 1
-
•
LO
I— 1
to
o
10
t-~
r— <
o
a>
o
f— i
o
£
•
LO
CXI
to
o
-
r*i
to
o
0
r-j
o
•-*
OO
OO
I— i
r— 1
-
LO
r*j
O
LO
s
LO
LO
t— t
1-1
o
1-1
f-
uO
o
f— 1
0
LO
K)
O
t~-
*
V
e
a
in
rt
oc
^
o
to
OO
C
a>
e
e
O
a:
*
Ul
i
u
o
in
Ul
u
c
Z
u
color
4->
C r
r
V-i •
03
P.C
a
rt
•
T-* C,
•t-"
4-J <
•i-f 4
i-H
"
VI t
n) )
i-H
O- I
rt
(A.
r-
i-H
CXI
K
O
u
o
u
4
>-
Z
_l
4
-J
4
>o
LO
I/I
ti ca
0)
rH Q,
0 3
1 tJ
(U
D CXI (j
H » rt
-* rH O
H i-H
> •- O
ft X
U ^ i
3 rH
H .-H •=»•
i rxi
3-0)
5
J V) LO
« o
•-) t-
0 0
LO CXI
fx) fxl
10
o
(J
o
> u
in
* ¥
y> °
« 5
r? 9
_i =
4 U
& 5
r- U
a) *t
>-. o
o c/;
(U (-<
M X
OC -. rt
-H ^ XI jC
c.
a
* eg to TJ- LO
10
o
u
in
<
in
u
§
o
8
Ul
J
a
O
O
S I
_l
3
U
-I
a
3
j
o
w
o
0.
59
-------
H
OJ
w
ra
0)
to
0
LH
ttt
CX
O
u
0)
u,
a
Q
O
O
c>
X
D
U
HARDNESS
X
_i
<
K>
!!
»-
o
z
u
FO
o
2
ro
O
u
ID
\O
in
00
o
rH
rH
V
O-
r- 1
r-(
O
CO
fO
*O
rH
rH
*
rH
LO
i-H
•^r
oo
m
Oi
i-H
K)
00
O
\O
LO
csi
o
Ol
*O
LO
\o
T-H
rH
rH
V
LO
rH
rH
O
o
*fr
VO
r-l
rH
*
(SI
O
\D
LO
rH
<3-
Ol
Q
rH
A
«t
rH
rH
O
CM
to
(SI
CTl
«
«
LO
O
OO
rxi
rg
00
to
o
\o
•£>
o
rH
O
\O
(SI
•*t
rH
•«•
(SI
O
cr.
^o
to
rH
*O
O
rH
O
T-H
A
LO
i-H
i— 1
O
*t
to
(SI
O)
*
«
VO
*N. Miami
CHARACTERISTICS OF RAW WASTEWATER **Gainesvi lie
sewage behaved as would a rai
eing softened.
le sewage "negative NCH" cami
unchanged.
,0 rH J*
• H -H OO
6 rH > 3
CTJ QJ t/) O
•H 4-) Q> r-<
2 cti ex
3 -H .p
n)
2 U
/ — S
to
O
a
U
in
n)
13
H)
rt
•H LO O d
P.
•H
O
0
r.
a,
00
2
OO O> OO
1 1 1
T— 1 (SI fO
f_4
rt
*~~>
O O O
f^~ OO O^
1 1 1
^t LO \O
e
o
o
* * *
* * *
o
10
O
o
o
in u
o
o in
«
o
_ o
" wo
o °
i * s
O M
O
P3o
0 < <
§
o
o
u
1- u,
o
o 8
o
*
Q.
o
60
-------
0
rt
c
rt
c
o
c
o
C
o
(1) U
ai u
3
0
O
2
o
U
HARDNESS
<
K)
0 0
O
Kl
h-
0
z
o
in
O
'i
10
o
0
CD
-------
o
1—I
.c
«
e
(3
Q
.c o
+-> <_>
O *-M
y. o
"O C
c o
fi .H
+J
n c-
•o s,
•H O
1-. Ifl
<~ O
O
o
>, E
^ c
o o
> •r-'
3
- U
O U
i-l O
t/1I—
d> O
c
•H 4-»
« U
U 0)
i/i .c
0) U
O
0)
V,
3
+-»
X
• H
IS
a
o
u
c
0
u
a>
Z
D
O
HARDNESS
*
_j
<
fO
S8
u
s§
o
I-
0
z
u
10
o
£
K)
o
u
00
«O I
r- (- <*
(/)
X
U
<
m
O
o
I
O
Xjipiqanj,
X
a
\
9
S
Ul
o
4
W
o
Q
1 (HO)
ED
£OD3W
CM
t>.
OO
VO
V
vO
rH
O
hO
•^~
in
CM
*t
00
CM
Tt
to
oo
vO
to
rt
1-
rl-
to
1
•*3-
rt
VO
VO
CM
rH
0
O>
CM
r>~
OO
VO
V
VD
rH
O
to
*}•
LO
CM
in
t~~
CM
rt
Kl
O
rH
00
to
oo
^1-
CM
to
OO
rt
OO
vO
CM
rH
0
o>
CM
t--
o
vO
V
vO
rH
O
to
*t
in
CM
r*4
VD
X
p,
x:
bo
•H
X!
4J
m
4-1
u
cd
4->
(3
O
U
I/I 4->
H cd
XiX!
+->
CM
C
0X1
C -H 4J
O 4-" -H
•H cd 3
4-" rH Q
O<3 4-> O
,-,000
o o cd
(/I O 4-* ^
XI rH C 0)
Cd <4-l O O
U 3
Cm T3
•H CO rH O
4J Cd r-.
0 3 C
in C o 4->
cd -H .H c
^3O< ECL, ininLOinLnin
-i3«fx edctf totototototo
do c/5
a • <+Hf~- cdcdcdcdcdcd
D i-( 4J O
O O -Mnj CCCCCri
34J&cd a> . ££££66
/*-N jJ'H Lnotnoi-oo
^1 P3E rHrHtO^vO
^ 0
H i— j CM tO n- LO \O
O CNJ OO O
un o ^t- vo
iM rH tO
K>
O
O
10
2 o •
^ « o ...
« S u
1 < ; . § „ ;t g I : :
< z ° ° S
x.auui-m'iii
-.2S-.«jrfJif'
?35?3?3?3 U 2
?23?Sg8gSii
62
-------
e
rt
o
- (XI
>-,
o
c
o
o c
o
.c rt •
n iz
~ 3
t O
C
C rt
o
C *J
O •"
u
3
o
I—*
0
u
o
o
u
o
z
o
0
HARDNESS
*
-1
4
on
<
H
V)
X
-J
4
»ft
si
o
sS
o
t-
u
z
o
10
o
¥
10
o
o
OI
1- a
in
0
I
O
Ajipiqatij.
mi'
X.
a
•X
?
2
UJ
I
a
e(no)
^3
£033W
i
in
O
LO
C
K^
t^i
O
0
L/l
rj
O
T— 1
O
r-j
\o
o
OJ
r-~
^t
r-t
O
00
^J
o
1
sO
oo
oc
LO
r I
K".
O
Cl
00
r^j
«3-
oc
^y
I— t
•
o
^t
to
o
o
, — (
CNJ
o
(O
Cvl
1^-
oo
Tt
LO
tv)
o
o
,
f^l
\c
^
NO
oo
r*j
C
c-.
•w
•^t
VD
t^)
O3
(XI
•
O
t^
fs)
C
M
t-~
O
<£>
0
t— 1
00
r-i
oo
hO
O
,
00
to
*3-
i—(
rsj
to
o
c-.
o
c
\D
to
r-)
Osi
1 — I
vtj
O
tsl
^a-
in
(X)
rH
^t
O
to
o
r-
00
^o
to
-*
*J-
o
r-t
1
rt
•*±
'-O
o
oc
I^-
o
e>
oo
(XI
o
LO
LO
*
O
LO
r-i
vD
in
o
C
rt
C
(8
O
O
U
o
I—
o
et
o
c
o
o
o
o
10
o
u
>
t-
z
4
o
K
a
o
u
O
K
i liJ
u z
j e
3 3
a
O
u
O
in
U
o
u
O
4
»-
O
a
o
o
s
_i
o
3
o
m
' O
m o.
X I
z o
63
-------
o
rH
O
t_>
O
u
en
2
o
0
HARDNESS
X
_i
<
K)
5I
t*y
- O
°y
u
t-
o
z
u
ro
O
¥
K>
o
u
ffi
00
CM
rH
OO
vO
LO
CM
T
r-H
r- 1
LO
CM
to
0
1^
r^
CM
*o
>£>
OO
*o
t
^t
LO
CM
o
rH
•*±
in
00
CM
r— t
OO
CM
0
en
o
CM
f— i
•*r
o>
•^T
CM
VO
T— t
r— 1
O
LO
tO
O
00
£
a
O (U rl
C f.
CTi *rH T3 CO
rH r-H S r<
C CS ID C
a e •>*-> o
•H in CJ r— 1 >H
i-H r< 66 -H +->
rH O 00 (H 4H CO
r< O t^ rH \O LO +->
JD U - « -r-H
Lfl \O ^ tO .rH
V£> flj - - *. -m
-rH tO ^f CM rH
-•H r-H CM in C 4-J
CM tn CD rt ctj
•H 01 in rH JS S
Wl > U U P.+J .H
r< O O G -P
CO rH r-H Cti Irt
1-3 U-, U^ CO OJ
^J" \O ^~ CM O
OO I — \O r-H O
rH LO
IO
o
o
o
K> O
O 10
o *t to -rt **
r-H
rH
1 1 1 1
LO \O t"-- OO
L^
cd
r-^
«^~ O "=3" O
rH Cn
1 1 I 1
rH CM tO ^d-
f-4
(C
0
o o
o
• u
-i o z i
Z 3 U 2
g 8 i i
64
-------
i- c
o c
E -n
O 4J
C "3
0 <
' C
w- r-i
O ~
^ C
LO LJ
a u.
K 3
p. O
C '^
x:
c *->
o ^H
U
3
13
Color
c
c
c*
X
o
U
HARDNESS
<
^
(J
^ o
o
h-
u
z
o
fO
o
o
o
CD
?r- a
_J
4
fO
o
o
I
o
A}ipiq.nij.
nir
I.
r
DOSAGE IN
91UI1
£Q3^W
li
o
rxi
r-.
LO
-
0
to
s
o
ri
r i
to
O
oc
CO
o
oo
r- <
£
^
o
o
I
rH
°
i-H
-
o
to
to
C~l
c
oo
c
o
CD
O
G
Os)
C
(X]
00
o
o
;
CM
LO
LO
-
0
to
to
rH
O
CD
OC
00
£
-
^
o
o
LO
rH
to
LO
LO
r-j
s
C-J
o
(NJ
S
o
cr>
s
LO
UO
^
^
LO
rH
^
LO
s
\o
00
to
rH
»C
to
O
00
oo
s
to
^
o
o
o
0
in
to
to
LO
00
o
to
5
D
C
• H
E
LO
i
X
• H
"
•o
• H
Cl
rt
~
S
Cu
o:
o
^t
(SJ
10
o
•M
3
C
• H
E
o
to
i
x
••H
"
^
0
1 — 1
c/:
4-1
CO
a
u
X
,x
r-l
•H
E
O
1— t
cu
X
t
«
c
•rH
rH
4->
4-1
O
CO
3
o
i— i
CO
1
r-j
••
rH
CO
rH
n
^
- clear and
oc
c:
• H
rH
4-1
4->
CU
CO
-o
..H
P.
a
IH
1
O
~
LO
A 4-1
•3- r1
- QC
tO -H
r-<
CO ,£:
r,
Ct
*-^
collecting
o
4-1
r-
o
•r-t
(H
p.
ni
C
•rH
rH
4-1
4J
O
CO
CO
0
4-1
3
C
• H
E
o
T— 1
surface
u
o
I-H
D
.0
"
C^J
to
•rH
•O
•H
jD
>-.
3
4U
4-1
U
z
HH
O "H
rH 4-1
P. «
6 r-H
CO d>
CO IH
•H C
OS o
•H -»H
U 4-1
O O
P. 4-1
CO -H
C-
CC-H
c u
•H f^
CO .-.
3 c.
W;
^
CO
rH-O
O -H
rH rH
0 0
U CO
t3
• 0
4-1 T3
CO
0)
4->
CO
• H
jr
4-1
C
• H
rH
a
>
o
E
O
r<
C
0
U
(^
Q
C
cu
p
CO
3
1/5
.
4-1
C
CU
i — 1
rH
0
U
X
0
r^
c:
>
o
E
o
i~
.
+
0»
O>
Ol
X
rH
rH
r:
u
• H
4-1
U
K
i-
P.
K
r-^
0
>
o
g
5
,-.
t>4
t^.
1
O>
rH
t
CO
^
<
C
o
to
73
o
4-1
U
0
rH
rH
O
u
o
i — i
p.
e
rt
u~.
, — ,
t^-
c
ra
^_;
CO
rt
* — '
*^3
0
4-1
~
4-1
• H
P-
U
o
t-H
^
cc
21
o
o
10
o
< 10
£ S
Z
o
ct
4
X
IO
o
u
_l
4
X
» «
— UJ
^ ,?
55
O
o
o
o
u
ID
3
-------
E
o
oc
4->
C
o
u
•rH
C
C
O
C
o
u
3
T3
Q
o
o
rH
o
u
Q
o
o
CJ
o
U
HARDNESS
X
_l
<
ro
••§
o
»o"
°s
r-
u
z
u
10
o
2
K)
O
U
CD
< O I
r-t- a
(O
*
_J
4
10
o
o
i
A}Tpiq.inj,
X
a
X.
9
2
UI
8
8
o
Z(HO)
BD
£033w
0
CM
t— i
0
o
rt
o
LO
I—I
CM
o
vO
o
LO
CM
o
rH
O
r-~
o
00
o
«*
rH
O
00
Ct
a>
r-
to
•^t
LO
r-j
rr
o
to
0
*d-
O
r^J
(NJ
•*
rvi
o
00
t
0
rH
OJ
a>
r-H
OO
\o
0
ai
o
CM
to
LO
(SI
^r
o
LO
r-l
^J-
O
vO
0
LO
Csl
\o
0
*o
\o
vO
<3
^
i-H
cxt
r--.
o
CTl
r-~
to
\o
o
LO
1-
0
LO
o
to
o
0)
r-l
r-i
OJ
o
«*•
r-i
vO
oo
vO
i-H
rJ
LO
o
01
v£>
«t
rH
0
LO
*t
O
LO
t— I
VD
Ul
X
nj
e
O -H
to LIH fx
" -H
rH fl} *O U
N tu 0
C CO r< r.
^ n) js CD a,
3 X 4->
I 1 1 1 1 Q
) J3 .rH 4->
) -
r-H n S~i Tl r*
L) n Crt (11 hf
D 4J 3 E
X XrH CO 0 3
)H T3 U C e -rt
O 3 O (U Ifl
> O +-> ^3 0)
rH t/1 J-* T3 C
LO O O 0) T) t»
- E O cs cfl
«*>•» i-t E
•* « OJ (/) 4-)
rH CM O O Q>
VO rH CX
lf> Ifl P. 4->
SH fn r. 6T3
03 rt CTi Cti *rH
i-J 1-5 1-5 CO O
VO ^f VO O
i— 1 LO TT rH
CM
to
O
R o
3 in "
oj MI
tls:
z i x 2
3 _, * ES
4 4 U Z
^ r- J 2
J 0 •< 4
4 r- U X
tx
p
• H
rH
4-1
(U
V>
O
4->
3
R
•H
E
i
LO
f — ^
K
O
u
rt
O
i/)
n!
V '
"0
4->
rt
i >
• H
P
'H
U
tv
rH
CL,
00
O
-3
^
-H
O
• o
Q °
O
TOTAL C
SOLUBLE
(H J-t JH
000
o o o
Cl^ CX CX
X M X ^ Xi3
»H O *H -H IM O
Q> O PH > 4H > bC
1 1 1 1 1 1
»H C^J bO »3- LO vO
^
o r^ o I-H •<* to
OO O* <7^.
1 1 1 1 1 1
rH CM K} ^f LO \O
fH
CO
^
u • o
o o
0 1- 0 - -• » * o*
? 3 ? 3 ro 2
g 8 ? S i i
66
-------
(U
M
X
I—I
o
C
O
u
E
o
c
at
rt
o
E
o
o
Coloi
Q
o
u
S
o
u
VI
LJ
Z
O
cc
<
z
*
4
O
<
r-
^
*
9
z
UJ
4
8
O
0
«
in
•1
m
•8
"I
0
0
z
«
o
z
K)
O
0
o z
K "
in
0
o
z
o
z
a
auitq
£OD»W
(i
O
LO
r-H
eg
0
s
(M
O
OO
O
LO
LO
O
LO
O
rH
O
-3-
O
00
r-H
(SJ
(VI
O
vO
CO
c:
-
O
a\
o
i-H
"
o
LO
o
LO
rH
(SI
O
00
rH
OO
o
oo
o
00
s
o
•-
"
LO
rr
o
LO
rH
K,
LO
O
CO
1 — 1
o
1— (
s
S
LO
OO
00
o
LO
^
O
o
LO
LO
rH
<*
0
en
c—
o
5
\o
LO
o
LO
o
00
o
Cl
*
^
LO
LO
o
o
fsl
LO
O
rj
Csl
«M
-
rsj
K;
oc
o
o
o
r-
LO
LO
o
-
f. K a
oc a z
si &.
4-" OS 3
05 E 03
^O O lH
i-H
L(H ^ r-
4H 0 « g
u E re
nj " cij t/j
C O J3
0 (2 rH CS
U .H < r-
OJ Of.
i-* 4-1 ^
3 4-1 xs:c
O 05 ^ <00
45 O
4-1 E O '
rH 0 O fl
C be • • )-c
X 4-> 00 O
_C MH _0 0
to
O
C_J
05
U
i/i
o!
•G
0
•M
05
,H
£^
•H
U
0
—
tc
(S] rH LO O OO
O^ CD fsl bo
1-1
I I i I 1
rH (SI K~. Tt LO
L4
05
C
o
E
o
u
4
*
. o — *
H K S *
§
8
m a.
z o
67
-------
g
. !
3 £
C c
• H tj
£
LO
CX] u
0
S C
=£ I
0 C
^" r"
1
•H £
s ••-
r-
^ 4-
O 4-
rH d
CO t/
rH
(XI
K
O
u
o
u
$
>-
r-
Z
_J
4
X
-1
4
H
3
H
U
H
i
0
n
0
j
j
3
•^
3
•< ;>•
• r-
lO'H
H rH
< 3
J
)
i
LD
h
O
o
o
> u
V
V)
en
u
z
o
e
<
z
3
t-
CAICIIIU A* c.rn. " 46 Standing - 2 hours - sample for COD
g supernatant remaining in the 2 jars was
composited to yield 1 liter - a second
1,700 dosage of MgCO^ was then added fjar 61
- flocculated, settled, and filtered
Montgomery, Alabama, 8:30 AM grab, raw
Color - 135
Mg pptd (as CaCO,)
P
o
f) o
0
(J
u
1
1
r>
c
o
u
5
£
1-
rH rH
tO LO
t— i rH
rH Cxi
rH
CO
1 ^8 g S
o 2
Ul K u • 111
5 4 5 i § ^
J ' j r- Ij 100.
O O O O O I '
10 r- W V- M * 0
68
-------
Before discussing these most interesting samples, two very important
parameters need to be brought into the evaluation of the magnesium
process. As early as October, 1971, the consistent superiority of
the magnesium process over lime alone had been visually observed in
the clarity of the jars containing magnesium. Heretofore, color and
turbidity removal had been only visually observed. Beginning with
Table 42, color and turbidity measurements are now being included to
demonstrate analytically these additional advantages previously observed.
Tables 43 and 44 show conclusively the advantages of the magnesium
coagulation process over lime alone. Although the COD difference
between the two processes is small on this wastewater, the color and
turbidity differences are vastly significant.
Tables 45 and 46 point out perhaps the most significant finding
to date, namely, the higher the COD the greater the removal of COD by
magnesium and lime over lime alone. Again, color and turbidity
removals are superior using the magnesium process against lime alone.
The desired pH range of 11.5-11.6 was not attained during these tests
using the magnesium process. Therefore, another sample was obtained
to further investigate this high COD wastewater and to evaluate COD
reduction by the magnesium ion (as in Tables 34-37).
Table 47 represents a brief test involving double flocculation
in an attempt to utilize the excess hydroxide gained in the first
coagulation. Any advantage to be realized by this procedure is not
evident in this test. The COD values for coagulated, filtered and
stabilized effluent for jars 1 and 2 of this test are even lower than
those in Table 45 and about 70 mg/Jl lower than when lime alone is used.
A third sample of the Montgomery, Alabama, wastewater was received
and showed only 2 mg/Jl natural magnesium, a COD of 1,400 mg/£, and a
color of. 160. Tables 48-49 show the results of the jar tests. Table
48 contains, in addition, final granular carbon filtration. The
reduction in COD from 1,400 to 4 mg/& and color from 160 to 0 mg/£
demonstrates the high quality effluent which may be obtained by
physical-chemical treatment of municipal wastes. The mg/& COD
removed/mg/& magnesium ion precipitated averaged 36.0 for the 12 jars.
Montgomery waste is the only one readily available whose magnesium
content is so low as to be negligible and which may be employed to
secure a family of "lime only" base line curves covering a rather wide
range of COD and BOD values in the untreated waste.
The data obtained in Table 50 must be compared with those of
Table 51 and Table 52 which follow. In Table 51, 50 ppm MgC03-3H20
is used in all jars and in Table 52, 100 ppm of MgC03'3H20 is used.
First, to compare tha data of Tables 50 and 51 only. In doing so,
the data from the jars as shown below should be compared.
69
-------
3
re
oi
re
E
re
-D
e
o
c/,
C
Gi
re
c
o
u
3
T3
0)
OS
O
O
U
•KJn
*O
T— t
O
u
•K
o
o
u
c
o
u
•KV-
0
1 — 1
0
S
0
u
V}
LU
Z
o
a:
<
X
^
"*
m
<
h-
<
~x
?
z
UI
i
o
'I
u
-
o
z
o
to
0
10
o
o
O I
*- <*
10
0
I
o
I
sunq
£033W
C o
o
r— 1
VO
LO
O
rxj
o
0
LO
c
LO
LO
0
O1
Ox)
o
-
•^
0
LO
LO
o
o
r-j
rH
Ox]
LO
O
0
en
i — i
rsj
*3-
">
^
o
in
LO
o
o
CXI
Cxi
O
00
O
ri-
cxi
-
Cxi
\D
i — 1
s
o
£
LO
LO
-
LO
o
LO
O
o
CX]
ro
rxj
o
Cxi
O
O
UI
-
LO
0
LO
O
o
-
o
-
\o
o
1X1
rxi
CO
O
c
0
CD
CTi
00
SO
^
^
o
0.
LO
O
o
Cxi
LO
K)
o
(XI
CD
$
LO
rr
^
o
o>
LO
o
o
CXI
\o
o£
C
•H
?3
C
re
4->
t/1
CX
C
• H
T— <
4J
O =«:
(/;
C
E
C 4->
o ra
4-> 3:
re
1-1 C
«•!-*
re 4->
0 «
4^
LH rH
0 -H
U-
<
-K
„.
»
in
<
J
(C
0
(O
»-
in
UI
1-
o
4
(E
4
Z
£
O
.O
tH
nj
U
T3
0
4-;
ej
>
4-J
re
^
re
— i
c c
re c
t-. •!-
re
in C
3 C
C- >-
re
— , L
— c
4-
c
o c
_C -r-
(/; *-
IT.
<
K
00
\O
K
U
0
o
4
>-
t-
z
J
4
X
_1
4
O
SU
f
o
o
o
)
U
a>
r— (
O
U I
tn
C
O
E
o
CJ
CX)
1
1
,c
3
a
a
-
c
3
c
S! S
ic r
— ci
^ c:
-- "3
SJ S
X X
p F
d 2
H C
C.i-
V
c/> C
P ;
3 c
d ••-
E
c:
J~3 fN
00
0
0 O
c
u
s
a
!
I
C£
C
H
'TJ
03
4-1
u".
t^
^<
3
r-j
c
-"
c
c
_
<
i
rt
4-1
ra
C
0)
en CX
0 3
O (/)
-1-1 1
X in
> D
re 4_)
0 3
J= C
• H
13 E
c
re LO
V
a) s-
S-i c£ c
el q r-
r— 1 -H (.
r-l
O +J ^
aj O r
-C y.
•4-
J-. -H t
rt p -
r-. rt V
0
o
u
UJ
_]
00
3
) O
tt
f-.
minu
o
rH
d
^
re
H 4-> T!
^ tn C
•i o a
J ^ 0
C--M
3 E in
3 (/> X
• H
J >- E
i-i-i 3
H "O O
H .-• I-H
3
H
O
K u
t- I
o c
K (/
C
(U
V
0
n
LO
1 — 1
S-i O
0) 4-J
4-> re
0 t-
c5 2
i-1 O
"O rH
4-i
C
3 ~^
r-t U~,
O r-
C 0
O rH
t
>
a
c
(E
J
1 ^
LO
O
N.
4J
• H
-3
1-
0
i-.
a
.
a
o
m
O
CO
70
-------
rt
H
H
_!
•a
e
rt
cd
C
O
-e
E
• H
Ifl
V
c
00
n)
c
o
u
•a
-
o
z
u
|O
o
X
10
o
u
Sox
i-l- a
u>
x
_i
<
IO
o
u
5
X^iptqani
X
a
*x
?
z
u
o
suiiq
£OD3w
1-1
^~
rH
0
to
*!•
v£>
i— (
vO
CTl
a>
CM
i-H
i-H
O
rH
tn
0
o
CM
to
o>
CM
rH
iH
0
rH
in
0
o
CM
Tt
OO
>0
i-H
O
rH
*O
iH
VO
CM
00
l/>
rH
OO
i-H
i-H
t^-
to
rH
rH
O
to
I/)
O
0
CM
Ul
f~
to
rH
rH
O
to
Ul
o
0
CM
vO
22
a, a,
05 05
ui i/> in in
4) CO 4) v +•> *J
3333
C C C C
•H -H -H -H
e e e e
to o o o
CM rH <71
I O
I I I \O
X rH
•H X OC OC
2 -H a c '
2 -H -rl
T3 rH 13 r<
•H 5 4-> (S O
P, O 4-> rt rH
rt rH (U 4-> O
O5 C/0 C/2 W U
c
a>
£
O
to
O
U
O CM OO
to to to
rH IO Ul
V)
rt
OO
vO
rH
0
U
a
o
>-
K
2
J
^
-1
O
vO
f>
O
0
c
1 O
5
en
g
U
_
H
S
Ul
.
0
I
e
«
U
o
t-
,
Z
f
g
CJ
^
U
_
g
•
1
Q
C
h-
' g
o
s
Ul
_i
r
_
s
2
i
i
i
o
0 O.
0
71
-------
,0
n)
H
n)
O
O
U
(D
O
rH
<
U1
n)
n!
E
cfl
X
1)
O
OS
•M
c
o
i
4->
a)
a>
o
rH
0
i
OS 0
»= 6
U
O
f-
OS >
o
o
a
u
s
o
o
in
kl
z
o
rH
O
O
CM
CM
rH
O
Kl
0
Kl
O
S
Ol
Kl
OO
O
vO
rH
00
OO
0
oo
o
00
00
i-H
0
CM
rH
rH
O
LO
O
*
0
Ul
LO
00
o
rH
0
CM
vO
0
rH
O
0
OO
Kl
CM
rH1
LO
rH
rH
O
O
in
0
LO
O
CM
f-
Kl
VO
oo
LO
Kl
rH
CM
vO
VO
O
03
O
O)
CM
LO
rH
VO
rH
rH
O
LO
0
VO
o
O
P.
00 O
•i-i in
•)-> C
4-1 O
0) -rl
(/> +->
a
r-t +J
rH rH
rt-H
S ^H
Ul
0}
w I-H
U O,
o e
rH C«
tt. W
in
4->
c
§
g
o
u
C n)
O *-"
U 0
H
C
'
c •
O I/I
•rl (3
+•> I-H
0
3
O •
OS i-H
O
CM CM K) rt LO f^
OO 00 OO 00 OO OO
LO
O O Kl ^- \O OO OO
r^ t — t*- t^ t^ t^
i
t*
O
rH
K) Kl r-. 00 O LO O
in LO in LO vo vo C3
•
o
^4
CO
r^
rH CM Kl Tfr LO VO
O
CM
CM
1*
O
u
o
u
s
2
ALKALII
vo
H
0
O
O
> u
tf
(0
k
Z
c
I
.
i
o
in
»
•
c
V
2
9
i
rH
C
O
0 U
e
u
S
a
3
«
1
LO
00
Ol
1
o
c
u
I
0
00
Kl
g
U
!
•
K)
rH
U
O
1-
5
c
• K
0
LO
U
H
li
_
•
:
S
g
•I
I
a
S
Ul
_
:
c
2
i
o
o &
1
0
72
-------
O
O
U
00
o
f-J
be
s:
o.
n.
•o
c
r-)
,c
o
4->
(/I
TO
TO
TO
.a
B
X
4)
O
«*-(
o
c
+->
TO
r«
o
i-H
cS
1 UC3
0
<*> E
u
o
I *O
o
*> 6
Q
o
U
9
2
o
U
HARDNESS
x
_»
<
10
s§
U
in
!s
r-
U
Z
U
m
o
X
PO
O
U
Sox
Pi: a
in
X
_l
<
rt
8
z
o
^jipTqani
x
&
V.
?
?
ft
«
o
o
*(HO)
BD
0ZH£
^OD^w
li
ra
M
M
i--
vO
K)
Ol
OO
rH
I— (
rH
OO
rH
O
**
OO
in
•3-
00
rH
•>*
(SI
O
0>
(N»
tO
i-H
^O
M
i-H
f>
r-t
I-H
f-H
t-H
O
O
r*
O
in
i-H
f-
(SJ
•si-
r-~
LO
to
o>
00
OO
0
i-H
*o
I-H
VO
rH
r-a
to
O
T*
rH
O
If)
O
CT1
»t
rsl
i-H
^d-
«*
rH
i-H
i-H
rH
rH
O
•q-
«*
o
LO
CM
0
CJ
LO
t--
K>
to
O
o>
r-J
o
r-t
00
*0
rH
rj
CM
•s-
LO
rH
O
to
O
a>
OO
c
rH
VO
vd
rH
CTl
•o
i-H
rH
t
0
CM
'
i— I
r-H
O
LO
LO
0
LO
rt
rH
C) rt
0 4J
U 0
H
C
•H
C
o •
•H I/I
+-> c
(JrH
3
13
»-5 Z
rH CM tO *t
TO 10
4-" O
•H U
P. TO rH IO i-H ^t
• H U to to «a- r-~ X
U T3
0) (/) 3
r< TO O
C
bC TO
S 4-1
TO
C
(U
S-i {X
0! rH CM to «3- 3
"-5 c/5
U X X4->
•wax
X 3 3 b£
!H O O -iH
0) rH 1-4 r.
> U U J3
1 1 1 1
(—1 OJ f^ ^~
tn
OO
i
j_,
O
i-H
0
u
O
CM
CM
H
0
u
o
u
2
ALKALIN
\D
H
O
U
O
* o
w
AROMESS
z
o
LO
•>
t
0
c
c
3
1
V
3
f-H
P
O
o o
a
u
V
|
c
1
OO
0>
1
a
O
u
_
4
£
i-
o
00
«
£
u,
!
«
«
to
i-H
U
o
t-
i
o
LO
u
8
u
s
a
i
o
c
i
a
s
u
_i
•
3
J
C
«
2
X
X
0
n a.
0
73
-------
O
O
•a
o
CM
to
O
u
00
P.
a
o
o
•a
n)
6
•H
J
£
0)
Ul
rt
n)
CO
,0
CO
O
00
4-i
O
S
4-J
(U
6
4->
rt
(U
f-.
o
rH
0
u
I •«
d)
o
o
°\= e
Q
0
u
9
a
O
HARDNESS
*
_i
<
10
«i
F)
Kt°
"8
r-
o
z
u
m
I
m
o
u
Sox
£-tt
¥
_J
<
m
8
S
AlTptqanx
X
a
X
f
Z
g
SI
o
0
Z(HO)
BD
0ZH£
SOD^N
Si
LO
CM
LO
t^
*!•
to
o
01
CO
Ol
oo
rH
O
CM
oo
to
CM
CO
rH
•*t
•st
O
en
00
CM
i-i
00
CM
1-1
01
i-f
r-l
LO
CM
•<±
O
0
1-1
I— 1
o
CM
vO
t^
CM
(O
O
CT>
O
O
rH
CM
rH
VO
I-H
OO
CM
0
LO
iH
vO
ro
o
C31
00
o
rH
\0
•«t
1-H
t-
CM
rH
LO
vO
Tt
O
O
rH
CM
LO
rH
00
1—
O
to
o
ai
r-
en
vO
vO
rH
CM
CM
•*
•*
rH
O
-*
O
CJl
vD
01
CM
f~
iH
LO
^f
rH
O
O
LO
O
o
i-H
ro
Lfl
rH
^1-
r-
LO
to
CTl
oo
o>
0
rH
OO
vO
rH
•ej-
CM
^0
to
rH
vO
to
O
01
o
CM
rH
«*
00
rH
I
vO
rH
O
LO
LO
O
o
rH
•»
•O
t
« o
4-> U
•H OS
cxu
•H
O VI
00
3 bo 60 bo
O -H -H -rt
l-H r< r-l r<
0X1 J343
UO
00
r<
O
o
u
o
CM
CM
K
O
t>
o
U
„
<(
r-
2
I
«J
X
_J
4
rt
^
N
O
U
O
> o
V,
4
M
0)
U
5
X
_
i
o
LO
»
«
C
c
3
m
=
5
Tt
rH
f
O
n c
a
0
V
a
I
f
LO
00
Ol
1
o
c
u
_J
1
o
oo
to
c
G
u
hi
a
=
c
V
t-~
to
rH
U
o
r-
_,
I
0
LO
jj
g
UJ
-
g
:
j
c
M
O
C
•
^
I
o
s
111
-1
a
c
2
.
3
i
,
o
n a.
' o
74
-------
Table 50 Table 51
Jar No. I Jar No. 1
Jar No. 2 Jar No. 2
Jar No. 4 Jar No. 3
Jar No. 6 Jar No. 4
% Reduction in
Residual COD
from Lime Treat-
COD COD ment Due to 50
Jar No. Table 50 Jar No. Table 51 ppm MgC03
1 180 1 111 38
2 180 2 108 38
4 160 3 102 36
6 135 4 116 14
The lower percentage reduction at the very high pH 11.7 is
probably due to stabilization of insoluble COD by the large excess of
lime at this pH value. This is consistent with the fact that values
for turbidity and color are not affected but steadily improve as Mg
precipitated and pH increased.
Table 52 is a repeat of Table 51 but with the dosage of
MgC03*3H20 doubled to 100 ppm. The resulting values for COD and TOG,
and corresponding percentage reductions, are only slightly better than
for 50 mg/£ MgC03'3H30, indicating that for this type of waste, the
optimum dosage of MgC03«3H30 is slightly greater than 50 ppm and much
less than 100 ppm. The percentage reduction figures perhaps show it
best.
% Reduction in
Residual COD
from Lime Treat-
COD COD ment Due to Add.
Jar No. Table 51 Jar No. Table 52 50 ppm MgC03-3H20
1 111 1 98 10
2 108 2 100 8
3 102 3 97 10
4 116 4 109 9
To further substantiate the value of the magnesium ion in
coagulating raw sewage, samples of low magnesium wastewater were
secured for jar testing. Tables 53 and 54 demonstrate the effect or
lack of effect of the added magnesium ion. The wastewater was
collected at the Catoma Sewage Treatment Plant at Montgomery, Alabama.
Table 53 was performed using a high COD wastewater and Table 54
utilized a low COD wastewater.
Table 53 employed a narrow range of precipitated magnesium.
A wider range would have been better, as noted below. The superiority
of the magnesium and lime treatment over lime alone is shown by COD
75
-------
.0
CB
E-
ctf
e
n)
J3
n)
rH
<
O
M
•4-1
d
o
o
o
o
e
O
e
a)
S
O
3
04
C3
O
U
O
rH
O
u
Q
O
U
9
O
O
in
o
CM
O
LO
CO
o
CM
OO
vO
t-l
1— 1
<-<
0
0
•«t
o
00
rH
LT1
t-H
CM
to
o
csl
O
^~
>o
r-
rH
OO
CO
CM
t— 1
I— I
O
CM
Tt
O
00
CM
O
rH
VO
CM
OO
rH
•£>
ff)
OO
\0
r- )
T3-
0)
tO
i— t
rH
0
to
*t
o
00
to
in
CO
r-H
1-1
CM
rH
\O
to
O
"9-
(—4
CM
rH
rH
•tt
1-H
rH
O
in
•*f
O
00
•ej-
LO
LO
rH
CM
rH
*!•
CM
"*
CM
rH
vD
CM
rH
LO
rH
rH
O
\D
Tt
0
CO
LO
o
1^-
i-H
vD
CM
rH
'J
to
CM
LO
rH
OO
to
i— 1
LT>
rH
0
LO
*±
<
1
vO
•M to
« Ol
•H nj
P.C-)
• H
U I/I
0 cd
r.
a,
ac
• O CM OO CO O
' LO LO LO LO
rH CM fO ^- LO \O
O
co
o
U
K
UJ
r-
^
Ul
r-
<
>
$
K
U.
O
(A
U
»-
CO
It
Ul
r-
O
^
K
<
X
U
CM
rH
CM
H!
0
U
D
U
in
<
>-
>-
z
j
4
X
J
4
O
UO
If
0
(J
e
u
in
4
v>
ARONE
X
«j
4
s
CO
to
)
0
o
u
•
u
5
a
a
9
u
CM
rH
K
0
> U
o
u
in
<
X
O
w
u
i
a
o
to
^t
rH
)
Q
0
U
J
s
r-
t-H
«*
to
O
8
UJ
J
03
3
-1
O
•)
U
o
H
I
u
r?
u
_)
O)
a
i
o
0
n
TOTAL
o
g
Ul
-i
SOLUB
z
i
K
X
Z
^
O
I 0-
i
0
76
-------
CD
n)
O
DC
0
+J
in
O
O
u
f
o
s
•(-4
1/1
o
c
eo
o
c
o
ra
rH
3
03
O
U
o
t-i
o
u
o
o
u
9
2
0
U
HARDNESS
*
_t
<
i*S
•'I
•f
0
r-
u
I
u
n
0
X1
K>
O
O
CD
4OZ
r-h- a
*
_J
<
IO
8
z
o
z
&
>N
f
z
U
ss
o
o
Z(HO)
BO
O^HS
£OD3w
i-H
rH
O
>*
to
0
0
rH
Tfr
LO
0
*J-
rs]
rH
\O
rH
rH
O
VD
IO
O
O
i-H
LO
LO
rH
O
LO
C*J
rH
OO
i-H
rH
O
VO
IO
I
vO
Cti
o x
o to
LO rH
« O 1
t-H
•0 in
in o
r< (-. rH
nj rt O
T3
(U
+•> K
rt O
4-» U
•H eB
fXU
•H
u in
0) ca
L^
Ps
bo
s
>
o oo oo -a- *a- o
to to «3- 1^- 1*^
^
nt
i-H M tO «* LO VO
(/>
4-"
(U
o
u
o>
rH
H
O
u
a
u
<2
£
z
ALKALI
^
K
O
u
c
V
in
-------
and color analyses. Comparison of Table 53 with Tables 48 and 49 show
that a much lower dosage of magnesium is about as effective as the
massive dosages of Tables 48 and 49, and far less lime is required.
This fact was shown in Tables 51 and 52 also.
Jar test data collected using low magnesium wastewaters have
been for medium and high strength.COD concentrations. A low strength
COD wastewater was desired to complete the picture on the effect of
the magnesium ion. Table 54 represents this testing program. This
type of wastewater would be encountered at most plants during the
early morning hours, holidays and weekends. The data indicate that
during these periods treatment with lime only would be sufficient.
A plant which has a substantial amount of magnesium in its raw water
would add lime to precipitate the magnesium for storage and recycling.
A plant having wastewater low in magnesium would not add magnesium
during these low COD periods. Several other samples of this type of
wastewater will be tested in the future.
The higher the raw COD the more beneficial is the magnesium
ion. A wastewater having a COD greater than 2,000 needs to be
examined to further substantiate this repetitious finding.
Precipitation of Metals
An analysis of six metals was accomplished using raw sewage
from Montgomery, Alabama. The sewage was subjected to coagulation,
sand filtration, carbon adsorption, and ion exchange by clinoptilolite.
The raw COD was 1,400 mg/£ and the final COD was 4 mg/£.
Raw Sewage Treated
Metal (mg/A) (mg/£)
Copper 0.06 <0.01
Zinc 0.33 0.02
Lead 0.07 <0.05
Total Mercury 0.000012 0.000004
Barium 0.43 <0.10
Aluminum 0.91 <0.20
The low solubility of most metallic hydroxides at high pH
values suggests the application of the magnesium treatment process
to industrial wastes high in metallic ions for recovery of valuable
raw materials as well as meeting effluent standards.
78
-------
SECTION VI. PILOT PLANT STUDIES
Pilot Plant
Figure 1 is a flow sheet of the pilot plant and Table 55 gives
the dimensions and capacity in gallons of each of the main units. Two
50 gpm pumps were mounted in parallel on a steel framework below the
surface of the influent wastewater in the comminutor discharge basins
of the waste treatment plant of the city of Gainesville, Florida.
When operating, they served as a continuous sampler, discharging
untreated waste into flow control (1) which is a baffled steel tank
fitted with an adjustable V-notch weir which could be set for any
desired flow up to 100 gpm. Excess waste was returned to the
comminutor basin through a drain pipe. This makes it possible to
operate at any desired rate up to 100 gpm.
Recycled magnesium bicarbonate liquor from storage tank (11)
was pumped by a calibrated positive displacement pump and added to
the waste as it discharged from the weir. The waste flowed by gravity
through a small rectangular baffled steel trough to rapid mixing
basin (2) and then to the bottom of the flocculator (3). Lime slurry
(12) was added in the discharge line between the rapid mix and
flocculator. Much experience with the large, heavy floes formed
by the reaction between the magnesium and the lime soon indicated
that the rapid mix was not needed, as the large heavy floes formed
a fluidized bed at the flocculator paddle speed of 1-2 rpm.
The coagulated and flocculated slurry passed to twin sludge
concentrators (4) and the clear settled effluent passed to a baffled
settling basin. Baffles provided a high pH contact basin (5), a
central carbonation basin (6) and a secondary settling basin (7) to
remove most of the CaCO^ precipitated in the carbonation basin.
Table 55
Dimensons of Main Units
Volume Volume
Unit Dimensons (cu ft) (gal)
Mix basin
Flocculation
Sludge concentration
High pH contact basin
Carbonation basin
CaC03 settling basin
30"
7' x
9 '3"
11'
3' x
10'
x
8
x
X
9
X
30"
i
9
'
9
10
6'
'3
3"
'3
x 5'
ii
8" x
11 x 8
x 8'
11 x 8
6'
'7"
7 "
i7.i
30
335
335
865
236
786
2
2
6
1
5
225
,500
,500
,588
,770
,895
79
-------
E
o.
D>
O
J
<
Q a:
Ul UJ
25
i^
O
^
1
CVl
o o m in
evi to
J g
f* _l Q
o o o
I- O m
UJ
2
UJ
X
o
CO
3
m 10
>o ro
Q
o
O
o
-------
The settled effluent passed to an effluent trough with ten
equally spaced V-notches on the receiving side and open at both ends.
A movable baffle made it possible to divide the effluent flow into
any two desired multiples of 10 percent. These two flows passed to
splitter boxes, each provided with an adjustable weir by which the
flows could be further divided when desired. In practice, the flow
of one splitter box passed to waste and the other passed through a
dual-media filter (8) of sand capped with anthrafilt and operated at
a rate of 4 gal/ft2/min. Figure 2 is a view of the pilot plant.
Magnesium Recovery
Sludge carbonation for magnesium recovery and recycling was
carried out on a "batch" basis. Settled sludge from the twin con-
centrators (4) was drawn by gravity into a 175 gal carbonation
tank (10) equipped with a rapid mix impeller. Pure C02 from a
refrigerated storage unit (9) supplied by the Chemetron Corporation
was passed through a calibrated flow meter into the rapidly mixed
slurry until the pH was reduced to pH 7.5. The carbonated sludge
was allowed to settle and the supernatant transferred to storage
tank (11) and its alkalinity as CaCO^ determined for each batch.
The procedure described above for the recovery of the
magnesium is not desirable since the dissolution of the Mg(OH)2
component of the sludge floes by C02 releases at least part of the
trapped organics and inorganics. This was not a serious problem using
pure C02 since alkalinity values of the magnesium liquor as high as
25,000-30,000 ppm as CaC03 were obtained. The volume of this very
rich magnesium liquor to be returned as magnesium make-up was relatively
small. However, in actual practice, and using kiln gas containing only
20 percent CC>2, the maximum alkalinity to be obtained is 16,000 ppm as
CaC03, thus requiring more recycled magnesium liquor.
In actual practice, the thickened sludge will pass to a
vacuum filter. The clarified filtrate will be recycled to the
influent waste to recover the high excess lime present. The cake
will pass to a multiple hearth furnace and be lightly calcined at
a temperature in the range 500°-600°C (900°-1100°F). At this
temperature the organics will be consumed supplying part of the fuel
needed and the magnesium will be converted to an "active" form of MgO
readily soluble in C02. The CaC03 will not be calcined at that
temperature.
The calcined sludge, now a mixture of CaC03, MgO, metal
phosphates, silicates and silica, will be slurried, carbcaated as
described and the carbonated liquor stored or recycled. It was not
possible to obtain a vacuum filter and small multiple hsarth furnace
for the pilot plant. However, these operations were carried out on
small batches of a few pounds using a laboratory muffle furnace for
81
-------
FIGURE 2
VIEW OF PILOT PLANT
82
-------
83
-------
-------
85
-------
the calcination process. The calcined sludge was an odor free, gray
to brown powder, with no hard lumps or pebbles and from which the
magnesium could be easily recovered by carbonation.
To determine the percentage of magnesium which may be recovered
by such a series of operations, a large laboratory sample of sludge was
prepared by coagulating five drums of raw wastewater. The resulting
sludge was collected on a vacuum filter and calcined in the laboratory
muffle furnace at 550°C. The cooled calcined material, friable and
light gray in appearance, was ground, suspended in distilled water
and carbonated to pH 7.5, at which point this present work and the
Dayton work have shown that all of the Mg(OH)2 has been converted to
the soluble bicarbonate. Analyses of the carbonated liquor showed
99% recovery. The amount of phosphate, if any, dissolved by the
operation was not checked. Should it prove to be appreciable, it
would be removed in the next pass of the recycling procedure and
would not pass into the settled effluent from the primary clarifier.
An Expanded Concept of the Magnesium Process
As previously discussed, the preferred method of magnesium
recovery is filtration of the thickened sludge and light calcination
of the filter cake or, perhaps the thickened sludge itself without
filtration. In either case, the calcined sludge would be composed
mainly of calcium carbonate, together with an active and readily
soluble form of magnesium oxide, metal phosphates (mainly Ca3(POit)2),
silica and silicates. It would be slurried, carbonated with 20% C02
from the furnace to pH 7.5, settled or filtered and the clarified
liquor containing magnesium bicarbonate recycled to the plant influent.
From this point, either of two courses could be followed and
a third, the most attractive, should be carefully studied.
Course 1, the simplest^—Convey the cake to landfill. This
would eliminate lime recovery and recycling and substantially increase
treatment costs.
Course 2, direct recalcinatiorr—This would result in the
production of lime of poor quality, containing all phosphates and
silicates. With the price of lime certain to increase substantially
due to the energy crisis, this becomes somewhat more attractive.
Course 3, not yet proved possible—This would involve two
steps. The first would be the separation, by selective flotation,
of the CaC03 from the phosphates and silicates. If this can be done,
a much higher quality lime could be produced by recalcining and
recycling the high dosages needed. The rejects, namely phosphates
and silicates with some CaCC^, would be disposed of as landfill, as
in Case 1, above.
86
-------
The second also deserves careful study since it would completely
eliminate all solid waste discharges from a physical-chemical waste
treatment plant. It would involve:
Removing NHs from the coagulated and filtered effluent with
the naturally occurring ion-exchange mineral clinoptilolite.
When the column is exhausted, acid stripping with I^SOi^ to
recover NH3 as (NH^^SOif. This acid solution containing
excess l^SO^, would then be used to acidulate the phosphate-
silicate slurry of Course 3.
Assuming a "normal" municipal waste containing 15 ppm nitrogen,
equivalent to 18 ppm NH^, and 9 ppm phosphorus, equivalent to
27 ppm PO^, calculations indicate that the above process would
theoretically yield for each million gallons of waste treated,
about 1,200 Ibs dry weight of a 10-12-0 fertilizer worth, at
Florida current prices, about $50 per ton. Cost of the HgSO^
needed at $30 per ton would be about $10.
Figure 3 is a flow sheet embodying the entire process,
including reduction of BOD and COD to effluent standards
using granular carbon filters.
The influent and effluent data shown on Figure 2 are derived
from a typical bench scale run.
Study of Carbonation and Recovery of
Since magnesium carbonate trihydrate is not now and never has
been commercially available in this country, there is little infor-
mation available concerning its physical and chemical properties .
Information in the scientific literature is old and often conflicting
in important details .
Accordingly, several pounds of very pure material was prepared
from USP grade Epsom Salts, MgSOi^HjO and caustic soda. The Epsom
Salts was dissolved in distilled water and Mg(OH)2 precipitated by
adding the NaOH slowly with continuous stirring. The voluminous,
snow-white precipitate was allowed to settle, the clear supernatant
drawn off and the Mg(OH)2 washed by decantation with distilled water
until free from NaOH. It was then suspended in distilled water, cooled
to 15°C and carbonated with pure C02- It was cooled only to 15°C since
below that temperature the pentahydrate, MgCOs'SI^O, is supposed to be
formed. Since the carbonation reactinon is exothermic, the plastic
carbonation drum is suspended in a slurry of cracked ice in a larger
drum to maintain the temperature below 20°C.
When all of the Mg(OH)2 was dissolved, the clear Mg(HC03>2
liquor was warmed to 40°C by suspending the plastic drum in a larger
drum of hot water and aerated with compressed air using a porous
diffuser plug until the alkalinity of the hot solution had dropped to
about 2,200 ppm, this being the solubility of MgC03-3H20 at the
temperature employed.
87
-------
— c>
o —
Q- Z
_l _l K
< < O
PI- _l Q O O
O O CO O O O O
I- I- to O 00 O I-
UJ
2
UJ
X
o
CO
ro
CO
to
ui
o
o
CK
O.
in
CM CO 10
in ff> —
a. z
-> _i ^
?£ SQOO
OOOTOOOO X
l-t-OTOOJOl- CL
88
-------
The hot slurry was filtered on a large Buchner funnel, using
No. 2 Whatman filter paper, washed with cold distilled water and
pressurized as much as possible on the Buchner funnel. The filter
cake was removed, chopped fine with a large spatula, spread out in a
thin layer on cotton towels and allowed to air-dry for 24 hours in
the air-conditioned laboratory. A 1 g/£ solution in C02~free distilled
water will have an alkalinity as CaC03 of about 695 ppm and contains
about 96% MgC03-3H20 and 4% free moisture.
It is almost identical in composition with much older material,
kept in closed containers, indicating that the material may be safely
stored under water plant handling conditions.
Table 56 shows the results obtained by carbonating a batch of
Mg(OH)2 prepared as described above.
Table 56
Product Recovery by Carbonation of Mg(OH)2
Time
(min. )
0*
5a
10a
15
20
25
30
40
60
80
90
100
Alkalinity
P T
0.5
2.9
12.0
2.6
3.3
3.8
4.5
4.1
0
0
0
0
0.5
6.9
32.5
8.5
8.7
10.9
15.5
17.0
19.9
24.3
25.6
26.2
MgC03
(mg/A)
40
580
2,400
5,200
6,600
7,600
9,000
8,200
Qb
0
0
0
Mg(HC03)2
(ing/ A)
10
110
850
3,300
2,100
3,300
6,500
8,800
19,900
24,300
25,600
26,200
Total
(mg/A)
50
690
3,250
8,500
8,700
10,900
15,500
17,000
19,900
24,300
25,600
26,200
a0.02 N acid. All others 0.2 N acid.
Increased C02.
Table 57 shows the rate of recovery of the MgC03'3H20 by
aerating the heated solution.
89
-------
Table 57
Recovery of Product MgC03*3H20
Time
(min.)
0
5
15
30
45
60
75
90
Alkalinity as
P T
(mi)
0
1.8
1.5
0.4
0.4
0.4
0.4
0.4
26.2
25.6
13.5
8.3
5.5
4.3
3.6
2.9
CaC03*
M
26.2
24.8
12.0
7.9
5.1
3.9
3.2
2.5
Mg(HC03)2
26,200
22,000
10,500
7,500
4.700
3,500
2,800
2,100
MgC03
0
3,600
3,000
800
800
800
800
820
Note: Initial temperature 42.5°C, final 36°C. Continued aeration
would have converted all remaining Mg(HC03)2 to MgC03'3H20 but with
little additional recovery.
*0.2 N acid.
Table 58 shows the data obtained by determining the alkalinity
of product prepared above with other samples as described.
Recovery of Magnesium Bicarbonate Liquor from Gainesville Tap Water
In order to have the storage tank filled with magnesium
bicarbonate liquor to be used as makeup in the first few cycles of
waste treatment, it is necessary to employ some convenient source
of magnesium. Gainesville unsoftened and softened municipal water
represent possible sources. Since essentially selective softening
is used, removing mainly calcium hardness, the softened water
contains about 40 ppm of magnesium as calcium carbonate. There is
no convenient source of raw water available to our pilot plant so
that the softened water was used. The pilot plant was placed in
operation, removing the magnesium as Mg(OH)2 by excess lime treat-
ment. The resulting sludge was carbonated and the carbonated liquor
stored for later use. This "shake-down" during several days of
operation provided an opportunity to check the calibration of all
feed pumps and locate and correct any operational difficulties.
90
-------
Table 58
Alkalinity as CaC03 of Selected Samples of MgC03*3H20
Sample
No.
1
2
3
4
5
Alkalinity as
17.4
17.5
17.25
17.4
17.5
CaC03
Alk.
696
700
690
698
700
% MgC03-3H20
96.0
96.5
95.0
96.3
96.5
% H20
4.0
3.5
5.0
3.7
3.5
Note: Solutions contain exactly 1.0000 g MgC03*3H20 per liter.
Description of Samples;
Sample 1: Prepared by A. P. Black from Dayton sludge in August, 1969.
Air-dried. Bottle 75% filled.
Sample 2: Same as above and ground. Bottle 40% full.
Sample 3: Another batch prepared by A. P. Black from Dayton sludge
in late 1969. This sample "cottony." Bottle 10% filled.
Sample 4: Material prepared by Black and Thompson from Dayton
sludge early in 1971. Bottle 50% filled.
Sample 5: Sample of material prepared by A. P. Black and Arley
DuBose in August, 1972. Air-dried. Bottle 95% filled.
On November 7, 1972, the pilot plant was placed in operation
using municipal tap water. This selectively softened water contains
about 40 mg/£ of magnesium as calcium carbonate. This was precipi-
tated by adding hydrated lime slurry, just as it will be when waste-
water is being treated. The resulting floe was very thin and light
as would be expected, since little or no CaC03 is formed using
softened water. The large flocculator paddles and the speed of
rotation destabilized the floe particles, preventing agglomeration.
The four 9-foot paddles were removed and replaced with 2-foot
paddles. The minimal speed of rotation (4 RPM) was known to be
excessive, but provisions for further reducing this speed were not
readily available. In order to build up the floe more readily,
Epsom Salts (magnesium sulfate) was fed into the tap water prior
to the addition of lime. The shorter paddles and additional
magnesium sulfate improved the floe size and settling character-
istics.
91
-------
The first batch of settled sludge was drawn into the carbonation
tank on November 9. This sludge (66 gallons) was carbonated and the
results appear in Table 59. The carbon dioxide used was determined by
weight difference in the cylinder. A set of scales has been loaned to
the project from the water plant for this operation.
Table 59
Carbonation of Sludge
pH
12.2
10.4
9.9
9.5
8.6
7.9
Time
(min . )
0
10
20
30
40
50
C02
(Ibs)
0
1
2
4
6
8
Alkalinity
OH CO 3
(ng/i)
338 76
696
1,520
1,440
480
0
HC03
—
304
720
2,000
4,880
5,280
The supernatant liquor containing the Mg (1^63)2 was removed
from the carbonation tank into a 55-gallon drum. Thirty pounds of
MgSO^ was then added to this solution prior to feeding it back to
the tap water.
A second batch of sludge was drawn into the carbonation tank
and allowed to settle overnight. Additional settling did take place
and the supernatant was discarded. This same process was repeated
twice in order to build a more concentrated sludge slurry. The
resulting thickened sludge amounted to 175 gallons and the carbo-
nation run is shown in Table 60.
An additional source of magnesium was provided by Dixie Lime
and Stone Company of Ocala, Florida, in the form of dolomitic ground
limestone. The magnesium concentration was 30% by analysis. The
limestone was mixed with the hydrated lime and fed to the tap water.
The hydrated lime utilized during this period was the dolomitic
variety which contributed an additional increment of magnesium.
Carbonation of the resulting sludge (Table 61) produced a stronger
magnesium stock solution.
After running the pilot plant 6 hours per day for three weeks,
a thick sludge had developed that was difficult to stir using the
92
-------
Table 60
Carbonation of Thickened Sludge
PH
12.2
10.4
9.8
9.7
9.0
8.6
7.6
Time
(min.)
0
20
30
40
50
60
80
CO 3
(Ibs)
0
2
4
6
10
12
Alkalinity
OH CO 3
(mg/A)
456 56
288
800
1,600
1,200
1,000
0
HC03
0
72
200
400
3,400
5,000
7,600
Table 61
Carbonation of Sludge
November 27, 1972
PH
9.2
8.3
7.8
Time C03 Alkalinity
(min.) (Ibs) C03 HC03
30 5 2,000 1,800
60 10 1,600 6,000
90 15 — 10,400
available equipment. Since dilution of the sludge with tap water would
elute an even greater quantity of. magnesium and provide for more effi-
cient mixing, a 2 to 1 mixture of sludge and tap water was carbonated
(Table 62).
Operation of Pilot Plant Using Raw Sewage
On December 6, 1972, raw sewage was started through the pilot
plant at 25 gpm. The plant was run 6 hours per day for one week.
93
-------
Table 62
Carbonation of Sludge
November 28, 1972
PH
9.3
8.6
7.8
7.5
Time
(min . )
30
60
90
120
CO 3
(Ibs)
5
10
15
20
Alkalinity
CO 3
2,400
4,000
—
—
HC03
(mg/4)
1,600
4,400
8,600
13,500
The floe was thick and settled well. The pH was easily maintained
between 11.4 and 11.6. The only difficulty encountered was the
breaking up of the floe by the overflow weir in the flocculation tank.
Some settling of floe particles took place in the flocculation tank.
The overflow weir will no longer be utilized. Two 8 in. pipes were
installed below the overflow trough to carry the solids directly to
the settling basins. In addition, two side-mounted mixing plates
have been removed from the flocculator tank. Carbonation of the
sewage sludge is shown in Table 63. No odor was present in this
sludge.
Table 63
Carbonation of Sewage Sludge
(66 gal. tap water, 40 gal. sludge)
pH
12.0
9.6
8.4
7.8
7.3
Time
(min.)
0
30
60
90
120
CO 3
(Ibs)
0
3
5
8
10
Alkalinity
OH CO 3
(mg/i)
520 80
1,200
3,200
0
0
HC03
0
1,800
4,000
10,700
12,000
-------
Pilot Plant Additions and Improvements
1. In order to evaluate the benefit of return sludge, a 2 in.
pipe was installed to transport the settled sludge back to
the rapid mix tank. Operation of the plant using returned
sludge did not visibly improve the clarity of the super-
natant. In addition, the use of the rapid mix was shown
to produce a smaller floe due to excessive mixing speed.
Thus, the rapid mix is no longer utilized.
2. A previously installed baffle' in the settling basin was
positioned at an angle to the inlet pipes (2"—8") from
the flocculator producing excessive velocities in the
settling basin. This baffle (4* x 8') was moved two feet
back into the settling basin and installed on a true
vertical to within 5 feet of the bottom. Operation of
the plant under this condition produced a marked improve-
ment in the settling of the floe and clarity of the
supernatant.
3. Two flowmeters were installed for more accurate measurement
of the C02 gas being utilized. The high rate flowmeter
(20 Ibs/hr) is used in the recarbonation of the settled
sludge, while the smaller flowmeter (7.5 Ibs/hr) is used
on the recarbonation basin for pH adjustment.
4. In order to operate the pilot plant continuously for more
than 24 hours an additional magnesium bicarbonate tank was
constructed and installed.
5. A wash water storage tank for backwashing the dual media
filter was constructed and installed. This tank is for
holding high pH (11.5) wash water. Provision has also
been made for washing the filter with tap water and the
normal operating procedure uses only tap water for back-
washing the filter.
6. A second 70 gpm submersible pump has been installed in the
comminutor. Thus, a spare pump is ready in case of pump
malfunction during extended runs.
7. During construction of the pilot plant no provision was made
for draining or wasting sludge from the settling basin.
The rapid buildup of sludge in the settling basin required
this action and a drain line was installed.
Alternate Methods for Operating the Reactor-Clarifler
The 2,500-gallon cylindrical tank, which is equipped with a
vertically mounted paddle flocculator, has been operated as a
95
-------
combined mixing and flocculating reactor. It has been found that
liquid turbulence created by the paddles at only 2 rpm is sufficient
for complete reaction and the subsequent growth of large floe
particles. Doubling the paddle speed to 4 rpm was found to
destabilize the floes and the unit was rebuilt to provide the
slower paddle rate.
The fluidized sludge bed is allowed to rise to the level of
two 8 in. lateral transfer pipes which permits the sludge to flow by
gravity and with minimum turbulence to the sludge settling basins.
It has been impossible to prevent some breaking up of the large floe
particles and the resulting fines pass from the sludge settling
basins to the larger combined settling and carbonation basin. It
has been found that the use of activated silica as a flocculant aid
provides a stronger floe with less fines.
An alternate and probable preferable way to operate the unit
would be to hold the level of the fluidized sludge bed below the
level of the two 8 in. effluent pipes and allow the clear almost
turbidity-free clarified liquor to pass directly to the settling-
carbonation basin. This is usual practice in up-flow solids-contact
reactors of water plants. However, in waste treatment plants the
surface of the clarified liquor is covered with a film of grease
and entrapped solid matter such that an effluent weir cannot be used.
It would be necessary to maintain the levels in the tank such that
the two 8 in. effluent pipes are above the sludge level but below
the dirty surface and this cannot be done in practice on a continuous
basis.
For this work, a flocculant will not be used to toughen floe
particles. Actually, the passage of a small amount of destabilized
floe would not normally be of concern, but in a wastewater treatment
plant it should be held to a minimum since it contains a certain
amount of the finely divided amorphous calcium phosphate which is
redissolved in the carbonation step which follows.
Extended Pilot Plant Operation on Weak Wastewaters
During the period April 11, 1973 to April 13, 1973, a 48-hour
run was conducted. The results are presented in Table 64. No carbo-
nation was carried out prior to filtration by the dual media filter.
Average chemical dosages were:
MgC03'3H20 (present in sewage and added) 90 mg/£
Ca(OH)2 (90%) 435 mg/£
A composite of the two-hour samples was analyzed for BOD. The
influent was 132 mg/£ and the effluent 16 mg/£.
96
-------
Table 64
48-Hour Pilot Plant Run
Date
4-11-73
4-12-73
4-13-73
Time
12 M.
2 P.M.
4 P.M.
6 P.M.
8 P.M.
10 P.M.
12 P.M.
2 A.M.
4 A.M.
6 A.M.
8 A.M.
10 A.M.
12 M.
6 P.M.
12 P.M.
6 A.M.
12 M.
Influ
COD TOG
336 96
302
398 113
291
312 85
325
224 59
112
78 54
67
220 60
329
336 95
370
213
67
434
ent
Total P
- mj
9.5
12.6
23.4
10.2
13.4
12.0
12.4
9.4
8.4
7.9
9.0
6.4
11.2
COD
y / n
11 x,
58
61
62
64
70
65
55
44
36
36
28
24
37
62
50
36
34
Effluent
TOC Total P
34 0.42
0.33
34 0.22
0.56
37 0.83
0.29
28 0.34
0.21
19 0.25
0.18
17 0.58
0.24
22 0.34
Note: Four-hour lag time (4 P.M. influent is 8 P.M. effluent).
91
-------
The values from each of the two-hour samples were averaged and
the results shown below:
Influent Effluent
pH 7.1 11.5
Alkalinity 116 mg/X, 215 mg/£
COD 260 mg/£ (total) 48 mg/£ (filtered)
Total P 11.2 mg/£ 0.37 mg/£
TOC 81 mg/£ 27 mg/£
An influent flow of 50 gpm was maintained for the entire
period. Approximately 4 gal./ft2/min passed through the dual media
filter.
During the period April 17, 1973, to April 20, 1973, a 72-hour
run was conducted. Chemical dosages were the same as the prior week.
However, carbonation was employed prior to dual media filtration.
The phosphate content in the effluent increased due to the redissolving
of fine floe by C02.
The filter run the first week was 40 hours, while the second
week a 60-hour run was observed. Table 65 presents the analytical
data collected.
The values from each of the three-hour samples were averaged
and the results shown below:
Influent Effluent
pH 7.1 10.0
Alkalinity 123 mg/£ 145 mg/£
COD 347 mg/X, (total) 56 mg/Jl (filtered)
Total P 6.7 mg/£ 0.48 mg/£
TOC 86 mg/£ 32 mg/X,
A composite of the three-hour samples was analyzed for BOD.
The influent was 180 mg/£ and the effluent 30 mg/£.
-------
Table 65
72-Hour Pilot Plant Run
Date
4-17-73
4-18-73
4-19-73
4-20-73
Time
3 P.M.
6 P.M.
9 P.M.
12 P.M.
3 A.M.
6 A.M.
9 A.M.
12 M.
3 P.M.
6 P.M.
9 P.M.
12 P.M.
3 A.M.
6 A.M.
9 A.M.
12 M.
3 P.M.
6 P.M.
9 P.M.
12 P.M.
3 A.M.
6 A.M.
9 A.M.
COD
472
493
448
469
134
78
540
505
450
495
405
225
89
56
378
473
562
402
403
302
134
112
549
Influ
TOG
123
126
90
60
41
25
132
96
113
131
101
64
37
18
95
158
124
106
101
74
40
30
85
ent
Total
7.6
7.9
7.8
5.7
2.8
2.1
8.4
8.8
9.1
8.2
7.7
5.8
2.8
1.9
8.9
9.4
8.4
9.2
7.8
5.7
3.1
2.8
7.7
P COD
61
62
61
63
50
49
54
63
65
81
72
56
53
40
43
27
56
62
65
62
63
45
36
Eff
TOC
42
39
39
39
37
25
23
17
42
36
36
34
33
20
23
22
37
38
39
35
37
23
21
luent
Total P
0.68
0.45
0.43
0.47
0.27
0.40
0.66
0.36
0.85
0.76
0.54
0.36
0.70
0.38
0.20
0.18
0.58
0.62
0.81
0.57
0.40
0.33
0.25
99
-------
Comparison of Pilot Plant Run and Table 27
Table 27 represents a close approximation to the raw sewage
encountered during the period of April 17, 1973 to April 20, 1973.
Comparison of the average COD, TOC, and BOD values for the pilot
plant run to this table shows quite favorable agreement. The raw
BOD of the table was 136 mg/£ and the pilot plant was 180 mg/£. The
filtered BOD of Table 27 was 19 mg/£ and the pilot plant was 30 mg/H.
The TOC of the pilot plant (raw) was 86 mg/Ji and Table 27 was 132
mg/£. The filtered effluents were the same. The COD of the raw
sewage for the pilot plant was 347 mg/£ and Table 27 was 387 mg/&.
The filtered effluents were the same.
The only significant difference between Table 27 and the
pilot plant was that carbonation was performed on the pilot plant
run and not in Table 27.
Comparison of Pilot Plant Runs and Table 28
Table 28 compares other physical-chemical processes to the
earliest jar tests of this research. Comparison of this table to
the pilot plant runs (average values) shows some improvement using
the magnesium carbonate process.
For example, comparison of the work of Hannah (Table 28) to
the pilot plant run of April 8, 1973 shows:
COD
265
66
BOD
— uig/ x,—
139
28
TOC
78
23
Hannah Before
After
Gainesville
Pilot Plant Before 260 132 81
4-8-73 Run After 48 16 27
Comparing the work of Bishop (Table 28) to the pilot plant run of
April 15, 1973, shows:
COD BOD TOC
-mg/Jl-
Bishop Before 347 142 118
After 66 31 26
Gainesville
Pilot Plant Before 347 180 86
4-15-73 Run After 56 30 32
100
-------
The detailed analyses of the two-hour samples of the April 11,
1973 to April 17, 1973, pilot plant run are shown on Table 64. Table 65
displays the analyses of the three-hour samples collected from April 17,
1973 to April 20, 1973. Comparison of these hourly samples with
Table 28 may be made by including a four-hour time lag between the
influent and effluent for the week of April 11 and a three-hour time
lag for the week of April 17.
Extended Pilot Plant Operation on Medium Strength Wastewaters
Table 66 displays a summary of five weekly runs. Each run was
conducted under increasing magnesium precipitated conditions. The
data shown are averages from Tables 67, 68, 69, 70, and 71. Figure 4
shows typical COD curves.
The municipal wastewater of Gainesville averages 40
magnesium (as CaC03). Therefore, a baseline for lime only treatment
had to be determined at the highest pH which would not precipitate
any magnesium ion (pH 11.1). The 5 mg/Jl magnesium precipitated run
required only a slight addition of lime to bring down half of the
magnesium occurring naturally in the sewage. The 10 mg/£ magnesium
precipitated run required adding makeup magnesium since approximately
12 mg/Jl magnesium (as CaC03) cannot be precipitated from the sewage
but remains complexed.
The data show increased improvement in the effluent character-
istics even though the raw sewage increased in strength over the test
period. Shock loads from 200 mg/£ to 1,000 mg/£ COD were encountered
each week. The process responded by showing a lower residual in the
effluent COD each week as the magnesium was increased. Phosphate
removal increased from 1.3 mg/£ residual for lime only to 0.1 mg/Ji
for 30 mg/£ magnesium ion precipitated.
The length of filter runs showed an increase from 30 hours for
lime only treatment to 72 hours for 30 mg/£ magnesium ion precipitated.
Bacteriological samples were collected during the weeks of
June 5, 1973, and June 26, 1973. The membrane filter technique was
used to determine total coliforms per 100 mis. Grab samples were
collected from the raw sewage entering the pilot plant at the flow
control box. The effluent from the dual media filter served as the
other collection point. Counts for the raw sewage averaged 4 million
per 100 mis. The filter effluent (pH 9.5) showed 0 to 20 coliforms
per 100 mis.
101
-------
u.
ZQ
-o
u> m — ro
in to o cvi
a:
o
o
LU ui
3 K
o
u. a
o
o
o>
E
o O
0 —
CM to
-Q'0'O iN3nidd3
102
-------
vO
vO
01
rH
tfl
H
T)
01
to
4-1
•H
a.
•H
U
0)
PH
•H
CO
GJ
a
00
CO
JTJ
'O
(U
(0
to
(11
o
c
M
m
0
a
0)
14_|
M-l
W
33
ft
4J
3
0
P-i
rH
to
O
H C
M
4J
j3
0
rj
0
H
C
H
4-1
3
O
O
O
PQ
C
M
4J
3
O
Q
0
U
C
M
OJ a)
Pj *^3
•H T)
rJ
O CSl
in in
o m
in ^**
CO CO
O m
CO
co r^
r^"* 1
I m
ON rH
1 1
in in
m
*
rH
rH
O
in
•
o
^j-
•
ON
m
CM
r^
rH
rH
CM
m
vO
^3-
CM
rH
ON
CM
in
in
OO
rH
•xf
O
i-H
CO
px
1
in
i
vO
m
rH
rH
rH
CM
•
O
O
•
oo
OO
CM
CM
00
m
oo
CM
CO
r-
^j.
m
m
o
rH
m
o
CM
CO
f^
I
CM
1
vO
m
•
rH
rH
CO
rH
•
0
CM
•
CO
r^
CM
m
*^-
rH
CM
CO
o
vO
CO
CM
vO
CO
ON
m
o
m
o
CO
CO
r^
1
CO
rH
I
00
103
-------
Table 67
Lime Only Treatment
Date
5-9-73
5-10-73
5-11-73
5-12-73
Time
12 M.
4 P.M.
8 P.M.
12 P.M.
12 M.
4 P.M.
8 P.M.
12 P.M.
4 A.M.
8 A.M.
12 M.
4 P.M.
8 P.M.
12 P.M.
4 A.M.
8 A.M.
12 M.
COD
603
423
342
117
738
540
459
333
153
225
1,116
612
1,656
396
126
252
423
Influe
TOC
181
127
79
31
150
141
158
78
50
48
275-
164
533
50
40
60
110
;nt
Total
8.9
9.2
7.0
3.3
11.6
12.2
8.6
7.1
3.1
5.3
12.0
9.5
8.7
7.1
3.6
10.2
12.0
P COD
mg/Jt— —
184
187
166
112
76
112
151
133
83
61
97
72
162
137
122
72
47
Eff
TOC
59
51
46
32
24
26
43
31
34
29
29
46
52
45
42
26
20
luent
Total P
2.40
2.18
1.84
1.00
0.48
0.93
1.35
1.82
1.42
1.26
1.12
1.42
1.68
1.15
1.11
0.62
0.42
Note: Four-hour time lag (4 P.M. influent is 8 P.M. effluent).
10U
-------
Table 68
5 mg/X, Magnesium Precipitated
Date
5-15-73
5-16-73
5-17-73
5-18-73
Time
4 P.M.
8 P.M.
12 P.M.
4 A.M.
8 A.M.
12 M.
4 P.M.
8 P.M.
12 P.M.
4 A.M.
8 A.M.
12 M.
4 P.M.
8 P.M.
12 P.M.
4 A.M.
8 A.M.
12 M.
COD
513
707
571
174
223
803
542
522
552
174
281
736
620
842
474
300
846
649
Influ
TOG
110
118
94
30
47
158
174
117
90
28
52
124
163
109
87
25
125
112
ent
Total
9.6
8.4
6.6
3.3
4.4
12.0
12.0
9.1
7.6
3.1
4.5
11.8
9.5
9.1
8.0
3.2
10.3
12.0
P COD
139
147
132
112
70
74
128
147
143
116
70
115
155
163
139
108
12
77
Eff
TOG
34
42
36
33
20
21
37
52
45
32
21
36
50
50
35
26
6
19
luent
Total P
0.82
0.64
0.64
0.54
0.43
0.51
0.83
1.07
0.69
0.61
0.48
0.91
1.20
0.91
0.67
0.52
0.05
0.52
Note: Four-hour time lag (4 P.M. influent is 8 P.M. effluent).
105
-------
Table 69
10 mg/Jl Magnesium Precipitated
Influent
Date
6-5-73
6-6-73
6-7-73
6-8-73
Time
12 M.
4 P.M.
8 P.M.
8 A.M.
12 M.
4 P.M.
8 P.M.
8 A.M.
12 M.
4 P.M.
8 P.M.
8 A.M.
12 M.
COD
553
669
535
295
821
803
749
312
491
455
500
437
517
TOG
118
142
119
59
130
160
127
109
113
98
93
113
146
Total P
10.7
9.3
9.2
7.5
11.0
9.8
8.6
4.4
11.2
9.4
9.1
10.0
12.0
COD
146
140
120
89
71
96
136
53
46
86
103
57
68
Effluent
TOG
29
40
38
18
18
25
39
16
20
27
14
20
27
Total P
0.46
0.71
0.57
0.46
0.54
0.53
0.51
0.44
0.28
0.55
0.57
0.41
0.50
Note: Four-hour time lag (4 P.M. influent is 8 P.M. effluent).
106
-------
Table 70
20 mg/& Magnesium Precipitated
Influent
Date
6-26-73
6-27-73
6-28-73
Time
4 P.M.
8 P.M.
12 P.M.
8 A.M.
12 M.
4 P.M.
8 P.M.
12 P.M.
8 A.M.
12 M.
4 P.M.
12 P.M.
COD
434
399
—
1,215
434
417
651
—
495
469
477
—
TOC
89
79
—
83
96
97
109
—
55
76
52
—
Total
8.0
6.7
—
4.4
10.6
8.7
6.8
—
9.5
7.9
9.3
—
P COD
— mg/x- -
—
77
74
70
94
87
72
66
50
80
56
—
Effluent
TOC
—
29
—
—
21
33
34
33
—
23
25
26
Total P
—
0.23
0.21
—
0.15
0.17
0.26
0.22
—
0.18
0.24
0.23
Note: Four-hour time lag (4 P.M. influent is 8 P.M. effluent).
107
-------
Table 71
30 mg/£ Magnesium Precipitated
Influent
Date
8-13-73
8-14-73
8-15-73
8-16-73
Time
4 P.M.
8 P.M.
12 P.M.
12 M.
12 M.
4 P.M.
8 P.M.
8 A.M.
12 M.
4 P.M.
8 P.M.
COD
482
580
500
700
1,356
819
382
527
482
498
591
TOG
173
107
95
183
412
116
87
100
108
118
96
Total P
8.8
6.8
6.3
10.5
24.9
5.7
4.2
3.9
5.4
6.0
—
COD
40
55
46
68
72
97
86
68
56
42
58
Effluent
TOC
—
—
20
21
30
31
28
31
26
24
32
Total P
—
.10
.07
.10
.14
.18
.16
.15
.13
.12
.18
Note: Four-hour time lag (4 P.M. influent is 8 P.M. effluent).
108
-------
SECTION VII. REFERENCES
1. Stephan, D. G. and Weinberger, L. H., "Water Reuse: Has It
Arrived?" Jour. Water Poll. Control Fed., 40, 529 (1968).
2. "Advanced Waste Treatment Research," FWPCA Publ. No. WP-20-AWTR-
19, R. A. Taft Water Res. Ctr., U.S. Dept. of the Interior,
Cincinnati, Ohio, 96 pp (1968).
3. Weber, Walter J., Jr., Hopkins, C. B. and Bloom R., Jr.,
"Physiochemical Treatment of Wastewater," Jour. Water Poll.
Control Fed., _42, 83 (1970).
4. Pearse, L. et al. , "Chemical Treatment of Sewage," Sew. Wks. J.,
I, 997 (1935).
5. Rudolfs, H., Setter, L. R., and Baumgartner, W. H., "Effect of
Iron Compounds on Sedimentation, Digestion and Ripe Sludge
Conditioning," Sew. Wks. J., _!, 398 (1929).
6. Waring, F. H., "Report of Investigation of the Calcar Process
of Treatment of Municipal Sewage at Circleville," Sew. Wks. J.,
_5, 199 (1933).
7. Eldridge, E. H. and Damoose, N. G., "A Study of Ferric Chloride
Treatment of Sewage at Grand Rapids, Mich.," Sew. Wks. J., _5_,
739 (1933).
8. Scott, L. H., "Treatment of Sewage at Oklahoma City with Iron,
Chlorine, and Lime," Sew. Wks. J., 7_, 506 (1935).
9. Hendon, H. H., "Experiences with Chemical Treatment of Sewage
at Birmingham, Alabama," Sew. Wks. J., J5, 231 (1936).
10. Rudolfs, W. and Gehm, H. W., "Chemical Sewage Coagulation," Sew.
Wks. J., jj, 195 (1936).
11. Rudolfs, W. and Gehm, H. W., "Chemical Coagulation of Sewage,"
Sew. Wks. J., JJ, 422 (1936).
12. Rudolfs, W. and Gehm, H. W., "Chemical Coagulation of Sewage,"
Sew. Wks. J., _8, 537 (1936).
13. Rudolfs, W. and Gehm, H. W., "Chemical Coagulation of Sewage,"
Sew. Wks. J., _8, 547 (1936).
109
-------
14. Rudolfs, W. and Gehm, H. W., "Chemical Coagulation of Sewage,"
Sew. Wks. J., IQ, 450 (1938).
15. Gehm, H. W., "Chemical Coagulation of Sewage," Sew. Wks. J., 10,
938 (1938).
16. Gehm, H. W., "Chemical Coagulation of Sewage," Sew. Wks. J., 11,
739 (1939).
17. Gehm, H. W., "Chemical Coagulation of Sewage," Sew. Wks. J., 13,
239 (1941).
18. Miller, E. C., "Chemical-Mechanical Treatment of Sewage and Sewage
Sludge at Dearborn," Sew. Wks. J., J>, 447 (1933).
19. Donaldson, W. et at., "Some Notes on the Operation of Sewage
Treatment Works," Sew. Wks. J., j4, 48 (1932).
20. Rudolfs, W., Discussion of Paper by Gleason and Loonam—"The
Development of a Chemical Process for Treatment of Sewage,"
Sew. Wks. J., J5, 267 (1933).
21. Jaffe, T. and O'Sullivan, J. H., "High-Rate Chemical and
Biological Treatment of Sewage," Sew. and Ind. Wastes, 24,
149 (1952).
22. Stevenson, R. A., "Chemical Sewage Purification at Palo Alto,"
Sew. Wks. J., J5, 53 (1933).
23. Banks, H. 0., "The Palo Alto Sewage Treatment Plant," Sew. Wks.
J., J3, 68 (1936).
24. Slechta, A. F. and Gulp, G. L., "Water Reclamation Studies at the
South Tahoe, PUD," J. WPCF, JJ9, 787 (1967).
25. Rand, M. C. and Neverow, N. L., "Removal of Algal Nutrients from
Domestic Wastewater," New York, Department of Health Research
Report No. 11, 1965.
26. Hurwitz, E. and Williamson, F. M., "The Use of Cooperas-Sodium
Silicate as a Sewage Coagulant," Sew. Wks. J., 12, 562 (1940).
27. Rudolfs, W. et at., "Chemical Treatment of Sewage," Sew. Wks. J.,
12, 1051 (1940).
28. Gehm, W. H., "Chemical Coagulation of Sewage and Proteins as
Coagulant Aids," Sew. Wks. J., L3, 1110 (1941).
29. Black, A. P. and Eidsness, F. A., "Carbonation of Water Softening
Plant Sludge," J. AWWA, 49, 1343 (1957).
110
-------
30. Black, A. P., Shuey, B. B., and Fleming, P. J., "Recovery of
Calcium and Magnesium Values from Lime-Soda Softening Sludges,"
AWWA, jx3 [10], 616 (Oct., 1971).
31. Thompson, C. G., Singley, J. E., and Black, A. P., "Magnesium
Carbonate: A Recycled Coagulant," J. AWWA, J54 [1] , 11-19 (1972).
32. Thompson, C. G., Singley, J. E., and Black, A. P., "Magnesium
Carbonate: A Recycled Coagulant," J. AWWA, j>4 [2], 93-99
(1972).
33. Standard Methods for the Examination of Water and Wastewater,
American Public Health Assoc., N. Y., Thirteenth Edition, 1971.
34. Morgan, J. J. and Stumm, Werner, Aquatic Chemistry, Wiley
Interscience, 199-201 (1970).
35. Bridger, G. L., Salutsky, M. L., and Starostka, R. W., "Metal
Ammonium Phosphates as Fertilizers," J. Agr. Food Chem., 10,
181-188 (1962).
36. Bube, K., "Uber Magnesium Ammonium-Phosphat," Z. Anal. Chem.,
.49, 525-596 (1910).
37. Szekeres, L. and Rady, M., "The Solubility of Some Calcium
Phosphates," Agrartud. Egyet. Mezogazd. Kar. Kozl, 165-168
(1959).
38. Uncles, R. F. and Smith, G. B. H., "Solubility of Magnesium
Ammonium Phosphate Hexahydrate," Ind. Eng. Chem., Anal. Ed.,
1J3, 699-702 (1946).
39. Methods for Chemical Analysis of Water and Wastes, Environmental
Protection Agency, Water Quality Office, Cincinnati, Ohio (1971).
40. Fishman, M. G. and Douns, S. C., "Methods for Analyses of
Selected Metals in Water by Atomic Absorption," Geological
Survey Water Supply Paper 1540-C (1966).
41. Murphy, J. and Riley, J., "A Modified Single Solution Method for
the Determination of Phosphate in Natural Waters," Anal. Chim.
Acta., 2J_, 31 (1962).
42. Lange, N. A., Handbook of Chemistry, Handbook Publishers, Inc.,
Sandusky, Ohio, Ninth Edition, 1956.
43. Lehr, J. R. et al, , "Preparation and Characterization of Some
Calcium Pyrrophosphates," J. Agricultural and Food Chemistry,
11, 214-222 (1963).
Ill
-------
44. Lehr, J. R. et at. , "Calcium Ammonium Ortho-Phosphates," J. Agric.
Food Chem., 12, 198-201 (1964).
45. Tennessee Valley Authority, "Crystallographic Properties of
Fertilizer Compounds," National Fertilizer Development Center,
Muscle Shoals, Ala. (1967).
t
46. Villiers, R. V. et al. , "Municipal Wastewater Treatment by
Physical and Chemical Methods," Water and Sewage Works,
Reference Edition, R-62-R-81 (1971).
47. Weber, W. J., Jr. et at., "Granular Carbon Treatment of Raw
Sewage," Water Pollution Control Research Report, Washington,
D. C. (May, 1970).
48. Smith, C. V., Jr. et al. , "Physio-Chemical Treatment of Sewage,"
Presented 63rd Meeting American Institute of Chemical Engineers
(Dec., 1970).
49. Hannah, S. A., "Chemical Removal of Phosphorous," Advances in
Treatment of Domestic Wastes, Southeast Water Laboratory,
Athens, Ga. (Oct., 1971).
50. Bishop, D. F. et al., "Physical-Chemical Treatment of Municipal
Wastewater," J.W.P.C.F., _44, 361 (March, 1972).
51. Sullivan, J., "Feasibility of Treating Wastewater by Distillation,'
E.P.A. Project No. 17040 D.N.M. (1971).
52. Mercer, B. W. et al. , "Ammonia Removal from Secondary Effluents
by Selective Ion Exchange," J.W.P.C.F. (Feb. R95, 1970).
53. Menar, A. B. and Jenkins, D., "Calcium Phosphate Precipitation
in Wastewater Treatment," E.P.A. #EPA-R2-72-064, December, 1972.
112
-------
SECTION VIII. APPENDIX
The laboratory studies were conducted using one-liter samples.
The dosages of magnesium carbonate trihydrate (MgC03«3H20) and 98%
calcium hydroxide (Ca(OH2) were weighed out on an analytical balance
in 15 and 25 ml beakers, respectively. A portion of the one-liter
wastewater sample was retained in a 200 ml beaker in order to prepare
slurries of the chemicals and rinse the slurries from the beakers. A
glass stirring rod with a rubber policeman was utilized to facilitate
preparation of a uniform suspension. The magnesium slurry was
quantitatively transferred to the rapidly stirred (70 rpms) sewage
sample. The beaker is then rinsed with two successive portions of
wastewater and the washings added to the jar. After three minutes
the lime slurry was added and the rapid mix continued for an
additional two minutes. The speed was then reduced to 35 rpms.
After twenty minutes the stirrers were slowly removed prior to
cutting off the jar test machine. The suspension was then allowed
to settle for ten minutes prior to direct sampling or filtration.
Magnesium carbonate trihydrate, MgC03'3H20, is a white powder
of fine particle size and a bulk density of 40 Ibs/cu ft. The air-
dried product contains about 4% moisture and a dry basis is about
99.7% pure. This chemical should always be weighed out and added as
a slurry. Water solutions of the material slowly decompose to form
the relatively insoluble "basic" carbonate, 4 MgC03'Mg(OH)2*3H20.
"Magnesium carbonate" purchased from any source will be the
basic carbonate, 4 MgC03'Mg(OH)2'3H20. The low solubility, 90 mg/£,
slowness to dissolve and extremely low bulk density, 5-8 Ibs/cu ft,
make it unsatisfactory for practical use. It should not be used for
jar tests.
113
-------
-------
A i < rsvifjN Number
2
Subject Fn-ld & Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Or ft,miration
City of Gainesville
P.O. Box 490
Gainesville. Fla. 32601
Title
PHYSICAL-CHEMICAL TREATMENT OF MUNICIPAL WASTES
BY RECYCLED MAGNESIUM CARBONATE
1Q Authors)
ur. A. r . u lack
Dr. A. T. DuBose
R. P. Vnnh
16
21
Project Designation
EPA Grant Project
12130 HRA
Note
22
Citation
Environmental Protection Agency report number, EPA-660/2-7U-055, June
23
Descriptors (Starred First)
*Waste treatment, *Physical-Chemical treatment, *Chemical coagulation,
*Chemical recycle, *Sludge treatment, *Nutrient removal.
25
Identifiers (Starred First)
*Magnesium carbonate, *Magnesium hydroxide, *Calcium carbonate, *Carbon
dioxide, *Lime, *Clinoptilolite, *Carbon.
27 Abstract
'The applicability to municipal wastes of the recently discovered magnesium
carbonate-lime water treatment process has been investigated. A sixteen-month
laboratory study was conducted and was followed by an eight-month pilot plant
study. Four wastewaters with COD values varying from 200 to 1 ,500 mg/1 were
examined. Bench-scale coagulation studies designed to compare the effect of
added MgCCh with treatment by lime only showed a 0%-30% greater reduction in
effluent COD residuals. Color and turbidity reduction by the magnesium-pius-
lime process averaged 50%-85% greater when compared to treatment by lime only.
A series of 72-hour pilot plant runs was conducted with the magnesium precip-
itated increased after each three day period. Effluent characteristics improved
as the amount of magnesium precipitated was increased. Influent and filter
effluent samples were collected every four hours and analyzed for COD, TOC,
total phosphorus, alkalinity, hardness, calcium, and magnesium. Values for BOD
were determined from composited samples. The percentage reduction in chemical
(COD) and biological (BOD) oxygen-consuming substances ranged from a low of
70%for no magnesium ion precipitated to a high of 90% for 30 milligrams per
liter of magnesium ion precipitated. Higher dosages have not yet been
investigated.
A b.sfractor
A. T. DuBose
Institution
City of Gainesville, Fla.
WR 102 (REV JULY 19691
WRSl C
SEND TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U S DEPARTMENT OF THE INTERIOR
WASHINGTON, D C 20240
* GPO: 1968-359-339
-------
------- |